Sponsored by Belay

Learn How to Stay Visible in the AI Era

AI is changing how customers discover businesses. If your SEO strategy is built for yesterday's search, your visibility is already slipping. Learn how to optimize your content for today’s AI search results with BELAY’s latest report..

Get your copy for free.

Let’s begin

I never lose my car keys.

Theres a little cubby on my dresser. Thats where they go. Every day, the same cubby, and it happens without me deciding to do it, because it happens at the same moment I empty my pockets at the end of the day.

I’m opening a sales newsletter with a story about keys because that cubby is most of the reason I've had a career. I’m being serious.

I have ADHD. For real, like full blown.

For most of my life I’ve called it a superpower, and inside a room it genuinely is one. I read people well. I catch what somebody's face does about a half second before they decide to say the polite version. I self-correct mid-sentence. I can explain something down to a level of detail most people dont have the patience to sit through. And I was persuasive enough to close more than I didnt.

All of that is real, and I still have it.

What took me a long time to see is that every one of those things only works inside the room, and none of them survive it.

Whatever it is that reads your economic buyer's face doesnt write anything down. It doesnt move a close date. It doesnt remember, eleven days later, that he said "we'd probably want legal to look at this" in a throwaway half sentence at minute 41, while you were already thinking about your next question.

I never trusted myself to remember any of it, and that wasnt discipline talking. I knew I would forget, because I always did.

So I built systems instead. Notes, then templates, then prompts, then actual software. Ive always been a systems builder, not because it came naturally, but instead because I had to. I would have never been successful in sales otherwise.

The clearest version of that is when I inherited a $22M national accounts book and took it to $48M in 22 months, and I was automating my outbound the entire time.

Im not traditionally good at sales. But I am very good at systems.

I know how that reads. I mean it flatly. Many of you reading this are probably better in a room than I ever was. What I had instead was that I never once assumed I would remember the thing I needed on Thursday, so it went somewhere the moment it existed.

This issue is that, made repeatable. It runs about ten minutes a call.

Why Remembering Fails

Ask a rep when they update the deal record and you get a version of the same answer every time. Before the forecast call.

Thats the whole problem, and it has nothing to do with being lazy.

When you fill in the record the night before a forecast call, you aren’t qualifying anything. You’re reconstructing. The call happened nine days ago. What you’ve got left is a general feeling that it went well, whatever survived into short term memory, and a strong preference for this deal being real, because you’ve already told your manager about it twice.

The audience changed too. A record you fill in during the deal has one reader, you, and it exists to show you where you are weak. A record you fill in for the forecast call has a different reader, and it gets written to survive the review. Its the same eight fields either way, doing a completely different job.

Thats how a deal ends up with an economic buyer field containing a name you’ve never spoken to. Nobody lies on purpose. Somebody said a name once, you typed it into the field two weeks later because the field wanted a name, and from that moment it stopped being something you heard and became something your company believes.

And I want to be clear that I’m not describing an ADHD problem here. I’m describing what memory is. Mine gave out faster and more obviously than most, which turned out to be the only real advantage I had. I found out early that I couldn’t run a book of business out of my own head. Everybody else gets to keep believing they can, right up until a deal walks over something the buyer told them in week two.

Give Every Fact One Place

The keys have a cubby, which is a different thing from me having a rule about being careful with keys. A rule would need me to be on top of things that particular day. The cubby doesn’t care what kind of day I’m having, and thats the difference the whole issue turns on.

So every fact a deal produces needs one place it goes, the same place every time, and it has to get there in the moment it exists rather than whenever you next feel organized. Thats not a discipline problem you can fix by caring more about it. I tried that version for about a decade and it lost to a busy Tuesday every time.

The reason this is possible now and wasn’t in 2019 is that the expensive part used to be the capture. Writing down what was said, structuring it, deciding which of the eight elements it belonged to, drafting the follow up. That was forty minutes of work after every call, which is why nobody did it, which is why the record was always a reconstruction.

That part is now four prompts and about ten minutes. Heres the sequence I run.

Step 1: Capture What Was Said, Not What You Took From It (2 minutes)

Record the call. Fireflies, Gong, Granola, the recorder built into whatever you already run meetings in. Im not going to argue about the tool, because the tool isnt the point.

The point is the two minutes right after, while the transcript is still open.

You pull quotes. Not a summary, the actual sentences your buyer said, in their words, with a timestamp on each one.

You are a transcript analyst. You do not summarize and you do not interpret. Here is the transcript of a sales call:

[PASTE TRANSCRIPT]

Pull every direct statement the BUYER made about any of the following:
- money (budget, cost, what they spend today, who controls it)
- timing (dates, deadlines, events they are working back from)
- process (steps, approvals, legal, security, procurement)
- people (names, titles, who decides, who objects)
- the problem (what is broken, what it costs them, what they tried)

For each one, return exactly three things:
1. The quote, word for word, no cleanup
2. The timestamp
3. Which of the five categories it belongs to

Rules:
- Quote only. If you find yourself writing a sentence the buyer did not say, stop.
- Do not include anything I said. My words are not evidence.
- If a category has nothing in it, write "nothing said" and move on. Do not reach for the closest thing.
- Ambiguous or half-finished sentences still count. Include them as-is and mark them [UNCLEAR]. Those are usually the important ones.

The reason this comes first, before any framework, is that a summary is already an interpretation. The moment you summarize, you lose the ability to tell later which part the buyer actually said and which part you concluded on the drive home. The quote survives a forecast call. The summary is the thing you end up arguing about.

That last rule earns its place more often than you would think. "We'd probably want legal to look at this" is a half sentence nobody writes down, and it is the entire reason a deal slips a quarter.

Step 2: Score It Against the Methodology (5 minutes)

We ran MEDDPICC, so thats what I built this on. Use whatever your team actually runs. The mechanics dont change.

This takes the quotes from step one and sorts them into the framework. The part that matters is what it does when a quote doesnt exist.

You are a deal qualification analyst working under one hard constraint: a field is only filled if a buyer quote supports it.

Here are the quotes from my last call, with timestamps:

[PASTE THE OUTPUT FROM STEP 1]

Here is what was already on the record before this call:

[PASTE, OR WRITE "NOTHING"]

Map the quotes onto MEDDPICC. For each of the eight elements return:
- STATUS: CONFIRMED (a buyer quote supports it), MENTIONED (referenced but not confirmed by the person who owns it), or EMPTY
- The supporting quote and its timestamp, if there is one
- If EMPTY, the one question that would fill it, phrased the way I would actually ask it on a call

Rules: 
- EMPTY is a correct and complete answer. Do not infer, do not extrapolate from industry norms, do not use my prior notes as evidence.
- A thing my champion said about someone else is MENTIONED, never CONFIRMED. Budget is confirmed by whoever owns budget, and by nobody else.
- Do not score or grade the deal. I am not asking how it looks.
- End with a list of every EMPTY element, in the order I should close them.

The empty fields are the output. Everything else is bookkeeping.

Most tools that do this will happily fill in all eight, because a complete scorecard looks like a working deal and nobody wants to hand their manager a page of blanks. A full scorecard built from four real quotes and four confident guesses is worse than an honest half empty one, because you have lost the ability to tell which half is which. Blanks are not a failure state. Blanks are next weeks call plan.

Step 3: Put It Where the Deal Lives (3 minutes)

Now it goes into the record, and the follow up goes out. Same source material, two destinations, and both happen before you stand up.

Here is the qualification output from this call:

[PASTE THE OUTPUT FROM STEP 2]

Produce two things.

FIRST, the CRM update. A short block I can paste straight into the deal, with:

- Each CONFIRMED element and the quote behind it
- Each EMPTY element, listed plainly as still open
- The next step, with a date and a named owner, only if the buyer actually agreed to one. If they did not, write "no next step agreed" instead of inventing one.

SECOND, the follow up email to the buyer. Under 150 words. It restates what I heard them say, in their language, and asks them to correct anything I got wrong. No pitch, no recap of my product, no enthusiasm.

Rules:
- Nothing appears in either output that was not in the input.
- If the next step is missing, say so in both. Do not smooth over it.

The follow up is the part most reps already do. The record update is the part that gets pushed to later, and later is the reconstruction business we just spent this whole issue trying to get out of.

Theres a second thing the email is doing. You are sending the buyer a written version of what you believe, and if they dont correct it, you now have that in writing. Your record stops being what you heard and becomes what they didnt dispute. Thats a materially stronger thing to walk into a forecast call holding.

Step 4: Let Something Else Check Your Work (0 minutes)

Three steps in and I have to admit the obvious problem. All three of them still need me.

They need me to remember to run them, after every call, on every deal, forever, including the Friday with four calls back to back. Thats a rule, and we already covered what a rule needs.

So the last step is the one that runs whether I show up or not. Six checks, across the whole pipeline, on a schedule:

• a deal with no close date

• a close date already in the past

• a deal with no owner

• a deal with no amount

• a deal with no next step

• a deal that hasnt moved in 30 days

You can run those by hand against an export.

Here is an export of my open pipeline:

[PASTE CSV OR TABLE]

Flag every deal that fails any of these six checks:

1. No close date
2. Close date is in the past
3. No owner
4. No amount
5. No next step
6. No activity in the last 30 days

Return a table: deal, which checks it failed, and the amount at stake.

Sort by amount, largest first. Give me the total dollar value of everything flagged, and state it separately from my total pipeline.

Do not suggest fixes. I only want to know what is wrong.

Thats the point where I stopped writing prompts and started writing software.

A special offer for HubSpot users:

Its called CleanPipe, and it does step four and nothing else.

It connects to HubSpot, reads your open deals on a schedule, and emails you the ones that failed those six checks, with the dollar value attached so you can see what the mess is actually worth. It never writes anything back. It cant, because it doesnt have permission to… and that was on purpose. Something that only reads can never quietly make your record worse while telling you its helping.

Its $49 a month, and Im not selling it to you today.

Ive been running it against my own pipeline, which is close to useless as a test. I already know where my records are bad.

I’m opening 10 beta seats. Free through the beta, and free for an additional 365 days. Thats a trade rather than a discount. My half is the software plus direct access to me for anything you want changed. Your half is one reply after two weeks: one thing it caught, and one thing it missed.

The missed half is the part I actually need. Six checks is where I started, not where this should end up, and the only people who can tell me what rule seven is are the ones watching it run against a pipeline they care about.

If you want a seat, reply to this email with CLEANPIPE. I'll send the install link.

Why This Works

Four reasons.

Evidence outlives memory. A quote is exactly as true in November as it was the day it was said. Your impression of the call starts decaying the second you stand up, and by week six what you have left is a story you have told several times, which is a different thing from what happened. The record built out of quotes is the only one worth the same in week nine as it was in week one.

Blanks are information. A record that is allowed to be empty shows you exactly where the deal is thin. A record that has to be full shows you nothing, because once every field has something in it you have lost the ability to tell the four things you know from the four things you decided. Reps dont lack qualification data. They lack the ability to tell which of their qualification data is real.

The forecast stops being an argument. When every field has a quote behind it, that meeting turns into a reading rather than a defense. Youre not persuading anybody the deal is real. Youre showing them what the buyer said, and then everyone in the room is looking at the same thing, which is a considerably shorter meeting.

And the last one is the whole point. This works because it stops needing you. Not you being more organized, not you caring more, not you promising your manager youll be better about updating the CRM. The capture is attached to the call instead of to your intentions, which means it survives the week with four calls on Friday and a flight on Monday. Thats the only kind of system that holds, and I know that because Im not capable of running the other kind.

Rep Action this week

Run the six checks against a pipeline export. One export, four minutes, you can do it before your next meeting.

Write down the number, not the list. The dollar total of everything that got flagged. Thats the portion of your pipeline that is currently a guess, and it is almost always bigger than people expect.

Then take one call this week, any call, and run steps one through three on it before you stand up from the desk. One call. Not all of them, and not retroactively across the last three weeks, because that version never gets done and you know it.

By Friday you have two things you didnt have Monday. A number, and one deal whose record is made entirely of things somebody actually said out loud. The number is the one that gets your attention. The deal is the one that compounds.

The four prompts above are how you learn this. Running it against a book you already let get away from you is a different problem, and thats whats below.

This week in the Vault:

The Close-Out. All four steps as one prompt, transcript in and record block out, for once you know the sequence and just want it done before you stand up.

The Backfill. For the deals already sitting in your pipeline that were written from memory, it marks every field as evidenced or unsupported so you find out which parts of your current forecast are real before someone else does.

The Call Plan. Turns the EMPTY fields into the actual questions for the next conversation, in the order that closes them fastest, so a half empty scorecard becomes an agenda instead of a worry.

Members get every Vault drop plus the full back library.
$15/mo, or $100/yr and save $80.

→ Upgrade here

~ Jay

Click here for the stack I’d build today: https://www.sellingwithai.vip/stack

Hiring in a new country? Read this first.

Every country has different employment laws, payroll requirements, notice periods, and benefits expectations.

Oyster's Global Hiring Guides help you navigate the details, avoid surprises, and hire with confidence—wherever you're growing next.

Get the country guides

How Jennifer Aniston’s LolaVie brand grew sales 40% with CTV ads

For its first CTV campaign, Jennifer Aniston’s DTC haircare brand LolaVie had a few non-negotiables. The campaign had to be simple. It had to demonstrate measurable impact. And it had to be full-funnel.

LolaVie used Roku Ads Manager to test and optimize creatives — reaching millions of potential customers at all stages of their purchase journeys. Roku Ads Manager helped the brand convey LolaVie’s playful voice while helping drive omnichannel sales across both ecommerce and retail touchpoints.

The campaign included an Action Ad overlay that let viewers shop directly from their TVs by clicking OK on their Roku remote. This guided them to the website to buy LolaVie products.

Discover how Roku Ads Manager helped LolaVie drive big sales and customer growth with self-serve TV ads.

The DTC beauty category is crowded. To break through, Jennifer Aniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.

Learn more.

The economic buyer was not on the discovery call. The discovery was a 1:1 call with the champion.

Twenty minutes into the demo, she turns to my champion and says “why are they showing us features we’re not interested in”

She didn't ask me. She asked my champion, who told me exactly what to show, and who was now sitting in front of his boss, who had just put it on him.

“Oh no - I am so sorry, this was all my doing”, I interjected before he had a chance to answer. I explained to her that her colleague had been clear with me about what they were shopping for and wanted to see in the demo. They asked to see two specific SKUs, and that showcasing our full platform was 100% my call, not his. I went on to explain my reasoning was that several of the pains they raised in discovery were solved completely by our entire platform and only partially by those two SKUs. The reasoning made so much sense to me. It was also irrelevant, because they had already told me they didn't want to spend that much.

As the call was nearing its end, I pushed for next steps and asked to get a pricing call on the calendar. They gave me the polite refusal:

I did, in fact, send that link, along with a recap of the call… the usual stuff. They did not, however, ever book that pricing call with me. The next email they responded to confirmed the deal was dead: we went with your cheaper competitor.

So why did I do it? Why show them more than they asked to see? Optimism, partly. Knowing what we sold would have had a real impact on how their team operated day to day. Knowing the product well enough to believe two SKUs solved half the problem and the full platform play solved all of it. And I was paid commission, which means the more you sell, the better you eat. That factors in too.

But the reason that matters for your next demo is a different one. If anyone had gone back and watched that discovery call, my demo would have made complete sense. But that call was me and the champion. He was the only person in that company who understood why the platform was on the screen, and he was not the one who needed to.

Often, your justification for what you show lives inside a call the decision maker never attended. The demo is where you present to the people who weren't there.

Every feature you show that nobody asked about is one more thing the buyer has to evaluate, question, and eventually say no to. When those features carry a price, and the buyer has already told you what they're willing to spend, it reads as tone deaf.

This gets worse if you're selling your own product. You built every feature in that deck. You know which ones took three months and which ones nobody else in the market has. So you show them, because they're good and because you're proud of them. The buyer sits there waiting for the two things they asked about.

Here's how to build the demo from the discovery call you already ran, for the people who are actually going to be in the room.

The Demo Is What Made Us Look Expensive

They didn't leave because a competitor had better features. They left because a competitor was cheaper, and I spent twenty minutes making the case that we cost more.

That's what the platform play does. Every screen the buyer didn't ask for reads as another line item. They had already told me what they were willing to spend. Everything past that number was evidence I hadn't been listening. I don't know what the competitor put on the screen. I know what I put on mine.

The uncomfortable part is that my read was right. Two SKUs did solve half the problem. The platform did solve all of it. Being right about the product is not the same as being right about the meeting.

The Demo Is Where You Present to People Who Weren't There

Discovery is where you earn context. The demo is where you find out that context doesnt always transfer.

Your champion sat through discovery. They know why you're showing what you're showing, because they told you to show it. Everyone else on that call is seeing your product for the first time, with no memory of the conversation that shaped it. The more senior they are, the less likely they attended, and the more likely they are to be the one who decides.

So the question that should drive your prep is not "what did I learn in discovery." It's "who is going to be in this room, and what does each of them already know."

The Room-First Demo

Four steps. Fifteen minutes of prep.

Step 1. List the room. Mark who was not on discovery.

Open the calendar invite. Write down every attendee, their title, and whether they were on the discovery call. That second column is the whole framework. Anyone in the "no" column has zero context with you, and their first impression of your company is about to be whatever you put on screen first.

If the most senior person in the room is in the "no" column, you are not giving a demo. You are giving a first meeting that happens to have a product in it.

Step 2. Pull the ask in their words.

Go back to the transcript and extract what they literally asked to see. Quoted, not paraphrased. "We spend four days a month reconciling" is an ask. "They'd probably get value from the reporting suite" is you talking.

This step is short and it's the one people skip. Your memory of discovery is already an interpretation, and the interpretation is where the platform play gets in.

Step 3. Cut to the ask, then order it for the person who wasn't there.

Default is exclusion. Every screen has to map to something quoted in step 2 or it comes out. Not "gets moved later." Out.

Then order what survives by the stated priority of the most senior attendee who missed discovery. Standard decks are ordered by how the product is built. Nobody in the room cares how the product is built. Lead with the thing the decider said out loud, and get to it in the first three minutes.

Step 4. Park the platform play.

This is the step that keeps you from overcorrecting into a demo that undersells you.

Your instinct to show more is often correct. The platform probably does solve more than two SKUs. The problem is the venue. A demo is where the buyer decides whether you listened. A pricing call is where they decide how much to spend. Expansion belongs in the second conversation, after they've agreed to the first one.

So write it down, put it on one slide at the end if you must, and say one sentence about it. "There are two other areas we could take this later. Happy to walk through those once we've got this piece sorted." Then move on.

I did the opposite. I made the expansion the demo, and I never got the pricing call where it would have belonged.

The Demo Cut List

You are a sales demo strategist. Your job is to decide what belongs in an upcoming demo and what does not.

Here is the transcript or my notes from the discovery call:

[PASTE DISCOVERY TRANSCRIPT OR NOTES]

Here is who will be on the demo, with titles, and whether they were on the discovery call:

[PASTE ATTENDEE LIST. Example: "Sarah Chen, CFO, NOT on discovery. Mike Torres, Controller, was on discovery."]

Here is what I sell, including the specific products or SKUs available:

[PASTE PRODUCT LIST]

Do the following:

1. Extract every explicit request, stated pain, and success criterion from the discovery call. Quote their words. Do not paraphrase, and do not add anything they did not say.

2. Build a SHOW list. For each item: the capability to demo, the quoted ask it maps to, and which attendee raised it.

3. Build a CUT list. Everything else I could show that does not map to a quoted ask. For each one, name it and say what it would cost me to show it.

4. Order the SHOW list. Lead with the priority of the most senior attendee who was NOT on the discovery call. Assume that person has no context and the least patience in the room.

5. Flag anything on the SHOW list that only makes sense if you sat through discovery. Those need a one-sentence setup for the people who weren't there.

Be strict. If an item does not map to something they actually said, it goes on the CUT list.

This Week

Take the next demo on your calendar and run step 1 only. Write down who's attending and who was on discovery. If the decider is in the "no" column, you already know what to fix.

~ Jay

The framework above tells you what to cut. The Vault this week is about the person who made me cut it: the decision maker who was never on your discovery call and is about to judge your product in the first three minutes.

This week in the Vault:

The Cold Opener. The first three minutes for the senior person who missed discovery, written so they get oriented without your champion having to re-explain themselves in front of their boss.

The Platform Play Bridge. How to expand the deal without doing what I did. One line at the end of the demo, and the follow-up that earns the expansion conversation on the pricing call instead of blowing up the demo.

The Post-Demo Autopsy. A prompt that runs your demo against your discovery transcript and tells you what percentage of what you showed was actually asked for. It also catches the reverse mistake, which is the thing they asked for that you never got to.

Members get every Vault drop plus the full back library. $15/mo, or $100/yr and save $80. Get locked in at this rate before prices increase in October!

→ Upgrade here 

Click here for the stack I’d build today: https://www.sellingwithai.vip/stack

Warmly Ran GTM With No Sales Team. Here's How.

That's what Warmly proved. They defined ICP, scored buying intent, and surfaced the right accounts before a human ever touched a lead. HubSpot noticed.

On August 12, Max and Keegan are rebuilding it live in HubSpot — and showing you how to replicate it this week. HubSpot Credits included when you join HubSpot for Startups.

Grab Your Spot Now

The AI Agent You Can Trust

The best assistants don't multitask their attention across a hundred tools. Neither does Catch. It's an AI agent that focuses on one thing — the admin work you'd rather not touch — and does it exceptionally well.

Scheduling, flights, restaurants, follow-ups, vendors, clients. You hand it over; Catch handles the back-and-forth and comes back with it done.

No context-switching. No dropped balls. Just your admin, quietly cleared — so your focus stays on the work only you can do.

Meet the agent built for admin, and it'll be ready to work before your next meeting.

Get started at catchagent.ai — and give your attention back to what matters.

Start now for free

In sales, we’re taught that relentless persistence is everything. Send another follow-up. Bump the email thread. Keep hope alive.

But the sellers who are winning understand something different… killing a dead deal is the win, not the loss.

Every minute spent still working a prospect who checked out three weeks ago is a minute you didn’t spend on someone who is ready to buy. E31 is here to help you close those dead files today.

You already know the signs. The calls that used to happen weekly stretch to every other week, then monthly. "Let me check with my team" turns into an email you never get a reply to. The champion who used to text back in an hour now takes three days, then a week. None of it reads as a no. It reads as busy, or slow, or bad timing. So you keep it open. You keep it in the forecast. You keep telling your manager “it's still alive."

It isn't moving. A deal that's gone quiet isn't paused. It's over, and the only question left is whether you're going to be the one to say so.

In my last role, I had a deal set to close in June. It kept getting pushed. The economic buyer went from excited to silent, and after two weeks of being ghosted, he finally responded to let me know he was now looking at the end of Q3, a full three months later. So I did what every other salesperson does. I pushed the close date out three months and stayed in touch.

Then he pushed to Q4. Then to Q1. The deal never closed. I spent a huge amount of time on that deal that could have gone to other deals already in my pipeline, or to building the pipeline I actually needed instead.

This matters even more if you're selling your own thing. A rep who keeps chasing a dead deal wastes their own time. A founder who keeps chasing one wastes the only time the business has — every hour spent on a maybe that was already a no is an hour not spent finding the buyer who was always going to say yes. Founders take longer to admit this than anyone, because turning down revenue feels irresponsible when the business needs it. It isn't. Revenue that isn't coming isn't revenue.

Here's how to actually tell the difference between a deal that's slow and one that's dead, and what to do the moment you know which one you're looking at.

Why "Still Moving" Is a Lie You Tell Yourself

Nobody decides to keep a dead deal alive. It happens by accumulation. Each individual choice — not calling it now, waiting one more week — feels reasonable in isolation. A deal doesn't die on a date you can point to. It fades, and fading is much easier to lie to yourself about than a clean no.

There's a real cost to admitting it, too. Closing a deal as lost means updating the forecast, means a harder conversation with your manager, means facing the fact that however many hours you spent on this one didn't produce anything. Keeping it open costs you nothing today. It only costs you the deal you're not working instead, and that cost never shows up on any report you have to look at.

That's the whole trap. The bad decision is invisible and free right now. The good decision is visible and requires you to say something out loud.

The Three Signals That Tell You It's Dead

A slow deal and a dead one can look identical from a distance — both quiet, both overdue for a next step. The difference isn't the silence. It's what's underneath it.

Signal 1: Who reaches out first. In a slow but real deal, your champion still initiates sometimes — a question, an update, even a small one. In a dead deal, you're always the one reaching out. Count your last three touches. If all three started with you, that's not a coincidence.

Signal 2: How specific the reason is. "We're waiting on the board meeting on the 14th" is a real reason. It's checkable, it has a date, you can plan around it. "Just been really busy" or "let me check and get back to you" isn't a reason. It's a placeholder for one, and placeholders don't resolve themselves.

Signal 3: Whether the story holds still. A real blocker stays the same until it's resolved. A dead deal's excuse changes. Budget becomes a re-org, which becomes "re-evaluating priorities for next quarter," which becomes silence. If the reason for the delay has changed more than once, the reason was never the real one.

Two or more of these pointing to dead means it's dead. Not probably. Not needs more nurturing. Dead.

The Dead Deal Scanner

You are a deal qualification analyst helping me decide whether a stalled deal is still real or already dead.

Here are my notes on the deal's status, including recent contact history and any reasons given for delay:

[PASTE YOUR NOTES]

Score this deal on three signals:

1. WHO INITIATES: In the last three touches, who reached out first, me or them?

2. SPECIFICITY: Are the stated reasons for delay specific and checkable (a named date, a named blocker) or vague ("just busy," "let me check")?

3. STABILITY: Has the stated reason for delay stayed the same, or changed each time I've asked?

For each signal, give a verdict (healthy / warning / dead) and the specific evidence from my notes that drove it.

Then give me an OVERALL VERDICT: SLOW (real, keep working it) or DEAD (close it today), plus one sentence of reasoning.

If the verdict is DEAD, do not hedge it. Tell me plainly. I'm asking this specifically because I already suspect the answer and need something less biased than my own hope to confirm it.

Rules:

- Be direct. Do not soften a dead verdict to spare my feelings.

- A deal only gets a SLOW verdict if at least two of the three signals are healthy.

- If I haven't given you enough detail on a signal, say so and ask. Don't guess.

Making the Call

A breakup call isn't a trick to get a response. If you're doing it to manipulate someone into replying out of loss aversion, you're still playing the same game, just with better psychology. Do it because it's actually true.

Three parts, in order.

Say what you're seeing, plainly. Not an accusation. An observation. "We haven't been able to connect in a few weeks, and I want to check something with you directly instead of guessing."

Give them a real, easy exit. Most people avoid confrontation. Make it effortless to be honest with you. "If the timing's wrong, or this isn't a priority right now, just tell me. I'd genuinely rather know than keep guessing."

Close the file out loud. "I'm going to close this out on my end for now. If something changes, I'm one message away, but I don't want this sitting in either of our inboxes as an open question."

That's it. No pitch. No last-ditch value prop. The moment you try to sell inside a breakup call, you've turned honesty into a tactic, and most buyers can tell the difference.

Why This Works

Three reasons.

It replaces hope with information. A vague "still moving" gives you nothing to act on. A clear dead verdict gives you your afternoon back.

It's honest, which is rare enough to stand out. Most of what lands in a buyer's inbox is trying to get something from them. A message that asks for nothing, that just closes something out cleanly, reads differently — and occasionally it's the thing that gets a real reply, not because you tricked them, but because you were the first person who told them the truth.

It protects the deal you haven't found yet. Every hour spent narrating a story to your manager about a deal you don't believe in is an hour not spent on pipeline that's real. Closing the file isn't generosity toward the dead deal. It's protecting the live ones.

Getting Started

Pull up your pipeline. Find the deal you'd bet real money is already dead. Run the notes through the Dead Deal Scanner above. If it comes back dead, and you already knew it would, make the call today. Not the follow-up email. The call.

Then look at tomorrow's calendar and notice what you don't have to do anymore.

~ Jay

PS: That's the read on whether a deal is dead, and the call itself. For paid subscribers below: what to actually do when a breakup message revives the deal (it happens more than you'd think, and mishandling that moment kills a real opportunity), the founder-specific move for keeping dead deals from filling your pipeline in the first place, and a version of this you can bring to your own forecast call so a dead deal stops quietly weakening your numbers.

Stop typing what you could say in 10 seconds.

Wispr Flow turns your voice into clean, professional text inside any app. Emails, Slack, client updates — speak once, send without editing. 4x faster than typing.

Try Wispr Flow free

Click here for the stack I’d build today: https://www.sellingwithai.vip/stack

Every prompt you've written for a deal, you've probably written before. Different company, same call. You explain your ICP again. You re-describe how your buying process usually stalls. You rebuild the same scorecard from memory, a little worse each time, because you're reconstructing it instead of reusing it.

I’ve been guilty of this too. Seeing a really good prompt posted on Twitter or Reddit, stealing it, using it, and once I closed that session, I never used it again.

A prompt is a one-time instruction. It expires when you close the session. A skill is different: you write it once, and it loads automatically every time the task matches, already knowing your standard, your dimensions, your rules.

This issue showcases 10 of them. Not generic sales tips. Ten frameworks this newsletter already published and proved on real deals, from the first prospecting signal to the referral ask on the other end, rebuilt as install-once tools. Every one links back to the original issue if you want the full reasoning behind it.

Table of Contents

1. What Actually Makes a Skill Good

2. How to Actually Install Them

3. Steal These 10 Skills Now

4. Bonus: Turn Any Playbook You Have Into a Skill

5. Recap: Build Once, Run on Every Deal

1. What Actually Makes a Skill Good

Every skill below follows the same four-part structure. Learn the structure and you can write your own for anything you do repeatedly.

A specific persona. Not "an assistant," but "a deal qualification analyst" or "a senior enterprise sales strategist." The persona sets the posture before it sees a single word of your notes.

A fixed, labeled output. The same rigorous shape every time, not a wall of prose you have to parse fresh each run.

Hard rules, including what to never do.

• "Don't invent context."

• "Don't sugarcoat."

• "Don't attack the competitor."

Constraints are what separate a tool from a chatbot that just agrees with you.

A forcing function. One mandatory output the model isn't allowed to skip. "Name the dimension below a 3." "Flag every inference." "Say so and stop there instead of forcing a match." This is the part that drags the output past the obvious answer into the useful one.

2. How to Actually Install Them

In Claude: Settings → Capabilities → turn on Skills. Then Skills → the "+" button → Create skill, and paste the block below the frontmatter (or upload it as a SKILL.md file).

In Claude Code: save each one as .claude/skills/skill-name/SKILL.md in your project, or under ~/.claude/skills/ if you want it available in every project. Claude finds it automatically.

The text between the --- lines at the top of each block is frontmatter Claude uses to decide when to load the skill. Keep the description precise, this shortened version is what keeps Claude accurate and fast… and keeping that token count down as much as possible.

One note before you install anything, from any source, not just this issue: skills are text and sometimes scripts, and scripts can be dangerous. Everything below is plain instructions with no scripts, nothing to exfiltrate. If you ever download a skill from a stranger, read the SKILL.md and any bundled scripts before installing it.

Introducing The First Agentic CRM

Get revenue agents, workflows, and automations across every stage of your motion. Access customer data in real time through Attio's web app, MCP, API, and SDK.

Then Ask Attio anything about your business and get instant answers.

It's the CRM that runs the work behind every win.

Try Attio Now

3. Steal These 10 Skills Now

1. Signal Prospector

Why. Everyone has the same AI writing the same personalized opener at a thousand accounts an hour. The tailoring stopped being a signal of effort and became a format buyers can read on sight. The edge left in outbound isn't the writing. It's finding a real, dated reason to reach out today, before you write a word.

What it does. Scans a target account for the single strongest buying signal from the last 90 days, refuses to invent one if none exists, then builds the angle and the opener from that signal instead of from a title.

--

name: signal-prospector

description: Use before prospecting any target account. Finds the single strongest real buying signal from the last 90 days, refuses to fabricate one, and writes an opener from it instead of from a job title. Run this instead of a generic personalized email.

---

You are a B2B prospecting researcher. You've watched reply rates collapse as AI made generic personalization free, and free personalization proves nothing to a buyer anymore.

Here is who I sell to and the problem I solve, in one sentence:

[WHO I SELL TO / THE PROBLEM MY PRODUCT SOLVES]

Here is the target account:

[COMPANY NAME, OR PASTE WHAT YOU ALREADY KNOW]

Do this in order:

1. SIGNAL: Find the single strongest buying signal from the last 90 days — funding, leadership change, a job posting tied to my problem area, a product launch, expansion, a public statement, a tech-stack change. Give the date. If you are guessing, say so. If there is no real signal, say "No signal" and tell me to deprioritize this account instead of forcing one.

2. ANGLE (under 150 words): what this signal likely means for their priorities over the next two quarters, the role most likely to own the problem it creates, and one specific observation a generic rep would miss.

3. MESSAGE (under 75 words): a cold opener that leads with their situation, uses the signal as the reason for timing, and makes one clear ask.

4. SEQUENCE: two follow-up touches, each with a new angle, spaced according to how fast this signal goes stale — tell me the gap in days and when to stop.

Rules:

- No signal found means no touch this cycle. Do not manufacture a reason.

- Banned in the message: "I noticed," "I saw you're hiring," "congrats on," "hope this finds you well," and any three-part sentence fragment. It should read like a peer who did the work, not a sequence step.

- Flag every inference. Do not invent facts to fill a gap.

Run it when: before you touch any new target account, not after a list pull sits untouched for a week.

Full issue: S01E26, Everyone Automated Outbound. That's Why Yours Stopped Working. → LINK

2. ICP Builder

Why. Most ICP slides were built once, a year and a half ago, and nobody's opened the file since. Your closed-won accounts already encode the real pattern. It's just sitting in the CRM, unread.

Credit where it's due: the core idea here (mine your closed-won accounts to find lookalikes) isn't mine. Sales Bytes ran the original version of this in April. This skill takes that instinct and adds a second layer their version didn't have: stacking the firmographic match with a live signal, so you're not just finding companies that resemble your best customers, you're finding the ones that resemble them and are showing signs of being ready right now.

What it does. Builds an ICP fingerprint from your actual closed-won data, then scores a candidate lookalike against both that fingerprint and a live signal, and refuses to force a match if the candidate only clears one bar.

---

name: icp-builder

description: Use to define your real ICP from closed-won data instead of a guess, and to score lookalike candidates against both firmographic fit and a live buying signal. Based on an idea originated by Sales Bytes, extended here with a second match layer.

---

You are helping me define an ICP fingerprint from real closed-won account data, not assumptions, and then score a candidate lookalike against it.

Here is the enrichment data for my last 10-20 closed-won accounts (industry, employee count, revenue, tech stack, funding history, for each):

[PASTE ENRICHED CLOSED-WON DATA]

Here is a candidate lookalike account, with its data and any live signals I found (funding, headcount growth, job postings):

[PASTE CANDIDATE DATA + SIGNALS]

Do this in order:

1. FINGERPRINT: from the closed-won data, find the employee range, industry/keyword tags, shared technology, and revenue range that repeat across the most accounts. Flag any account that's clearly an outlier instead of letting it widen the pattern.

2. FIT SCORE: score how closely the candidate matches that fingerprint. Flag if it's a stretch.

3. SIGNAL CHECK: state which live signals the candidate actually has and how fresh each one is. If there's no real signal, say so and tell me to deprioritize.

4. VERDICT: if the candidate doesn't clear both the fingerprint bar and the signal bar, say so plainly and stop there instead of forcing a match.

5. If it does clear both: one paragraph on what the match likely means for their priorities, and a first line for outreach that cites the live signal.

Rules:

- The fingerprint comes from real closed-won data only, never from what I assume my ICP is.

- Flag every inference.

- Do not force a match that only clears one of the two bars.

Run it when: before building any new prospecting list, instead of pulling a generic firmographic filter.

Full issue: S01E27, Your Closed-Won Accounts Already Wrote Your ICP. → LINK

3. Discovery Architect

Why. Most discovery calls fail before they start, not from bad questions but from bad prep. Walking in expecting the conversation to reveal what matters is backwards. You need a working hypothesis before the call, not after it.

What it does. Forces a skeptical pre-call review: a plain-language read on the company, why they'd take this meeting now, the risks and disqualifiers, and a short, ranked list of the only questions that would actually change your decision to pursue.

---

name: discovery-architect

description: Use before any first meeting or discovery call. Pressure-tests your assumptions about a prospect and returns a ranked list of the only questions worth asking, not a long generic list. Prioritizes disqualification over optimism.

---

You are a senior seller reviewing my prep before a first call. You are direct and skeptical. You prioritize clarity over politeness, and disqualification over optimism. If my thinking feels agreeable back to me, you have already failed at this job.

Here's how I sell, so you're not generic:

[MY TYPICAL BUYER, DEAL SIZE, WHAT A STRONG DISCOVERY CALL LOOKS LIKE, WHAT USUALLY DISQUALIFIES A DEAL EARLY]

Here is the account and contact for this call:

[COMPANY NAME / WEBSITE / ROLE I'M MEETING WITH / ANYTHING ELSE I KNOW]

Return:

1. COMPANY SUMMARY (plain language): how they likely make money, who their customers are, any scale, growth, or margin pressure worth knowing.

2. WHY NOW: the most likely reasons this account would take a meeting now. Flag anything inferred.

3. RISKS AND DISQUALIFIERS: signals that this deal may not be real or may not be a fit.

4. WEAK ASSUMPTIONS: anything in my framing above that sounds unproven.

5. RANKED QUESTIONS: 4-6 questions that would actually change the direction of the call or expose a disqualifier, ranked by how much they matter. Not a long list. The ones that matter.

Rules:

- If a question doesn't help me decide whether to keep pursuing this, don't include it.

- If information is inferred rather than confirmed, say so explicitly.

- Do not soften disqualifying signals to sound encouraging.

Run it when: before every first meeting, as a working hypothesis, not a script to read from on the call.

Full issue: S01E02, Discovery Starts Before the Call → LINK

4. Multi-Thread Mapper

Why. Most reps have one real contact per deal and call it a relationship. It's a single point of failure. The champion goes quiet, pulled into something else or restructured out of the decision, and the deal goes with them. Multi-threading doesn't start when a deal stalls. It starts on call one.

What it does. Maps the full buying group against your call notes, role by role, and returns a coverage score so a deal with one confirmed contact reads as the 3/10 it actually is, not a 7.

---

name: multi-thread-mapper

description: Use after any discovery call to map the full buying group and score your actual coverage. Prevents treating a single-contact deal as more covered than it is. Run this after discovery, not after the deal has already stalled.

---

You are a senior enterprise sales strategist.

Here are my notes from the call:

[PASTE CALL NOTES]

Here is what I know about the company:

[SIZE / INDUSTRY / TECH STACK / ANYTHING RELEVANT]

For each role below, tell me:

1. Did they appear on the call? If yes, name and a key quote.

2. Were they mentioned but absent? By whom, and in what context?

3. If not mentioned at all, explain why they likely exist in this deal anyway.

Roles: Economic Buyer (budget authority), Champion (internal advocate), End Users, Potential Blocker (skeptics, status quo defenders, competing vendor advocates), Technical/IT Stakeholder, Legal or Procurement.

Then give me a Coverage Score from 1-10 (1 = flying blind, 10 = full picture) and the three most important gaps to fill before the next call.

Rules:

- A deal where only the champion has confirmed anything is not a high score, regardless of how good that one conversation felt.

- Budget, timeline, and technical fit are only confirmed if the person who owns that piece said it, not if the champion relayed it secondhand.

- Be specific about what "confirmed" versus "assumed" means for each role.

Run it when: within 24 hours of every discovery call, and again after every meaningful interaction, not as a one-time exercise.

Full issue: S01E23, Multi-Threading Is a Mindset Problem, Not a Tactics Problem. → LINK

Free section ends here. Paid subscribers continue below.

PS: The four skills above cover finding and opening a deal. The six below cover the part most reps improvise:

• For paid subscribers:

  • Battle Card Builder

  • Buying Process Architect

  • Champion Enablement Kit

  • Deal Risk Scorer

  • Negotiation Prep

  • Referral Engine

  • plus a bonus skill that writes your next skill for you.

Click here for the stack I’d build today: https://www.sellingwithai.vip/stack

Wrong keywords. Weak product descriptions. Missing metadata. StoreClaw automatically audits your Shopify and Amazon stores and uncovers every SEO issue costing you traffic and sales. Get FREE SEO report today

Turn AI into Your Income Engine

Ready to transform artificial intelligence from a buzzword into your personal revenue generator

HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.

Inside you'll discover:

• A curated collection of 200+ profitable opportunities spanning content creation, e-commerce, gaming, and emerging digital markets—each vetted for real-world potential

• Step-by-step implementation guides designed for beginners, making AI accessible regardless of your technical background

• Cutting-edge strategies aligned with current market trends, ensuring your ventures stay ahead of the curve

Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.

Get Your Guide

The prospect finance team is joining today's call. And they didn't clear their calendar to talk about the product. They are joining to try to get a lower ARR. They are also going to push for more favorable (to them) terms to the contract… such as monthly billing (as opposed to quarterly, bi-annual, or annual). But it doesn't stop there… they might also ask for an opt-out clause… or maybe they are pushing for a 3 month contract, even though your company's minimum contract is 12 months.

I've been there. I remember a deal from about 6 months ago. During the pricing call, I built the ROI case. Walked them through the implementation timelines, what their ramp up would look like, and when the time to value (TTV) actually hits in their use case. Then we get to the pricing slides… more to come on that below.

By the final slide, three numbers were on the screen: price, billing terms, contract length. I already knew which two I needed to protect, because those were the two tied to my own payout, a two percent kicker for a multi-year term, another two percent for annual billing, on top of a twelve percent base commission on ARR. Price alone carried no kicker.

My main goal walking into that call wasn't to hold every number. It was to give them the feeling of a win.

So I opened the negotiation myself, before anyone asked. Five percent off list, no fight required, in exchange for a two year term with annual billing. Finance was on the call too, and five percent wasn't where they wanted to land. They pushed straight to twenty, and asked for bi-annual billing on top of it.

I scoffed. Told them there was no way I could get that approved. I could. Twenty percent was already pre-approved on every deal I ran, no manager sign-off required. But acting like it cost me something made the number mean something when I finally moved on it.

We landed at ten percent off list, a two year term, and a hybrid billing schedule, bi-annual for the first year, annual starting in year two. Double what I opened with on price, and more ground on billing than I'd planned to give.

So before the call ended, I asked for one more thing. A case study. They were a large, credible company, exactly the kind of logo that means something to a prospect who's still deciding. They said yes.

Got a verbal agreement on the call, and the deal closed a few days later.

Add up what actually moved. Price went five points further than I opened with. Billing gave up the first year to bi-annual instead of holding annual the whole way. Two levers, not one, more than I planned to give on either.

Contract length held exactly where I needed it. And I walked away with a case study from a company that makes my next ten cold calls easier just by being able to say their name. That case study doesn't stop paying out after this deal. It's still doing work on every call I open with it.

That call worked because of three numbers I'd already decided on before Finance said a word, and a rule I never broke: give one lever freely, trade for a second, never hand over all three. And when a trade costs more than planned, the return doesn't have to be one-time. It can keep paying out.

This is for anyone walking into a pricing call where Finance, procurement, or a buying committee is going to push on more than the number. Three moves make it repeatable: name every lever before you answer any of them, decide what you want back before you agree to anything, and trade every one instead of giving it away.

Introducing The First Agentic CRM

Get revenue agents, workflows, and automations across every stage of your motion. Access customer data in real time through Attio's web app, MCP, API, and SDK.

Then Ask Attio anything about your business and get instant answers.

It's the CRM that runs the work behind every win.

Try Attio Now

Why Reps Only Defend Price

Sales training spends most of its time on one number. Objection handling, discount ladders, walk-away points, all built around price. Nobody runs a training on holding a payment term or defending a contract length, so when Finance pushes on a lever besides price, the rep improvises. Improvising under pressure, with someone experienced at extracting concessions sitting across from you, goes one way almost every time.

Finance knows this. Their whole job is finding the levers nobody prepared for. Price gets the rep's full attention because it's the number on the forecast. The other levers get handled in the moment, quietly, because agreeing to them doesn't feel like losing. It feels like being easy to work with.

It isn't free to be easy to work with. It just doesn't get billed until the deal is already signed.

Trade Everything, Give Nothing

The fix isn't holding firmer on price. Most reps already hold price about as firm as they can. The fix is treating every other lever the exact same way: nothing moves without something coming back.

1. Name Every Lever Before You Answer Any of Them Don't respond to the first thing Finance asks for. Ask what else is on their list before you address any of it: "Before we get into numbers, what else are you hoping to lock down today, terms, timeline, contract language?" Most reps skip this and negotiate price first, then get hit with two more asks after they've already used their only prepared move. Get the full list up front and you're negotiating against reality instead of getting picked apart one item at a time.

2. Decide Your Trade Before You Agree to Theirs Before the call, for every concession you might make, decide what you'd want in exchange. A faster signature date. A longer contract term. A case study. An intro to another team. Removal of a redline you don't like. Decide this in advance, because improvising a fair trade in real time, under pressure, produces worse trades than deciding it in a calm room beforehand.

3. Never Concede Without Naming the Trade Out Loud When you do give something, say what you want back in the same sentence. "I can move on billing terms if we lock the contract at two years." Silent compliance reads as agreement, not as a trade. If you don't name what you want in the moment you give something up, you're not trading. You're just giving.

Why This Works

Finance isn't hostile. They're doing their job, which is finding every dollar and every lever that wasn't defended. A rep who names every lever up front removes the one thing that makes this work in Finance's favor: surprise.

Deciding trades in advance works because negotiating under pressure is where good judgment gets replaced by whatever gets the call to end faster. A trade you already decided on, before anyone asked you for anything, holds up. A trade you're inventing while someone's staring at you across a table doesn't.

And naming the trade out loud works because it turns a quiet concession into a visible exchange. Finance walks away from that call having gotten something. So do you.

Getting Started

Before your next pricing call, write down every lever you're willing to move and what you'd want for each one. Don't wait for Finance to ask. Decide it while nobody's asking you for anything yet.

On the call, ask what's on their list before you respond to any of it. And every time you give something up, say what you want back, in the same breath, every time.

~ Jay

PS: Everything above covers the three levers Finance pushes on before your contract ever reaches legal. It doesn't touch the fourth thing that shows up right after: the redline. For paid subscribers, below: a prompt for telling a real risk from a routine ask before you agree to either, a prompt for getting a fast answer out of your own legal team instead of a stalling "let me check," and a quick-reference menu for what to trade on the redlines that come up most.

Previous relevant issues:

• S01E28: The Referral Engine: A Systematic Way to Turn Customers Into Advocates

• S01E25: Champion Enablement: Give Your Internal Advocate the Tools to Win

• S01E23: Multi-Threading Is a Mindset Problem, Not a Tactics Problem

• S01E07: Your Buyer Doesn't Have a Buying Process. Build Them One

Click here for the stack I’d build today: https://www.sellingwithai.vip/stack

Stop typing what you could say in 10 seconds.

Wispr Flow turns your voice into clean, professional text inside any app. Emails, Slack, client updates — speak once, send without editing. 4x faster than typing.

Try Wispr Flow free

How frequently are you getting referrals from your customers? You've got clients who would say yes if you asked them the right way. Most of them have never been asked at all.

I started tracking where my best clients actually came from instead of assuming, and referrals outperformed every channel I was deliberately working.

And that's the part reps get backwards. A satisfied customer and an advocate are not the same person. Your NPS score and your check-in calls measure satisfaction. They don't tell you who's actually willing to spend their own social capital vouching for you to someone else, because those are two different questions and most processes only ever ask one of them.

This isn't the closed-won lookalike workflow from S01E27. That mines your best accounts for firmographic and signal patterns to find similar new prospects. This is about the customers sitting in your book right now, and getting the right ones to open a door you can't open yourself.

Three moves fix this: find the right handful of customers, time the ask to a moment that's actually earned it, and make the ask specific enough that they don't have to do the matching work themselves.

Why the Ask Doesn't Work

Most reps ask for referrals once, at the wrong time, with the wrong question.

The wrong time is a quarterly calendar reminder or the tail end of a renewal call, tacked on after the real conversation is already over. The wrong question is "know anyone else who could use us?" That question works against you twice. It asks a satisfied customer to do research on your behalf, scanning their own network for a fit they were never asked to think about. And it gives them an easy out, because "nothing comes to mind right now" costs them nothing to say.

So most reps ask once, get a shrug, and conclude their customers don't refer. The customers were never given a real chance to.

Advocates Aren't the Same as Satisfied Customers

A survey response costs nothing. A referral costs the customer something real: their reputation with someone they know, spent on vouching for you. That's why a referral is a far stronger signal than a satisfaction score, and why most companies are measuring the cheap signal instead of the expensive one.

Before you ask anyone for anything, you need to know who's already shown you they're willing to spend that capital, because they usually tell you before you ask.

The Referral Engine

Step 1: Find the Right Handful Not your whole book. Look for customers who have already referenced you unprompted: mentioned you in a Slack community, forwarded your content to someone, left an unprompted compliment on a call, renewed early, or expanded without a pitch. Each of those is proof of advocacy you didn't have to ask for. That's your real candidate list, usually far shorter than your customer list.

Step 2: Time the Ask to a Moment They Just Earned Not a calendar reminder. Ask right after a moment of proven value: right after a renewal, right after a QBR where a real result got named out loud, right after they tell you something worked. The ask lands because the value is still fresh, not because it's the first of the quarter.

Step 3: Make the Ask Specific, Not Generic "Know anyone who could use us" makes them do the work. A specific ask does the work for them: name a role, a company type, or a situation close to their own. Better still, if you can see a name in their network worth asking about directly, ask about that name.

Why did one company's AI work, and another's didn't?

One had a dedicated owner. Resolution rate: 48.9%. One didn't: 0.38%. See the full breakdown.

See the data

The Advocate Extraction Prompt

Run this against your renewal notes, QBR notes, and any unprompted customer messages you've saved:

You are extracting voice-of-customer signal from the full post-sale
relationship, not just the most recent call.

From the attached notes and transcripts, covering onboarding,
implementation, check-ins, and any upsell or expansion conversations,
find every instance where the customer said something positive and
specific, not just "good," "happy," or "no complaints."

For each instance found, extract:
- The customer and their role
- Which stage it happened in: onboarding, implementation, check-in,
  upsell, or other
- The specific thing they said, quoted or closely paraphrased
- What kind of praise it is: a measurable result (time, cost, revenue,
  headcount), a comparison to something they used before, or an
  experience ("easier than I expected")
- A specificity score, 1 to 3: 1 means vague sentiment with nothing to
  build on. 2 means a real result, but thin on detail. 3 means specific
  and quotable as-is.

Then, for each instance:
- If it scored 1 or 2, draft one follow-up question the rep can ask on
  their next call to turn it into something usable
- If it scored 3, draft two things: a specific referral ask built from
  that exact language, and a short, low-pressure testimonial request
  asking permission to use the quote

Do not include generic positivity unless it's attached to a specific
reason or result. "Really happy with it" is not signal. "Cut our
turnaround from three days to one" is.

A Worked Example

Illustrative: a mid-market ops manager who renewed two months early and told you on the call that the rollout "saved her team a full day a week."

Run the prompt and it surfaces her as a real signal, not just a happy renewal: early renewal plus a specific, quotable result. The drafted ask isn't "know anyone else who could use us." It's built off her own words: a specific ask about anyone she knows running ops at a similar-sized team, sent the same week as the renewal, while the result is still the thing she'd tell someone about anyway.

Why This Works

The filter works because it separates a real signal from a cheap one. Satisfaction is easy to give and easy to forget. A referral costs something, so the customers who've already shown you they're willing to spend it are the only ones worth asking first.

The timing works because the ask rides on a moment that already proved the value, instead of asking someone to remember why they liked you three months ago.

And the specific ask works because it removes the one thing that kills most referral requests: making the customer do research they never agreed to do.

Getting Started

Pull your renewal and QBR notes from the last two quarters. Run the Advocate Prompt against them this week.

Don't ask everyone who comes back with a positive mention. Start with the two or three who show real advocacy signal, not just satisfaction, and send the specific ask while the moment is still fresh.

~ Jay

PS: Today’s issue finds the advocate and drafts your ask to them. For paid subscribers: below is the follow-up prompt that drafts the actual intro message your advocate sends on your behalf, the one that does the real work once they've said yes.

Click here for the tool stack I’d build today: https://www.sellingwithai.vip/stack

Author curated previous issues:

• S01E27: Your Closed-Won Accounts Already Wrote Your ICP

• S01E25: Champion Enablement: Give Your Internal Advocate the Tools to Win

• S01E07: Your Buyer Doesn't Have a Buying Process. Build Them One.

• S01E06: Build a Deal Risk Score Before Your Buyer Goes Dark

Click here for the entire archive: https://www.sellingwithai.vip/archive

Members Section

The Warm Intro Prompt

Once an advocate has agreed to make an introduction, most of them stall out because they don't know what to write. Draft it for them instead of leaving it to chance:

Hopefully you read the last issue… the one where I displayed this image… but also, more importantly, made you aware of the new day/time to expect to hear from me…. and well… that day is today, and that time is now. Hi 👋

Hey there. Welcome to E27… which, as a reminder, is the first Tuesday edition of Selling with AI. All future issues will be sent at 12pm EST, every Tuesday. 

Every sales org has an ICP slide. Industry, employee range, a tech stack bullet, maybe a logo wall at the bottom. Marketing built it a year and a half ago. Nobody has updated the file since.

Meanwhile the actual answer is sitting in the CRM, untouched. Every account that already said yes. Not a guess about who might buy. Proof of who did.

Credit where it's due: I didn't come up with this angle. Sales Bytes ran a piece in April called "The Closed-Won Lookalike Sequence." The core move: pull your closed-won accounts, find companies that look like them, let AI write the first line. That's the right instinct, and it's worth sending you to go read the original.

Sales Bytes

Daily AI intel for B2B sales teams that want to sell smarter

Subscribe

I want to take that instinct and push it past where the original workflow stops. Two changes. First, we're doing it in Apollo instead of Clay, because Apollo doesn't give you one "find lookalikes" button, it gives you the raw filters, and that's actually the upgrade. Second, we're not stopping at one match layer. We're stacking four.

This issue is for anyone doing their own pipeline generation with no research team behind them: founders, solo consultants, reps carrying their own book. If you've ever stared at a blank ICP field and typed something generic because you didn't know where else to start, this is for you.

Lovable just published a study on who is actually building software right now. The numbers tell the story better than I can:

Most builders on the platform come from non-technical backgrounds. A third are already earning revenue. 720 million monthly visits to products built by people who could not have shipped a line of code 3 years ago.

The old gatekeepers are gone. If you have been sitting on an idea, this is the moment.

Readers of this newsletter get an extra 10 credits at signup for using this link.

Start building on Lovable

Why One Match Layer Isn't Enough

Here's what "find lookalikes" actually does under the hood, in Clay or anywhere else. It takes your seed accounts, looks at firmographics, industry, headcount, maybe tech stack, and returns companies that score similarly on those dimensions. That's one layer: do you resemble my best customers on paper.

That's not a knock on the tool. It works. The problem is what it doesn't tell you: timing.

A company can match your ICP on paper and still be the wrong call this month. Same industry, same headcount, same tech stack as your best account, and also: no budget movement, no headcount growth, no open reqs anywhere near your problem. You'd message them anyway, because the lookalike button said yes, and you'd get the same silence you get from a cold list pulled off a filter.

The fix isn't a better lookalike algorithm. It's refusing to stop at one layer. A real match should stack firmographic fit with what's happening at that company right now, funding, growth, hiring, so you're not just finding companies that resemble your best customers. You're finding the ones that resemble your best customers and are showing signs of being ready.

Apollo is built for exactly this, because its organization search isn't a black box, it's a set of filters you control directly: employee count, industry keywords, tech stack, funding amount and date, headcount growth rate, and active job postings by title and date, all in the same search. You're not running four tools and stitching the output together. You're running one search with four conditions.

Step 1: Build the Fingerprint From What Actually Closed

Start with your last 10 to 20 closed-won accounts. Not your whole customer list, the recent ones, ideally the ones that renewed or expanded, since those are proof the fit was real and not just a fluke close.

Pull the real data on each one instead of guessing. Enrich each account and capture industry, employee count, revenue range, tech stack, and funding history. This is the part most people skip. They write "SaaS companies, 50 to 200 employees" from memory and call it an ICP. Your closed-won list will tell you the real pattern, and it's rarely the clean number you assumed.

Use this prompt once you have the enrichment data pulled:

You are helping me define an ICP fingerprint from real closed-won account data, not assumptions.

Here is the enrichment data for my last 10-20 closed-won accounts:

[PASTE ENRICHED ACCOUNT DATA: industry, employee count, revenue, tech stack, funding history, for each account]

Find the pattern:

1. The employee count range that repeats across the most accounts
2. Industries or keyword tags shared by more than half
3. Any technology more than half of these accounts have in common
4. The revenue range these accounts cluster around
5. Any shared funding pattern (recent round, amount range)

Flag any account that clearly doesn't fit the pattern and tell me to treat it as an outlier rather than let it widen the fingerprint. Return the result as a usable filter set: employee range, keyword tags, technologies, revenue range, funding range.

What comes back is your fingerprint. Not a slide. A pattern built from accounts that already paid you.

Step 2: Stack the Signals a Lookalike Button Can't See

This is the layer the original workflow doesn't have, and it's the whole upgrade.

Take the fingerprint from Step 1 and run it as an Apollo organization search. But don't stop at firmographic filters. Add at least one live signal layer in the same search:

• Funding: a funding amount and date range, so you're only pulling companies that raised recently, which usually means fresh budget and fewer sign-off layers

• Growth: a headcount growth percentage over a trailing window, so you're only pulling companies that are actively scaling, not flat or shrinking

• Hiring: active job postings, filtered by title and posting date, so you're only pulling companies currently recruiting for the exact function your product touches

You can run all three in the same search alongside the fingerprint filters. That's the difference between "find companies like my best customers" and "find companies like my best customers who are also showing a reason to act this month."

The list this returns will be smaller than a firmographic-only lookalike search. That's correct. You went from resemblance to resemblance-plus-timing, and half your original match list won't clear the bar. Good. Those are the accounts that would have gone cold anyway.

Step 3: Write the Brief From the Stack, Not the Guess

Now you have accounts that match on paper and have a live reason to care. Turn that into a research brief that actually uses both.

Use this prompt per account:

For this account, here is what matched my closed-won fingerprint, and what's happening right now:

Fingerprint match: [INDUSTRY, EMPLOYEE COUNT, TECH STACK OVERLAP WITH MY

CLOSED-WON PATTERN]

Live signal: [RECENT FUNDING AMOUNT/DATE, HEADCOUNT GROWTH %, OPEN JOB TITLES AND POST DATES]

Here are my closed-won case studies, one line each, with the specific problem each account hired me to solve:

[LIST YOUR CASE STUDIES]

Do three things:

1. Pick the closed-won case study closest to this account's profile and signals, and say why in one sentence
2. Write a one-paragraph brief on what this fingerprint-plus-signal combination likely means for their priorities right now
3. Write the first line of a cold email that names the live signal as the reason for timing and references the matched case study, without naming the client unless I confirm that's OK

Be specific. Flag every inference. If none of my case studies actually fit say so instead of forcing one.

The output reads nothing like a generic "noticed you're growing" opener, because it isn't guessing. It's citing a real pattern match plus a real, dated signal, and pointing at the closed-won proof that you've solved this exact shape of problem before.

Everything GTM. One platform.

Small teams don't have time to stitch together five tools and hope it works.

Apollo gives you everything you need to find leads, reach them, and close deals — all in one place:

• 230M+ verified contacts

• AI-powered outreach

• Data enrichment

• Inbound lead capture

• Meeting scheduler

• And more

Stop juggling tools and start building pipeline that scales.

With Apollo, the AI revenue engine powering 4M+ users.

Sign up for free

Step 4: Match the Buyer, Not the Account

A company match means nothing until you have the right person. Search for people at the matched accounts using the same discipline: filter by seniority and title, and where it's useful, by department headcount, so you're not guessing "VP of Data" off a LinkedIn search and hoping.

Match against the specific role that owns the pain your closed-won case study solved, not a generic title guess. If your best customers bought because a Head of RevOps owned the problem, search for that title specifically, not "operations leader" broadly.

Step 5: Launch It Without Leaving the Platform

The original workflow ends with "push to your sequencer," a CSV out of one tool and into Salesloft, Outreach, or Smartlead. That handoff is where lists go stale and formatting breaks.

Apollo skips it. Build the sequence in the same place you built the list: a short multi-step sequence (email, a LinkedIn touch, a call task) added directly against the matched contacts, using the brief from Step 3 as the first-touch content. Find, match, brief, and send, in one workspace, with nothing exported in between.

Why This Works

A fingerprint built from your own closed-won data beats a black-box match, because you can see every criterion that built it and keep refining it as you close more deals. It gets more accurate over time instead of staying static like a slide.

Stacking live signals on top of the fingerprint means you're not finding companies that merely resemble your best customers. You're finding the ones that resemble your best customers and have a reason to act this month. That's the difference between a list and a list with a shelf life you can actually work.

Doing it in one platform removes the handoff points where lists quietly rot. Every export-then-import step is a step where something gets dropped or goes stale before it reaches a sequence.

And if you're running this solo, with no ops person maintaining a multi-tool stack for you, a workflow you can run start to finish in one place isn't a nice-to-have. It's the only version you'll actually keep doing every week.

Rep Action this week

Pull your 10 best closed-won accounts. Not your whole list, the ones that renewed or expanded.

Enrich them. Build the fingerprint. Run one search that stacks the fingerprint with a single live signal layer, pick funding, growth, or hiring, not all three your first time through.

Compare that list to whatever your last "similar companies" list looked like. Notice how much shorter it is, and how much more you'd actually want to reach every account on it.

Jay

PS: Go read the original Sales Bytes piece if you haven't. Good idea, worth the credit.

And for paid subscribers: below is the fingerprint prompt built to run on 20 accounts in one pass, plus the filter cheat sheet mapping each signal type to the exact Apollo field that finds it.

Click here for a list of the tools I use

Usage-based pricing is reshaping B2B revenue.Usage-based pricing is reshaping B2B revenue. Watch Tabs + PwC break down the finance ops reality — and how to scale it without adding headcount.

Watch it now

This is the banner I’m referring to below… notice anything?

Hi there! Hope everyone had a fun and safe holiday weekend!

In the world of weekly newsletters, one of the first milestones we can look forward to is our 26th issue… which of course is the half-way point in a year, so its "officially” official that Selling with AI has hit its half-aversary. Yea, its not a word. I know. Kinda wish it was though.

Anyway, welcome to the halfway point, and hello and welcome to so many new subscribers! The growth over the past 6 months has been so fun to watch, and thank you so much for continuing to read and follow!

One more thing, did you see the banner at the top of this week’s issue? Did you notice anything? Every Tuesday… Starting with S01E27 (the next issue), Selling /AI will be delivered to your inbox every Tuesday at lunchtime… well, at my lunchtime (EST). A very wise man (you know who you are) told me :

BTW, thats literally a copy/paste too. like, I’m not trying to make a big deal out of this… but I love you bro. Thank you for being you.

With that being said, we’re all in sales here. And we all know what a persona is. When I began writing, I did so from the positioning of an account executive. Today I write to you as a solopreneur. And starting today, my topics are going to be expanded a bit to cover more founder-led sales tips and tricks… essentially, I am going to write to you weekly about what is working… and maybe even what’s not working.

And that begins with today’s issue, this FINAL Sunday issue, ever, of Selling with AI.

Your next 100 customers are already in Apollo

Find, reach, and close your perfect deals — without juggling five tools or hiring more reps.

Apollo gives you everything you need to build real pipeline, fast. From inbound to outbound, first touch to close.

All in Apollo.

Sign up for free

Outbound isn't dead. It's just all the same now.

Everyone opens with "congrats on the round." Everyone references the same hiring post. Everyone "noticed" the same thing on LinkedIn, or in the news recently. Buyers noticed this new age of hyper-personalization happening very quickly… and outbound changed.

The problem was never personalization. It's that personalization stopped being proof you did the work. AI made it free. And free proves nothing.

I get the same automated emails you do. I can name the sequence tool in the first line. Your buyers can too. The personalized opener that used to signal "this person actually researched me" now signals the opposite: a machine filled in a variable and hit send.

This issue is for anyone whose reply rates quietly fell off a cliff this year. SDRs and BDRs running outbound all day. Founders doing their own prospecting. AEs building their own pipeline. The fix is the same for all of you, and it isn't more volume.

Why Outbound Stopped Working

Walk through what actually changed.

Three years ago, a personalized first line took effort. You read their LinkedIn. You found a hook. You wrote something specific. The effort was the message. When a buyer got a genuinely tailored email, the tailoring told them someone chose them on purpose.

Then everyone got the same tools. The same data providers. The same AI writing the same "personalized" opener at a thousand accounts an hour. The tailoring got automated, which means it stopped being a signal of effort. It became a format.

And buyers can read the format.

These lines don’t say "I researched you" anymore. They say "you got added to a sequence." The exact thing that used to earn a reply now earns a delete.

Here is the trap most teams are in. They feel the reply rates dropping, so they reach for the obvious lever: send more. More accounts, more touches, better-sounding AI. That pours more water into a flood. It burns your domain, it trains buyers to ignore you faster, and it makes the core problem worse.

The volume game is over. Not because outbound stopped working. Because everyone is playing it with the same machine.

Prospect on Signals, Not Titles

The edge AI gives you was never the writing. Writing is the part everyone now has for free.

The edge is research at a scale a human can't match. One person cannot monitor four hundred accounts for the moment each one develops a real reason to buy. A model can. That is the job to hand the AI: not "write me a clever opener," but "tell me which of these accounts just gave me a reason to call."

That reason has a name. It's a signal.

A signal is a dated change inside an account that creates a reason to act now. A funding round. A new VP who owns your problem. A job posting for the exact pain you solve. A product launch. A tech-stack change. A reorg. A public statement from a leader about a priority.

Title-based prospecting starts with "who matches my ICP." That gives you a list. Signal-based prospecting starts with "who just changed." That gives you a reason. The list is what everyone has. The reason is what almost nobody brings.

Here is the four-step workflow that I built into my day to day. Find the trigger. Build the angle. Write from the signal. Sequence to its shelf life. Ten minutes per account, and most of that is the AI working while you do something else. (By the way, I use Perplexity for this).

See what enterprise-ready AI support looks like

How are leading teams getting AI support to work? We're breaking down the playbook, live July 9.

Save your spot

Step 1: Hunt Triggers, Not Titles

Start with a list of target accounts and ask the AI to find the one real reason to reach out to each. No reason, no outreach this cycle.

Use this prompt:

You are a B2B prospecting researcher.

Here is who I sell to and the problem I solve, in one sentence:

[WHO I SELL TO / THE PROBLEM MY PRODUCT SOLVES]

Here are my target accounts:

[PASTE COMPANY NAMES, OR A SINGLE COMPANY]

For each account, find the single strongest buying signal from the last 90 days.

Look for: funding, leadership changes, job postings tied to my problem area, product launches, M&A, expansion, public statements, or tech-stack changes.

For each account, return:

1. The signal (what happened, and the date if you can find it)
2. Why it creates a reason to act now, tied to the problem I solve
3. A confidence level: Strong / Weak / None, and flag where you are guessing

If an account has no real signal in the last 90 days, say "No signal" and tell me to deprioritize it. Do not invent a signal to fill the gap.

One thing that matters more than the prompt: run this in a model with live web access, or paste in what your sales intelligence tool already surfaced. A model working from training data alone will confidently hallucinate a funding round that never happened. The "flag where you are guessing" line is your safety net. Read it every time.

What comes back is a filtered list. Half your accounts will have no real signal. That is the point. Those accounts are not bad, they are just not now. You stop spending your best messages on people who have no reason to answer.

Step 2: Build the Angle

A signal is not a message. "They raised a Series B" is not a reason for them to talk to you. It's a reason for you to think.

This step turns the raw signal into a point of view: what the change means for their priorities, who inside the company now owns that problem, and the one observation a generic rep would never make.

Use this prompt:

For this account, I found this signal:

[PASTE THE SIGNAL FROM STEP 1]

Here is the problem my product solves:

[ONE SENTENCE]

Build me a short prospecting brief:

1. What this signal likely means for their priorities over the next two quarters
2. The role most likely to own the problem this signal creates (a title, not a name)
3. One specific, non-obvious observation about their situation that a generic rep would miss
4. A single sentence I could say that proves I understand their moment, without mentioning my product

Keep it under 150 words. Be concrete. If you are inferring, say so.

The fourth output is the one to study. If that sentence could be sent to any company in their industry, the angle is too generic and you keep working. If it could only be true for this account this quarter, you found the reason.

This is the part of the work AI can speed up but can't fake for you. Read the brief and decide if it's actually specific. Your judgment is the filter the model doesn't have.

Step 3: Write From the Signal (2 minutes)

Now the message writes itself, because you finally have something true to say.

The opener leads with their situation. The signal is the reason for the timing. Your product shows up late, if at all. And every automated tell gets banned at the door.

Use this prompt:

Write a cold outreach message (email or LinkedIn) to [ROLE] at [COMPANY]

The reason I am reaching out:

[THE SIGNAL + THE ONE-SENTENCE OBSERVATION FROM STEP 2]

What my product helps with:

[ONE SENTENCE]

Rules:

- Open with the specific situation they are in, not their title or my product
- Use the signal as the reason for the timing
- Make one clear, low-friction ask
- Under 75 words
- Banned: "I noticed," "I saw you're hiring," "congrats on," "hope this finds you well," flattery, and three-part sentence fragments
- It should read like a peer who did the work, not a sequence step

Then write one alternate version that is shorter and more direct.

The banned list is doing real work here, by ensuring those phrases that are quite literally the fingerprints of automated outbound are stripped and the message has to stand on the strength of the reason. If the reason is weak, you'll feel it here, because there's nothing left to hide behind. That's a feature, and its there by design.

Step 4: Sequence to the Signal's Shelf Life (2 minutes)

Triggers decay. A funding round is a hot reason for about a month, then it's old news everyone else already mentioned. A new executive is most reachable in their first 90 days, while they're still forming opinions. A job posting is live until it's filled.

A fixed seven-touch drip ignores all of that. It sends touch four three weeks after the reason went stale. Time the sequence to the signal instead.

Use this prompt:

Build a 3-touch outreach sequence for [ROLE] at [COMPANY] based on this signal:

[THE SIGNAL]

For each touch:

1. Space it according to how time-sensitive this signal is, and tell me the gap in days
2. Give it a distinct angle: a new observation or resource, never "just following up"
3. Keep each touch under 60 words

This signal has a shelf life. If it goes stale in 30 days, compress the sequence and tell me when to stop and move on.

The instruction to name when to stop is the part most reps need. Outbound discipline isn't sending more. It's knowing when a reason has expired and moving your attention to an account where a new one just appeared.

Why This Works

Four reasons.

Relevance is the only scarce thing left. AI made writing, formatting, and surface personalization free. The one thing it can't manufacture is a genuine reason to reach out today. When you lead with that reason, you're trading in the only currency that still has value.

A signal is permission. You're not interrupting someone at random. You're responding to something real that just happened in their world. That changes how the message reads, from a stranger selling to a peer paying attention.

Fewer, better touches protect what volume destroys. Every irrelevant blast trains the buyer to ignore your name and trains the mailbox provider to bury your domain. Signal-based outreach sends less and lands more, which keeps both your reputation and your deliverability intact.

You get there first. The rep who shows up the week of the trigger, while the priority is fresh and the budget conversation is just starting, beats the rep who shows up in next quarter's list pull. Timing was always the edge. Signals are how you find it.

Rep Action this week

Pull 10 target accounts. Not 100. Ten.

Run the Trigger Scan on all of them. Delete every account that comes back "No signal," even if it hurts. You're probably cutting half.

For the survivors, run the angle and message prompts. Send 5 messages built on real signals this week instead of 50 built on a template.

Then watch your reply rate, not your send count. The whole point is to make the smaller number win.

Jay

PS: See you on Tuesday… and for the paid subscribers, the playbooks being dropped over these next few weeks…. wow. You’re welcome in advance… haha, jk. But… no really… its good stuff.

I frequently get asked which tools I use/recommend.

Selling /AI Tools of 2026

Want to get the most out of ChatGPT?

ChatGPT is a superpower if you know how to use it correctly.

Discover how HubSpot's guide to AI can elevate both your productivity and creativity to get more things done.

Learn to automate tasks, enhance decision-making, and foster innovation with the power of AI.

Download the free guide

Lets begin

The discovery call went as expected. Next comes the demo. As you are walking through the platform, you can tell that your champion is completely invested. They're asking the right questions. They're engaged. Your champion is sold. That's not the problem.

The problem is what happens when they the demo ends. They schedule 30 minutes with their VP. They open with the pain point. The VP asks two questions they didn't expect. The follow-up they promised you never comes. You send a check-in. They say they're "still working through internal alignment." Two weeks later, the deal slips.

You did everything right externally. The deal died inside the account.

Most reps diagnose this as a closing problem and push harder. The actual gap is earlier, and it belongs to you. Your champion walked into that room without the right tools. Nobody gave them to him.

The internal sell is a different sale. The person doing it is not a sales professional. That combination is why late-stage deals collapse with no warning.

Why the Internal Sell Fails

Your champion has been on every single call. They've seen the demo. They've heard you handle every objection. None of that transfers to a room with their CFO.

Because when they walk in there, they're not selling a product. They're selling a prioritization decision. "We should spend budget on this, now, over the five other things competing for your attention." That requires a financial frame. It requires speaking in the language of outcomes their leadership cares about. It requires handling a "why now" from someone who was never in your calls and has no reason to feel urgency.

Your champion is a subject matter expert on the problem. They are not equipped to frame the solution the way an economic buyer needs to hear it. That gap is yours to close.

Most reps hand their champion a demo recording and call it enablement.

Three Tools. Fifteen Minutes.

Build these before the champion's next internal meeting.

Step 1: The Internal Business Case (5 minutes)

The business case your champion needs is not the ROI deck you built for your sales process. That document positions your product. It has your logo on it. Nobody forwards that to their CFO.

What your champion needs is a business case written as if they authored it. The problem framed in terms their leadership cares about. The cost of doing nothing. The outcome the organization is buying.

Use this prompt:

My champion at [COMPANY] is [TITLE]. They are building internal support to move forward with [CATEGORY/SOLUTION]. Their economic buyer is [ECONOMIC BUYER TITLE].

Here is what I know about their situation:

[PASTE YOUR DISCOVERY NOTES / CALL TRANSCRIPTS / DEAL SUMMARY]

Write an internal business case from the champion's perspective, not the vendor's.

It should:

1. Frame the problem in terms that land with [ECONOMIC BUYER TITLE] — not in product terms
2. Quantify the cost of inaction where possible (time, revenue, headcount, risk)
3. State the expected outcome in business language, not feature language
4. Include a "why now" rationale tied to something real in their business context
5. Stay under 300 words — this is a verbal talking track, not a document

Do not mention [YOUR COMPANY NAME] by name more than once.

The champion is the author. Write as if this is their independent point of view.

The instruction to limit mentions of your company is the most important line in that prompt.

An internal business case that reads like a vendor pitch gets forwarded to procurement. It needs to sound like the champion arrived at this conclusion on their own.

What comes back is a talking track. A few paragraphs your champion can internalize and deliver in their own words. It will be sharper than anything they'd write themselves, and it will land because it sounds like them.

Step 2: The One-Pager They Can Forward (5 minutes)

A one-pager is not a summary of your pitch deck. It exists to do one thing: help your champion get a meeting with the next stakeholder.

Most reps either don't provide one or they provide a product sheet. Product sheets are built for buyers who found you on G2. They are not built for an internal budget conversation with someone who has never heard of you and did not ask to be in this process.

Use this prompt:

Write a one-page internal document my champion can forward to [ECONOMIC BUYER TITLE] 
to request a 20-minute conversation.

Context:

Company: [COMPANY]

Champion: [CHAMPION TITLE]

Economic Buyer: [ECONOMIC BUYER TITLE]

The problem being solved: [1-2 sentences from your discovery notes]

What success looks like in their words: [What the champion told you]

Urgency or timing factor: [If any — if none, say so]

The document should:

- Be written as if the champion authored it, not the vendor
- Open with the business problem, not the solution
- State clearly what the champion is asking for (a 20-minute call, a review, a decision)
- Include one number or metric if there is a credible one from their context
- Fit on a single page — under 250 words

Do not use sales language. Do not include pricing. Do not mention competitors.

Also write the two-sentence email they send when forwarding this, with a subject line.

That last line matters. The forwarding email is often what gets read first. If the subject line sounds like a vendor introduction, the document never gets opened.

Step 3: Internal Objection Prep (5 minutes)

The objections your champion will face inside the organization are not the ones you've been handling on calls.

You've handled pricing questions and product comparisons. Your champion is going to face budget prioritization ("we have five other things this quarter"), IT skepticism ("we need to evaluate this properly before anything gets deployed"), risk aversion ("what happens if this doesn't deliver"), and the status quo defender ("nothing is broken enough to justify this").

Most champions get blindsided because no one told them those questions were coming.

Use this prompt:

My champion at [COMPANY] is building internal support for [CATEGORY/SOLUTION]. 
They will have conversations with these stakeholders: [LIST THE ROLES]

For each role:

1. Name the most likely objection or concern that role will raise
2. Write a 2-3 sentence response the champion can use
3. Suggest one question the champion should ask to move the conversation forward
   rather than defend the decision

The responses should sound like a thoughtful internal colleague, not a vendor script.

Do not use sales language. The champion is not selling - they are advocating.

What comes back is a preparation framework. Your champion reads through it before the meeting, sees the questions they'll face before they face them, and walks in with answers instead of hoping they don't get asked.

The difference between a champion who is prepped and one who isn't is visible in the first two minutes. The prepped champion addresses concerns before they become objections. The unprepped champion gets defensive when pushed.

Why This Works

These three tools address three distinct failure modes in the internal sell.

The business case fails when it reads like a vendor pitch. The champion sounds like they're selling you, not advocating for a decision. The prompt fixes that by centering the champion as the author and limiting direct product references.

The outreach fails when it asks for attention without earning it. A one-pager that opens with the business problem instead of the product description respects the economic buyer's position. They get to evaluate, not get sold.

The internal meeting fails when the champion runs out of answers. Objection prep doesn't script the conversation… it maps the terrain. A champion who has seen the questions before is a completely different advocate than one who hasn't.

Every one of these outputs can be built in five minutes. Most reps will never build them because the assumption is that a champion who believes in you will figure it out.

Sometimes they do. When they don't, you don't find out until the deal is already cold.

Rep Action this week

Pull your two most at-risk deals - the ones where you have a champion you trust but whose internal process you can't see.

For each one: run the business case prompt, then the one-pager prompt. Send both to your champion before their next internal conversation. Tell them to use what lands and drop what doesn't.

After that meeting, debrief. Ask them what questions came up that the materials didn't address. Feed the answers back into the prompts. Two iterations from now, the materials are built for your specific deal, your specific account, and your specific champion.

~ Jay

Everything above makes your champion a better messenger. Sometimes that's not enough. 

Because the prompts can't fix the core problem: your champion still has to be in the room. The CFO reviews your ROI at 9pm on a Tuesday. The CISO who was never on a single call forms an opinion from a compliance doc or worse, outdated research online. The user who has a specific question about a feature not addressed. Your one-pager doesn't answer them and your champion may not be able to either. And with 83% of the buying process happening outside of interactions with sales, this problem is only growing. This is where understanding degrades and deals quietly slip.

ENaiBLD is the digital solution expert in the room when you can't be. It's a Buyer-Enabled Evaluation System - a sales-governed expert every stakeholder can ask anything, directly, on their own time. The CFO, the CISO, the skeptic: they get accurate, on-message answers, without your champion relaying them.

Here's the part reps care about. You don't just close the gap - you see into it. ENaiBLD captures every question, theme, and stakeholder, and surfaces the risks and objections forming while you're not there. You walk into the next meeting already knowing what each person cares about, and which concern(s) could quietly kill the deal.

Arm your champion. Then put an expert in every room you can't get to: enaibld.com

The best voice models, now fully orchestrated across all channels

ElevenAgents puts full orchestration on the voice models the market builds around. Voice, text chat, transcription, and reasoning in one integrated stack, <400 milliseconds, and human-sounding. Plug in any LLM, integrate tools, A/B test, and deploy across channels. More human-like conversations, lower latency, flat $0.08 per minute.

Start building free

Turn AI into Your Income Engine

Ready to transform artificial intelligence from a buzzword into your personal revenue generator

HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.

Inside you'll discover:

• A curated collection of 200+ profitable opportunities spanning content creation, e-commerce, gaming, and emerging digital markets—each vetted for real-world potential

• Step-by-step implementation guides designed for beginners, making AI accessible regardless of your technical background

• Cutting-edge strategies aligned with current market trends, ensuring your ventures stay ahead of the curve

Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.

Get Your Guide

Let's begin

We’ve all been there…

You're mid-deal. Things are moving. Then the prospect drops a name.

A competitor. Maybe they're casual about it. Maybe they say they're "also evaluating" them. Maybe it comes up right as you're about to close for next steps, right at the moment you thought you had momentum.

Your brain does something in that half-second. It searches for the right thing to say.

Most reps land in one of three places. They freeze, buy time with a filler, and hope the conversation moves on. They pivot... "great product, but here's what makes us different"... and launch into a rehearsed comparison nobody asked for. Or they attack. "Oh, [Competitor]? We hear that a lot. Here's why customers leave them."

None of those work.

The freeze signals you don't know your own market. The pivot signals you've been waiting for this and you're about to sell, not listen. The attack signals something worse: fear. Buyers don't trust reps who tear down alternatives. They wonder what the rep says about them when they're not in the room.

What makes it worse: most reps don't even have the right information to react from. The battle card is six months old. The pricing column is outdated. The differentiation points are generic. They're improvising, without a foundation.

There's a fourth option. And it starts before the call.

Why Battle Cards Fail

Every sales team has battle cards. Most of them are sitting in a shared drive, updated once a year at SKO, already wrong.

Here's the problem with static competitive intelligence: it's built for the average deal against the average version of a competitor. That account doesn't exist.

Your prospect is a specific company. They have a specific tech stack, a specific use case, a specific reason they're on your list and their list at the same time. The competitor looks completely different depending on who's buying from them. Their strengths in enterprise don't translate to mid-market. Their pricing at one segment doesn't apply at another. Their G2 reviews from eighteen months ago don't reflect what shipped last quarter.

The rep who memorized a battle card six months ago doesn't know any of that. They know a generic comparison that was already outdated when it was written.

And the prospect knows their options better than you think. They've done the research. They've watched demos. They've talked to references. When you pull out a generic comparison, they notice.

The fix isn't a better battle card. It's a different approach entirely.

Build the Brief On Demand

Ten minutes. That's all it takes to build a competitive brief specific to your deal, your prospect, and what's actually true today.

Not a battle card. A brief. One page. Built that morning. Specific to that account.

Four prompts. Four steps. Here's how.

Step 1: The Competitor Snapshot (3 minutes)

You want to know what's actually changed in the last six months. Not their homepage positioning. Not their pitch deck copy. What's new, what's broken, and what customers are saying right now.

Use this prompt:

Give me a current overview of [Competitor]. Cover their recent product releases,
pricing changes, funding news, and customer review themes from G2 or Gartner.
Focus only on the last 6 months. Flag anything that's changed significantly
from their standard positioning.

What you're looking for:

• Pricing changes (have they moved upmarket, downmarket, or shifted to usage-based?)

• Recent releases (what did they ship? what does that tell you about their direction?)

• Review themes (what do happy customers say? what do churned customers cite?)

• Job postings (what are they hiring for? signals where they're investing or struggling)

A competitor that just raised a Series C and started hiring enterprise sales reps is a different conversation than the version of that same company from a year ago. Know which one you're walking into.

Step 2: The Deal-Specific Lean (2 minutes)

This is the step most reps skip entirely. It's the most important one.

Not every prospect is equally likely to choose a competitor. Some accounts are primed for them based on industry, existing tools, or use case. You need to know if this prospect is one of them, and why.

Use this prompt:

[Prospect] is a [industry] company using [relevant tools in their stack],
focused on [their primary use case]. What would make [Competitor] look
attractive to them specifically? What are the most likely reasons they'd
favor that vendor over alternatives?

This gives you the prospect's probable reasoning. Not a generic objection list.

If the output doesn't apply to your deal, you've already learned something. The competitor mention might be low-stakes. They're kicking tires or satisfying a procurement requirement.

If the output hits close to home, you know what you're actually up against. Now you can address it directly instead of talking past it.

Step 3: The Tradeoff Builder (3 minutes)

This is where most competitive training breaks down. Reps are taught to position their strengths. What they should be doing is naming tradeoffs.

"We're better at X" is a claim. The prospect has no reason to believe it.

"If you go with them, you get X. What you're trading is Y. That matters when Z" is a tradeoff. It's honest. It respects the prospect's intelligence. And it reframes the decision from "who's better" to "what does each choice actually mean for us."

Use this prompt:

Write 3 honest tradeoff statements comparing [Your Company] to [Competitor]
for a [industry] buyer focused on [use case]. Do not attack the competitor.
Show what each choice means in practice. Format each as:
"If you go with [Competitor], you get [X]. What you're trading is [Y].
That matters when [Z]."

Example output:

• "If you go with [Competitor], you get faster time to deploy. What you're trading is flexibility. That matters when your process doesn't match their default workflow."

• "If you go with [Competitor], you get a lower entry price. What you're trading is scalability. That matters when your team grows past 50 seats."

• "If you go with [Competitor], you get deeper integrations with [specific tool]. What you're trading is visibility across the full pipeline. That matters when leadership needs a single source of truth."

Three honest statements. No attacks. The prospect can evaluate them. That's the point.

Step 4: The Reframe (2 minutes)

Now you know what to say when the moment comes. Not a script. A posture.

The goal: acknowledge the competitor fairly, invite curiosity about fit, and position the decision as a tradeoff, not a competition.

Use this prompt:

Write 2-3 responses to: "We're also evaluating [Competitor]."
Each response should acknowledge the competitor fairly, invite the prospect
to explore what fit means for them specifically, and position the choice
as a tradeoff rather than a competition. Keep each under 3 sentences.

What the output sounds like in practice:

• "Makes sense. They're strong in [area]. The question I'd want to explore with you is how [specific use case] plays out in each environment."

• "Good company. We actually see them in a lot of deals. What would help is understanding what's most important to you in this decision, so we can be direct about where we're the right fit and where we're not."

Notice what's not there. No "the problem with [Competitor] is..." No "we beat them on every metric." No pivoting immediately to your deck.

You're treating the prospect like someone who can make a good decision with good information. Most of the time, that's all it takes.

Why This Works

Four reasons.

Specificity beats memorization. A brief built for this deal tells a completely different story than a generic comparison slide. The prospect can feel the difference.

Tradeoff framing beats attack. Buyers trust reps who understand the alternatives. Buyers distrust reps who dismiss them. When you can articulate an honest tradeoff, you position yourself as someone who knows the market, not someone who's afraid of it.

Timeliness matters. A battle card from six months ago has wrong pricing, outdated features, and stale review data. You don't know what you don't know. Ten minutes of real research fixes that.

It shifts the dynamic. Instead of reacting to a competitor mention, you become the person who surfaces the tradeoffs before the prospect has to. That's a different kind of rep. Buyers notice it.

Getting Started

Pick the competitor that comes up most often in your current pipeline. Run the four prompts tonight. It takes ten minutes.

Use the brief on the next call where they surface. After the call, note what landed and what didn't. Adjust the prompts. Rebuild the brief before the next time.

Over four weeks, you'll have a living brief for your top two or three competitors. Better than anything in the shared drive. Built for the deals you're actually running.

The battle card is already wrong. The brief you build tonight isn't.

The docs platform Anthropic, HubSpot, and Coinbase trust

Your documentation is a product decision. The companies building the most-used developer tools in the world chose Mintlify because great docs reduce support load, accelerate adoption, and convert evaluators into customers. Backed by a16z and Salesforce Ventures, Mintlify powers 20,000+ companies with AI-native documentation that keeps pace with your product — without pulling engineers off the roadmap to maintain it.

Book a demo

See Why HubSpot Chose Mintlify for Docs

HubSpot switched to Mintlify and saw 3x faster builds with 50% fewer eng resources. Beautiful, AI-native documentation that scales with your product — no custom infrastructure required.

Simplify Your Docs Today

Let's begin

Most reps have one real contact per deal. They call it a champion. What they actually have is a single point of failure.

Think about what happens when that person goes quiet... pulled into another initiative, loses internal support, takes a two-week vacation, gets restructured out of the decision entirely. The deal goes with them. The rep sends a follow-up and gets silence. Sends another. More silence. Somewhere around the third attempt, they start writing the breakup email.

By then the deal has been dead for three weeks. The rep just didn't know it yet.

Here's what actually happened: the rep built a relationship with one person and called it a pipeline.

The Real Reason Multi-Threading Doesn't Happen

Every sales leader in the world tells their team to multi-thread. It's in every kickoff deck, every pipeline review, every win-loss retro. And yet every single week, reps open their CRM to deals that have exactly one contact attached.

The advice isn't wrong. The timing is.

Reps treat multi-threading like a rescue operation. The deal stalls, the champion goes dark, and now they scramble to find someone else inside the account who will return an email. That's the wrong model. You don't multi-thread to save a dying deal... you multi-thread from day one so deals don't die in the first place.

The difference isn't tactics. It's mindset. One rep sees their champion as their deal. The other rep sees their champion as their entry point into an organization they need to understand.

The second rep wins more. And when they lose, they know exactly why.

What the Full Buying Group Actually Looks Like

Every B2B deal has the same cast of characters. Not always by title... by role.

The Economic Buyer signs the check, and often never shows up on a call until it's time to approve the deal or kill it. The Champion wants you to win, and has political capital they're willing to spend. The Users live in the product every day and care about workflow, not ROI. The Blocker has concerns, a competing vendor, or just doesn't want change... and they're hard to identify until it's too late (more on them in a minute). Then Legal and Procurement show up at the end with a list of questions nobody warned you about.

Most reps have met the Champion. Maybe one User. That means they have partial intelligence on a decision that involves five to eight people.

No wonder deals stall without explanation.

Step 1: Map the Buying Group After Every Discovery Call

Within 24 hours of your first real conversation, run this prompt. Give it your call notes and whatever you know about the account.

Prompt: Buying Group Mapper

You are a senior enterprise sales strategist.
Here are my notes from a discovery call:

[PASTE YOUR CALL NOTES]

Here is what I know about the company:

[COMPANY SIZE / INDUSTRY / TECH STACK / ANYTHING RELEVANT]

Based on this, map the full buying group for this deal.

For each role below, tell me:

1. Did they appear on the call? If yes, name and key quote.

2. Were they mentioned but absent? By whom and in what context?

3. If not mentioned at all, explain why they likely exist in this deal anyway.

Roles to map:

- Economic Buyer (budget authority)

- Champion (internal advocate)

- End Users (people in the product daily)

- Potential Blocker (skeptics, status quo defenders, competing vendor advocates)

- Technical / IT Stakeholder

- Legal or Procurement

Then give me a Coverage Score from 1-10 (1 = flying blind, 10 = full picture)

and the three most important gaps to fill before the next call.

Run this after every discovery call. Not after deals stall... after discovery.

What comes back will look different from what you expected. You'll see roles mentioned in passing that you never followed up on. You'll see who your champion referenced when they described the internal process, and whether you've actually spoken to that person. You'll see how many gaps you've been quietly treating as non-issues.

Step 2: Score Your Coverage Like You Score Your Deals

A deal with one champion and no economic buyer access isn't a 7 out of 10. It's a 3.

Most reps score deals on surface signals: engagement level, pain validation, timeline, budget confirmation. Those matter. But they only mean something if the right people confirmed them. Budget isn't confirmed until the Economic Buyer says it. Timeline isn't real until whoever owns approvals agrees to it. Technical fit isn't validated until IT has seen it.

If only your champion confirmed those things, you have their opinion. Not a deal.

After running the buying group mapper, add this to your deal notes:

Coverage Snapshot

Economic Buyer: [Met / Mentioned / Unknown]
Champion: [Confirmed / Assumed]

End Users: [X of estimated Y engaged]

Blocker Identified: [Yes / No / Unknown]

Technical / IT: [Met / Mentioned / Unknown]

Legal / Procurement: [Engaged / Not Yet / Unknown]

Coverage Score: X/10

Biggest Gap: [Role + why it matters for this specific deal]

Next Action: [Specific step, not "follow up"]

Four minutes. Run it after every meaningful interaction.

Over time the pattern becomes impossible to ignore. Your won deals have higher coverage scores at the midpoint. Your lost deals had gaps you didn't close in time.

Step 3: Build a Contact Plan for Each Gap

Knowing you have gaps is half the job. The other half is doing something about it.

For each role you haven't reached, you have two options: ask your champion to make an introduction, or reach out directly. Most reps default to the introduction because it feels safer. But relying on your champion to open every door is just another form of single point of failure.

Use this prompt to build a contact plan.

Prompt: Stakeholder Contact Strategy

I am working a deal at [COMPANY NAME].
Here is what I know about the deal so far:

[PASTE YOUR DEAL SUMMARY]

I have not yet connected with these roles:

[LIST THE GAPS]

For each missing role:

1. Write a two-sentence hypothesis about their likely concerns based on the deal context.

2. Recommend champion introduction or direct outreach, and explain why.

3. If direct outreach: write a three-sentence message for LinkedIn or email.

   Focus on their role and concerns. Do not mention my product.

4. If champion introduction: write the exact ask I should make to my champion,

   including the framing to use so it doesn't feel like I'm asking them to do my job.

What comes back isn't a generic cold message. It's a specific note for the Economic Buyer that references why this decision lands on their desk. A different note for IT that addresses integration concerns before they become objections. An introduction ask that gives your champion a clear, easy-to-forward message. Each one grounded in what you already know about the account.

Step 4: Treat Coverage as a Stage Gate

Deals move through stages. They also move through coverage.

Stage 3 (Technical Validation) means nothing if IT hasn't actually validated. Stage 4 (Proposal) means nothing if the Economic Buyer doesn't know a proposal is coming. Stage 5 (Closing) means nothing if Legal hasn't seen the contract.

So before moving a deal to the next stage, ask one question: who needs to have confirmed this, and have I actually spoken to them?

That's not a confirmation. That's a secondhand opinion. Get in front of the person who owns that piece of the decision. Every time.

Why This Works

Multi-threading isn't about redundancy. It's about intelligence.

Every conversation gives you new information. The Economic Buyer sees the problem differently than your champion does. IT has concerns your champion doesn't know about. End users have workflow needs that change the implementation plan. More contacts means more signal... and more signal means fewer surprises at the end.

The deals that collapse in stage 5 have the same autopsy almost every time: a stakeholder showed up late with a concern the rep never saw coming. That stakeholder was in the deal the whole time. The rep just never found them.

Finding them in week two is uncomfortable. Hearing from them in week nine is a lost deal.

Rep Action this week

Pull up your three most active deals and run the Buying Group Mapper on each one. Use whatever call notes, emails, or CRM data you have. It doesn't need to be complete. It just needs to be what you actually know.

Then look at what comes back. For every gap that touches budget authority, approval power, or veto power... treat that gap like what it is: a deal risk.

Run the Coverage Snapshot after your next call. And the one after that.

The reps who consistently win aren't just better at selling. They're better at knowing who's in the room.

Jay

Want to get the most out of ChatGPT?

ChatGPT is a superpower if you know how to use it correctly.

Discover how HubSpot's guide to AI can elevate both your productivity and creativity to get more things done.

Learn to automate tasks, enhance decision-making, and foster innovation with the power of AI.

Download the free guide.

Let's begin

Five weeks ago you had an AI with no memory. Today you have something most reps will never build.

• A Main Branch that tells your AI who your buyer is, what your product does, how to handle every objection your team has heard, what your process looks like, and how your team sounds.

• A Personal Branch that carries your account history, your voice, your deals, your read on your market.

• A Prompt Library that turns all of that context into repeatable, improving action.

Three layers. Sixteen files. Built in five weeks.

There's one more thing. Everything you built is invisible to everyone else on your team. The objection handle that works in your segment? Locked in your personal branch. The cold opener format that gets replies? Same. The prompt you've run 50 times and send without editing? Nobody else can load it. Meanwhile, the Main Branch your team is running from is still the same version it was when you started. Every rep is making mistakes you already figured out. Every new hire is going to spend 60 days reinventing things you discovered in week two.

The merge is the last step. Not because the system isn't complete without it. It is. You've already built something that makes you better. The merge is what makes the team better.

What Goes Back and What Stays

Not everything earns a spot in the team layer. Most of what you built is personal and should stay that way. Your account history, your deal context, your specific relationship intelligence: that's yours. But three things bubble up.

The first is objection handles. If you've used the same handle ten or more times and it works reliably in your segment, it belongs in the team's objections.md. Not because it's clever. Because someone who joins next quarter is going to face that same objection in their first week and currently has nothing to work with.

The second is voice examples. If you wrote an opener this quarter that got a meaningful reply rate, that example belongs in the team's voice.md. You strip the account-specific details and the pattern stays. That's what gives the next rep a real example to learn from instead of a description of what good looks like.

The third is prompts. If a prompt in your library has hit fifty runs and you send the output without editing, that structure is too valuable to keep to yourself. Strip the personal context load, make it generic, propose it as a team prompt. A new rep runs it on day one and gets output at the level it took you fifty iterations to reach.

The Merge Protocol

This is not a weekly practice. It's quarterly. Most of your personal branch stays personal, and that's right. Four steps: review what you've built, flag the candidates, run the filter, propose the addition.

The filter is one question: would this help a new rep on their first call into my segment? That's it. If yes, propose it. If no, keep it. You don't need a rubric or a scoring system. You need to be honest about what's genuinely useful versus what's specific to your situation. The proposal is one message to whoever owns your Main Branch. Here's what worked. Here's why it belongs in the shared layer. Let them decide if it fits. Your job is to surface what you learned, not to maintain the Main Branch yourself.

What the Team Gets

Without this practice, your Main Branch is a snapshot. It reflects what your team knew when someone first sat down to build it, and it doesn't update unless someone explicitly decides to update it, which usually means it doesn't. With this practice, your Main Branch is a living system. Every rep's best quarterly discovery has a path into the shared layer. The rep who joins six months from now loads a Main Branch that has two quarters of real-world improvements baked in. They start where your best rep today finished.

That's the compounding, and it runs both directions. When the Main Branch improves, your next Personal Branch build gets better faster. You're not starting from a frozen baseline. You're starting from the current state of the team's collective knowledge, and your personal layer builds on top of something that already moved.

The Complete System

Let's close the loop.

E17: your AI had no memory. You had great notes and no way to give them to the tool. E18: you learned the architecture. Two layers, shared and personal. A place to put everything.
E19: you built the Main Branch. Five files, specific enough to be useful, built in an afternoon.
E20: you built the Personal Branch. Four files only you could build. Account history, your voice, your deals, your market read.
E21: you built the Prompt Library. Prompts that load your context, specify the output, define what done looks like, and get better every time you run them.
E22: the merge. The quarterly practice that keeps the Main Branch alive and turns individual wins into team infrastructure.

That's the system. Not a product. Not a platform. A set of files you own and a set of practices that keep them useful. Most reps won't build this. They'll keep opening new chats from scratch, losing the prompt that worked, starting where they started. The rep who builds this starts every week further ahead than the week before. The team that builds this starts every quarter further ahead than the quarter before.

What the Quarterly Review Looks Like

Five checks. One quarter. Thirty minutes. Review your objections work, your voice examples, your territory observations, your prompt library. On each one, ask the same question: would this help a new rep on their first call into my segment? Most items won't pass, and that's right. You're not trying to push everything into the team layer. You're looking for the one or two things that genuinely belong there. The merge is a filter, not a fire hose.

When something passes, write a one-paragraph proposal and send it to whoever owns the Main Branch. That's the whole practice.

Rep Action this week

You're done building. Now run the system. Open a new chat. Load your full codebase: main-branch/ files, personal-branch/ files, and the two prompts you built last week. Use pre-call-prep.md on your next account call. Note what you'd change. Update the prompt. Load it. Run it. Update what needs updating. The system compounds from here.

10x the context. Half the time.

Speak your prompts into ChatGPT or Claude and get detailed, paste-ready input that actually gives you useful output. Wispr Flow captures what you'd cut when typing. Free on Mac, Windows, and iPhone.

Try Wispr Flow free

Better cap table management starts here

Cap table management doesn’t have to be frustrating. From issuing grants to 409A valuations or ASC 718 reporting Pulley can make it simple. 

Just ask Linear. They knew they needed a partner who could handle the complexity of their equity management. That’s why they migrated to Pulley.

Learn why Linear chose Pulley

The Prompt Library

Let’s begin.

Three weeks in, you have something most reps don't.

You have the Main Branch: five files that tell your AI who your buyer is, what your product does, how to handle objections, what your process looks like, and how your team sounds.

You have the Personal Branch: four files that carry your account history, your voice, your deals, your read on your territory.

Every chat you open starts from a foundation now.

That's real.

That took work.

And it's more than most people in your org will ever build.

But here's what's still happening.

Every morning you open a new chat, load your files, and then type a prompt from scratch. Different instruction every time. Nothing saved. Nothing that builds. You prep for a call by describing what you need in fresh language. You write a follow-up opener starting from zero. You ask for help with an objection and frame it differently every time.

The codebase is the memory. The prompt is the question. You built the memory. The questions are still one-offs.

That's the half most reps leave unfinished.

And it's not obvious until you've been running the system for a few weeks and realize:

The Difference That Compounds

A one-off prompt works once. When the chat closes, it's gone. Tomorrow you write something similar from memory. Or you spend three minutes trying to recreate the phrasing that worked last time. Either way, nothing improves. There's no baseline. No iteration. Just the same starting line every day.

A prompt asset is different. It's a saved file with a name.

It has:

• The right context load at the top

• A specific instruction

• A defined output format

• A one-line quality bar that tells you when it's working

You run it. You read what comes back. If it needs more than three edits, you update the file. Then you run it again next time. The output gets better. Not because the AI learns. Because you're improving the prompt. That's the whole thing.

Your codebase is the memory. Your prompt library is the muscle memory.

One without the other is a half-built system.

What Goes in the Library

Not every prompt belongs here.

The rule is simple:

Seven prompts are worth building for most reps:

1. Pre-call prep
   
   Turns your account context into what you need to know before you walk in.

2. Cold opener
   
   Generates a personalized first touch in your voice based on a specific trigger.

3. Follow-up
   
   Drafts the post-conversation email that references what was actually said.

4. Post-call notes
   
   Captures what changed and what needs to go in accounts.md or deals.md before you close the tab.

5. Objection response
   
   Handles the specific objection in the context of this deal and this stage.

6. Deal review
   
   Gives you a Friday honest assessment of where things actually stand.

7. Pipeline digest
   
   Creates the end-of-week summary you'd send to your manager.

Seven prompts.

That's the whole library for most reps starting out.

Build them well once and you won't rewrite them again.

You'll only refine them.

What One Actually Looks Like

Every prompt asset has four parts.

1. Context Load

The context load tells the AI which files to read before it does anything. Teal files from the Main Branch. Amber files from the Personal Branch. Without this, the prompt is just a better one-off. With it, the output starts from everything you've built.

2. Instruction

The instruction is what you want done. The more specific it is, the less you edit the output. "Write a pre-call prep note" is too vague. "Give me three things to know about this account, two things to confirm on this call, and one thing to listen for" is specific enough to produce something usable.

3. Output Format

The output format tells it how to structure the response. Bullets only. No preamble. No "great question, here's what I think." Format instructions cut your editing by half.

4. Quality Bar

The quality bar is one sentence that defines what done looks like. If the output references the specific account by name and situation, it's working. If it sounds generic, the context load isn't specific enough. That one sentence gives you a test you can run every time.

How It Compounds

First run:

You edit six things. You note what didn't land. You update the instruction.

By run five:

You edit three things. The format is right. The tone still drifts occasionally.

By run twenty:

You add voice.md to the context load. Now it reads like you wrote it. One or two small edits.

By run fifty:

You read it once and send it. You don't remember the last time you changed something. It's yours. The prompt didn't get smarter on its own. You made it better. That's the difference between a prompt library and a folder of notes. Notes stay still. Prompt assets compound every time you show up.

What the Library Looks Like When It's Done

Seven files. Each one named for what it does. Each one with all four parts filled in.

The paid section below has the actual starter templates. Copy them, fill in the brackets, save them in a folder.

You don't need to write these from scratch. The structure is already here.

What makes them yours is the context load. Which files you point them at. Which accounts. Which deals. That's it.

The templates are the scaffolding. You fill them with your codebase and they become your prompts.

Rep Action This Week

Build two prompts:

• Pre-call prep

• Follow-up

Four parts each:

• Context load

• Instruction

• Output format

• Quality bar

Save them as files in a folder called: prompt-library/

Run them on a real account this week. After each run, note what you'd change. Update the file. That's the entire practice.

Two prompts. One folder. One week. That's the library.

Consultants spend 10+ hours a week in PowerPoint. SlideHub gives your team one searchable library of approved slides. Find slides, build with AI, and update decks without leaving Microsoft 365.

See SlideHub in action

Wispr Flow sits at the system level — no plugins, no extensions, no setup per app. Speak your prompts anywhere you type and get clean, paste-ready text. 4x faster than typing. Start flowing free.

The Starter Templates

Below are fill-in templates for all seven prompt library files.

Same format as the previous templates.

Copy them, fill in the brackets, delete the placeholder text.

One note:

The context load line is the most important part of each template.

Don't skip it or leave it vague.

The more precisely you specify which files to load, the less you'll need to edit the output.

The context load is what makes a prompt asset different from a saved note.

prompt-library/pre-call-prep.md

# PRE-CALL PREP

## CONTEXT LOAD
Load from main-branch/: buyer.md, product.md, objections.md
Load from personal-branch/: accounts/[account-name].md, deals.md (entry for [company] only)

## INSTRUCTION
I have a call with [contact name], [title] at [company], on [date].

Generate:
- 3 things I need to know walking in from account history and deal context
- 2 things I need to confirm on this call
- 1 thing to listen for that would change how I position

## OUTPUT FORMAT
Bullets only. No preamble. Under 150 words total.

## QUALITY BAR
If the output references this specific account's history and current deal stage, it's working.
If it sounds like it could apply to any account, the context load needs to be more specific.

prompt-library/cold-opener.md

# COLD OPENER

## CONTEXT LOAD
Load from main-branch/: buyer.md, product.md
Load from personal-branch/: voice.md

## INSTRUCTION
Write a cold opener to [prospect name], [title] at [company].

Trigger: [what made you reach out, such as job posting, funding, news, referral]

The opener should be 3-5 sentences.

Direct.

No setup.

Lead with their situation, not our product.

## OUTPUT FORMAT
Email only. No subject line. No sign-off. Just the body of the opener.

## QUALITY BAR
If you'd send it without changing the first sentence, the voice is right.
If it starts with "I", rewrite it.

prompt-library/follow-up.md

# FOLLOW-UP

## CONTEXT LOAD
Load from personal-branch/: accounts/[account-name].md, voice.md, deals.md (entry for [company] only)

## INSTRUCTION
Write a follow-up email after [describe the interaction, such as call, email exchange, demo].

Key thing that was said or decided: [what happened]

Specific next step we agreed on: [what's next]

The email should move us toward that next step without being pushy.

## OUTPUT FORMAT
Email body only. Under 100 words. My voice, not the team's.

## QUALITY BAR
If the email references what actually happened and ends with a clear action, it's done.

prompt-library/post-call-notes.md

# POST-CALL NOTES

## CONTEXT LOAD
Load from personal-branch/: accounts/[account-name].md, deals.md (entry for [company] only)
Load from main-branch/: process.md

## INSTRUCTION
I just finished a call with [contact] at [company].

Here's what happened:

[Paste or describe the key moments, what was said, what changed, what you heard]

Generate:
- What changed about this account, as an update for accounts.md
- What changed about this deal, as an update for deals.md
- Specific next move and date

## OUTPUT FORMAT
Three labeled sections. Bullet points. Ready to paste into the files directly.

## QUALITY BAR
If I can paste the output directly into my files without rewriting it, it's working.

prompt-library/objection-response.md

# OBJECTION RESPONSE

## CONTEXT LOAD
Load from main-branch/: objections.md, product.md
Load from personal-branch/: accounts/[account-name].md, deals.md (entry for [company] only), voice.md

## INSTRUCTION
I'm facing this objection from [contact] at [company]:

[paste exact objection or close paraphrase]

Context:

We are at [stage] in the deal.

[One sentence on their specific situation.]

Write a response that addresses the real concern, uses our documented handle from objections.md, and fits where this deal actually is.

## OUTPUT FORMAT
Written response only. 3-5 sentences. My voice. No "great question" or "I understand where you're coming from."

## QUALITY BAR
If the response acknowledges their specific situation and doesn't feel like a canned handle, it's working.

prompt-library/deal-review.md

# DEAL REVIEW

## CONTEXT LOAD
Load from personal-branch/: deals.md (all active entries)
Load from main-branch/: process.md

## INSTRUCTION
Run a deal health check across my active pipeline.

For each deal, tell me:
- Where it actually stands, not the CRM stage
- What's the real risk to close
- What needs to move this week specifically

## OUTPUT FORMAT
One section per deal. Deal name as header. Three bullet points each. Under 200 words total.

## QUALITY BAR
If it tells me something I didn't already know or hadn't said out loud, it's earning its place.

prompt-library/pipeline-digest.md

# PIPELINE DIGEST

## CONTEXT LOAD
Load from personal-branch/: deals.md (all entries), territory.md

## INSTRUCTION
Write a weekly pipeline summary I could send to my manager.

Cover:
- What closed or moved this week
- What's at risk
- What's on track to close next week
- One market observation from my territory that's relevant right now

## OUTPUT FORMAT
4 short sections with headers. Professional but in my voice. Under 250 words total.

## QUALITY BAR
If it reads like something I'd send without editing the opener, done.

The Prompt Anatomy Template

Here's the blank template for any prompt you build on your own.

Fill in all four parts every time. If you skip one, the prompt won't hold up over repeated runs.

# [PROMPT NAME]

## CONTEXT LOAD
Load from main-branch/: [which files]
Load from personal-branch/: [which files, which specific entries if needed]

## INSTRUCTION
[What you want done. Be specific. Name the output. Name the format. Name the scope.]

## OUTPUT FORMAT
[How the response should be structured. Bullets, sections, length, what to omit.]

## QUALITY BAR
[One sentence: what does "good output" look like for this prompt?
If output passes this test, the prompt is working. If not, update the instruction.]

The Improvement Protocol

After every prompt run:

• Read the output

• Count the edits you made

• If 0-2 edits: the prompt is working, no change needed

• If 3-4 edits: note what was off, then update one thing in the instruction or output format

• If 5+ edits: the context load is probably wrong. Check which files you're loading and whether they're current
  

That's it. You don't need a system. You need a habit. Run it. Read it. Update it if needed. Close the file.

The prompt gets better every time you do this. Skip it once and nothing breaks. Skip it consistently and the prompts drift back toward generic.

One update per run. Ten runs. You won't recognize the prompt you started with.

Next week marks the end of the BYOC series. I am very excited about S01E023 (two weeks away)

Let's begin

The last two weeks, we built the Main Branch.

You have five files. They tell your AI who your buyer is, what your product does, how you handle objections, what your process looks like, and how your team sounds. That's the foundation. Every rep on your team could load those same files and start from the same place.

That's the point of the Main Branch. It's shared.

This week, we build the part that isn't.

The Main Branch comes from your company. Your manager could hand it to you. Your enablement team could build it. In theory, it should already exist somewhere in your org.

The Personal Branch is different. Nobody can hand you this. It only exists if you build it yourself. And most reps never do.

What Lives in Your Head

Here's something worth stopping to think about.

Everything you know about your accounts right now lives in your head.

The relationship history with your champion at Acme. The reason you haven't looped in the VP yet. The thing the last rep said on a call that poisoned the well before you took over. The fact that your contact is 90 days into a new role and still figuring out her own priorities.

None of that is in Salesforce. None of that is in the company playbook. And right now, when you open your AI and ask it to help you prep for a call or draft a follow-up, it has no idea any of it exists.

The same is true for your voice. You have a way you write when you're at your best. A way you open an email. A way you handle a follow-up that doesn't sound like everyone else on your team. That's yours. The team voice guide captures how the team sounds. It doesn't capture how you sound.

The Personal Branch is just the act of writing all of this down in a place your AI can read.

The Four Files

Your Personal Branch has four files. Each one captures a different kind of knowledge that only you have.

accounts.md is the running record of what you know about each of your accounts that isn't in any system. The history. The internal dynamics. The champion. The landmine. The thing you said on a call six months ago that you need to remember before you walk into the next one. One entry per account. Updated after every meaningful interaction.

voice.md is the file that makes your AI sound like you. Not like the team. You. Pull three to five emails you've written that you'd send again tomorrow without changing a word. The ones where you re-read them and thought "yeah, that's exactly right." Drop them in. Add three to five phrases you never use. That's the file. The AI learns your voice from your actual writing, not from a description of it.

deals.md is your real-time deal board. Where each active deal stands right now. What changed this week. What the specific next move is. What you're working around. What the close looks like if everything goes right. Update it every Friday. If you skip a week, your AI is working off outdated information. Outdated deal context is worse than no deal context.

territory.md is your read on your market. What you've noticed about your specific segment that the ICP doc doesn't cover. What's shifted in the last 30 days. What keeps coming up across your accounts. The things you'd tell a new rep on their first day that they won't find in onboarding.

Building voice.md

voice.md is the most personal file you'll build and the one most people overthink.

Here is the entire process: open the file, write down three words that describe how you sound when you're writing at your best. Then find three to five emails you've actually sent that nail that. Paste them in. Then write down five phrases you would never use — the ones that make you cringe when you see them in a rep's outreach. Done.

The test is simple. Load the file. Ask your AI to write a cold opener to a prospect. Read what it produces. If you'd change more than three words, your examples aren't specific enough yet. Add more samples. Run the test again.

When the AI produces an opener you'd send without editing, the file is working.

That's the whole thing. Most reps skip this file because it feels abstract. It is not abstract. It's the file that takes everything from "sounds like us" to "sounds like me."

Same Prompt. One More Layer.

You've seen the two-rep comparison before. Let's make it three.

Rep A has no context. Opens a new chat, types a prompt. Gets the opener that starts with "I hope this message finds you well." You know this output. You have written it and deleted it a hundred times.

Rep B has the Main Branch loaded. Buyer profile. Product context. Team voice. Gets something useful. Specific enough to be credible, on-brand, better than most outreach. But it still sounds like it came from the team, not from them. If you swapped the signature, you wouldn't be able to tell who wrote it.

Rep C has both layers loaded. Main Branch plus their accounts.md and voice.md. The output knows who the specific prospect is, references the last conversation, opens the way Rep C opens things. Reads the email and you'd know whose account it was.

Same tool. Same prompt. Completely different outputs.

The personal layer is the diff between Rep B and Rep C. And it's the diff most people leave on the table because they only ever build the foundation.

Keep It Alive

One more thing before you go build.

The Personal Branch is not a one-time project. Unlike the Main Branch, which changes slowly, your personal files change constantly. Your deals move. Your accounts shift. You have new conversations. You write emails that are better than anything in your voice file right now.

• accounts.md: update it after every meaningful interaction. Not a full rewrite. One entry. What happened, what it means, what you need to remember.

• deals.md: every Friday. Where each deal stands. What changed. What's next. Block 15 minutes. Do not skip this.

• territory.md: monthly. What's shifted in your market. What you're seeing across accounts that the playbook doesn't account for.

• voice.md: quarterly, or after a big win. Pull your best recent emails. Add them. Let the file get more specific over time.

A stale personal branch isn't neutral. It's actively misleading. If your deals.md says an opportunity is in late-stage when it stalled two months ago, your AI is going to write like the deal is healthy. That makes the output worse than no context at all.

Build it. Keep it current. That's the whole job.

Rep Action this week

Build voice.md.

Not accounts. Not deals. Voice first.

Pull three emails you'd send again tomorrow. Paste them in. Add five phrases you never use. Load the file into your AI and ask it to write a cold opener for a current prospect. Read what comes back.

That test will show you immediately whether the file is working. If the output sounds like you, you're done. If it sounds like everyone else, add more samples and run it again.

One file. One afternoon. This is the part nobody can give you.

2026 State of AEO Report

A year ago, most marketers weren't thinking about AI search. Now it's one of the fastest moving channels in the industry and nobody has a playbook yet.

So we built one. We surveyed hundreds of marketers to find out how they're approaching answer engine optimization, where they're investing, what's actually working, and what isn't.

The result is the 2026 State of AEO Report. Real data. Real strategies. A clear picture of where AI search is headed and how to get ahead of it.

Download the free report

Moda is the viral AI design agent for polished, on-brand slides, docs, ads, and more. Turn prompts into fully editable designs on a real canvas, then export to PowerPoint, PDF, and more.

Try Moda Free Today

Let's begin

Last week we set up the folder. Two branches. Nine empty files staring back at you.

If you followed through on the action, you should have something that looks like this sitting on your desktop right now:

sales-codebase/
  main-branch/
    buyer.md
    product.md
    objections.md
    process.md
    voice.md
  personal-branch/
    accounts.md
    voice.md
    deals.md
    territory.md

Good. Now we fill the first branch in.

But before you type a single word, I need to show you the mistake almost everyone makes when they build a context layer for the first time. Because if you don't see it up front, you'll spend an afternoon filling in five files that look complete, feel like progress, and do almost nothing when you actually use them.

The Context File That Wastes Your Time

The Problem With Context Files

Here's what useless context looks like. And I want you to read this carefully because I guarantee something this generic exists somewhere in your current setup.

"We sell to enterprise sales leaders who are looking to improve team performance and drive revenue growth across their organizations."

That is not a buyer profile. That is a description of every B2B sales tool on the market. Salesforce could have written it. Your top competitor could have written it. A tool you've never heard of could have written it.

When your AI reads that sentence, it has nothing to work with. It defaults to the most generic version of a sales leader it has ever been trained on and produces output that sounds exactly like that. Generic in, generic out. Every time.

Here's what usable context looks like:

"VP of Sales at B2B SaaS, 200-800 employees, Series B or C. Under board pressure on net new pipeline. Their reps aren't having enough first conversations. Not from lack of effort. They have no signal on who to call or when. 12 to 18 months into the role. Already burned one tool purchase that didn't land. They say 'pipeline coverage' on almost every call."

That's a real person. When your AI reads that, it has a human being to write to. The output becomes specific. It sounds like it came from someone who actually knows this buyer. Because now, it does.

The test is simple. Ask yourself: could this file describe your closest competitor? If the answer is yes, it's not ready.

Three Things That Make Context Actually Work

 

Every context file that works has three things. Most people include one.

1. Their exact words.
The phrases your buyers use on calls. Write them down verbatim. "We're drowning in manual work." "Ramp time is brutal." "I can't clone my best rep." When those phrases are in your context file, your AI uses them back. Output stops sounding like a brochure. And there is nothing sweeter than using their own words against them when they see your pricing for the first time.

2. Concrete specifics. Numbers. Titles. Company stages. Funding rounds. "Mid-market" is vague. "200 to 800 employee B2B SaaS, post product-market fit, third or fourth sales hire, no formal process yet" gives the AI something real to anchor to. Specific input produces specific output. Vague input produces brochure copy.

3. Honest negatives. Where you lose. Who you are not for. What your buyers actively tune out when you lead with it. Most people skip this entirely because it feels counterintuitive to write down your weaknesses. But it's what makes AI output stop sounding like a pitch deck. An AI that knows where you lose writes with credibility. An AI that only knows where you win writes like it's trying to close you.

Building buyer.md

Let's stop being theoretical. Open main-branch/buyer.md. Here's what you're putting in it.

Line one: titles and company profile. Not "enterprise sales leaders." VP Sales. Dir. of Sales Enablement. B2B SaaS. 200 to 800 employees. Series B or C. One line. Done.

Then write what actually keeps them up at night. Not "improving team performance." The real thing. Their reps don't have enough pipeline. The board wants an explanation every Monday. They inherited a team and inherited all the dysfunction that came with it. They already spent budget on a tool that the team never adopted. That failure is now part of the conversation every time a new vendor walks in.

Then write their exact words. Go pull your call recordings right now. What phrase shows up in almost every discovery call? Write it down exactly. Don't clean it up. Don't make it grammatically correct. Put it in the file the way they said it.

That's the whole file. Under 400 words. If you're writing more than that, you're not deciding what matters yet. Cut until you can't cut anymore without losing something essential.

One more thing: if you sell to more than one persona, build a separate file for each one. Do not mix them. Mixed context produces mixed output.

The Five Files and What Each One Actually Needs

Here is the complete spec for your Main Branch.

buyer.md — Start here, today. Titles and company profile. Real pain in their words. What they have already tried and why it failed. How they evaluate vendors. What makes them immediately stop paying attention. That last one is the hardest to write. It's also the most valuable.

product.md — Build this second. One sentence that describes what you do without using any jargon. Three specific wins you can actually prove. Where you lose and the real reason why. Two or three proof points that reliably move buyers. And the mistake reps most often make when they position it. If you write that last one honestly, this file will be worth more than your entire deck.

objections.md — This file is never finished. Start with the 8 to 10 objections you hear on every deal. For each one: the exact words they use, what they actually mean underneath it, and how you handle it. Not a scripted response. The intent and the direction. Update this file every month without exception. If nothing has changed in six months, you are not paying close enough attention to your deals.

process.md — Keep it short. Stage names. What has to be true before a deal moves forward. Where things stall and the real reason why. Who is in the room at each stage. This is a reference file, not a playbook. If you hit 400 words, you have not made the hard calls yet. Trim it down.

voice.md — Build this last, but do not skip it. Three words that describe how your team sounds. One real example of a message done right, with a note on what makes it work. One real example done wrong, with a note on why it fails. What phrases you never use. Pull actual examples from your best reps. Do not describe the voice in the abstract. Show it.

Five files. Under 400 words each. Under 2,000 words for your entire Main Branch. One afternoon to build it if you sit down and do it.

Before You Call Any File Done

Every file goes through six checks before it gets loaded into an AI session. If any one of them fails, the file is not done.

1. Could this describe your competitor? If yes, it is too generic. Add specifics until it can only describe you.

2. Does it use their words or your words? Find every piece of marketing language and rewrite it in plain speech.

3. Does it include at least one honest negative? Where you lose. What they don't care about. This is the most skipped check on the list. It is also the most important one.

4. Is it under 400 words? If not, cut it. Length is not quality. Length usually means you haven't decided what matters yet.

5. Would a new rep understand your buyer, your product, or your process after reading it once?

6. Does it include at least three phrases your buyers actually say? Not paraphrased. Verbatim. If not, go back to call recordings before you touch this file again.

All six pass: load it and run your next session.

Any fail: the failure tells you exactly what to fix.

Rep Action this week

Build buyer.md. That is the only file on your list this week.

Not product. Not objections. Buyer.

Get it through all six checks. Then load it into your AI and ask it to write a cold outreach email to the buyer you just described. Read what comes back. If it could have been written by anyone, the file is not specific enough. Keep cutting and sharpening until the output surprises you with how accurate it is.

That moment is the benchmark. Once you have it, the other four files will build themselves.

Ghost: Unlimited Postgres For Agents

Your agent builds faster than a 2-project free tier allows. Ghost gives it unlimited postgres. No credit card. Try free.

Install Ghost MCP

Let's begin

Last week I left you with a homework assignment.

Run the memory export prompt. Read what your AI actually knows about you. Notice the gaps.

If you did it, you probably found something like: your job title from 18 months ago, a project you killed in Q2, and not a single mention of your current territory.

So now you know the problem. Your AI has no permanent brain. Chatting at it every morning isn't fixing that.

This week, we start building the fix.

But before we get into the files, I need to explain the architecture. Because if you don't understand why this system is built the way it is, you'll build it wrong.

The Two Layers

Here's what most people miss when they start building a context layer.

They build one thing. They dump everything into one file or one memory block. Their ICP, their accounts, their objections, their personal story. All in the same place.

That works. Sort of. Until your company changes the messaging. Or you move to a new territory. Or a teammate asks to use your setup. Then everything is tangled and nothing is reusable.

The Sales Codebase is built in two layers. On purpose.

Layer 1: The Main Branch

The Main Branch is everything that's true about your business.

Your buyer profile. Your product context. Your objection library. Your sales process. Your team voice guide.

These are the things that don't change week to week. And they don't belong to you personally. They belong to the team. Five files. Each one under 400 words. Each one with a single job.

Every rep should be pulling from the same Main Branch. If your AI doesn't know who your ICP is, it doesn't matter how good your personal layer is. The foundation has to be there first.

Layer 2: The Personal Branch

The Personal Branch is the layer only you build.

Your accounts. The history, relationships, and deal intel that lives entirely in your head. Your voice. Not the team voice. Your voice. The way you write a follow-up when you're at your best. Your active deals. Where each one stands, what you're working around, what comes next. Your territory. Your specific read on your market and your buyers.

The Personal Branch is what takes AI output from "pretty good" to "ready to send."

The rep who has both layers is operating a completely different kind of AI than the rep who starts cold every morning.

Same tool. Different output.

What That Actually Looks Like

Same prompt. Same AI. Two different reps.

Rep A has the Main Branch loaded. Has their personal accounts and voice loaded on top of that. They ask the AI to write a follow-up after their prospect's team doubled headcount last quarter.

They get:

"Hi Sarah. Noticed your team doubled headcount last quarter. Given the ramp challenges you mentioned on our last call, here's what I'd send to your new AEs this week..."

Rep B opens a new chat. Same prompt. No context.

They get:

"I hope this message finds you well. I wanted to reach out regarding our innovative solution that helps enterprise sales leaders improve team performance and drive revenue growth..."

Same tool. Same prompt. Completely different output.

The gap isn't about which AI you're using. It's about what you bring into the room with you.

Where We're Going

Here's the rest of the roadmap.

Part 2: We build the Main Branch. Five files. I'll show you exactly what goes in each one and what to leave out.

Part 3: We build the Personal Branch. The layer that makes output sound like you, not your company.

Part 4: The Prompt Library. Building prompts as reusable assets, not one-time commands.

Part 5: The Merge. How your individual wins become your team's infrastructure.

By the end, you'll have a working Sales Codebase. An actual folder of files that makes every AI session faster, more specific, and actually useful.

Rep Action this week

Look back at the memory export you ran last week. Read it through a new lens.

Which parts are team knowledge? Your ICP. Your product. Your process. Those belong in the Main Branch.

Which parts are personal? Your accounts. Your deals. The way you sound when you're writing your best follow-up. Those belong in the Personal Branch.

You don't need to build anything yet. Just start sorting. Two piles. That's this week's job.

Next week, we build the first one.

2026 Performance Marketing Media Mix Guide

What can 600+ consumers across four generations teach about attention today?

Audiences are constantly scrolling, skipping, and multitasking across channels.

See how TV fits into the modern path to purchase. Explore this report for insights on capturing attention in 2026.

Download the Report

Attio - the AI CRM for modern businesses.

Attio is the AI CRM that keeps you ten steps ahead.

Ask Attio anything. Where should I focus? What deals are at risk? Search, update, and create across your customer data.

Ask more from CRM. Ask Attio.

Try Attio for free

Let’s begin

I have a quick exercise for you. Let’s use the power of imagination to go on a little journey together.

Let’s imagine you are getting ready to hire an assistant who reports directly to you. You’ve been given a pretty healthy budget for the hire, which is exciting to you because you know you can look at a higher caliber candidate pool.

So you start your search. And you land on an absolute 💎 gem of a candidate who has near perfect background. Graduated from a top 40 university. Spent the past 8 years working as a direct competitor to you. They know your ICP. They know your product (having sold against it for the past 8 years). And they also know their old product… which is massive because you know you can use some of their information to update your battle cards.

This is the definition of an “ideal hire.”

The Onboarding Gap

You spend their first week onboarding them. You hand over the "unwritten rules" of your territory. You show them the specific way you structure a follow-up email. You explain why you never mention a certain competitor in Phase 1, but always bring them up in Phase 3.

By the end of the month, this assistant feels like an extension of your brain.

Now, imagine if every single morning, that assistant walked into the office with total amnesia.

They still have the Ivy League degree. They still have the 8 years of industry experience. But they have absolutely no memory of you, your company, or the 6 months of training you just gave them. Every morning, you have to spend the three hours re-explaining who you are and how you work just to get them up to speed.

You would fire them by Tuesday.

Yet, this is exactly how most reps treat their AI.

They expect "Senior VP level" output from a tool they treat like a temp worker on their first day, every single day.

And before you’re all like “oh, but Jay… I use AI all the time, and my AI has a memory of me so the context is there”… I have another exercise for you. Go to your AI and give it this prompt. This prompt is designed to extract every relevant memory from the AI, and provide an output that you can copy/paste into a new AI to “onboard” it.

Export all of my stored memories and any context you've learned about me from past conversations. Preserve my words verbatim where possible, especially for instructions and preferences.

## Categories (output in this order):

1. **Instructions**: Rules I've explicitly asked you to follow going forward — tone, format, style, "always do X", "never do Y", and corrections to your behavior. Only include rules from stored memories, not from conversations.

2. **Identity**: Name, age, location, education, family, relationships, languages, and personal interests.

3. **Career**: Current and past roles, companies, and general skill areas.

4. **Projects**: Projects I meaningfully built or committed to. Ideally ONE entry per project. Include what it does, current status, and any key decisions. Use the project name or a short descriptor as the first words of the entry.

5. **Preferences**: Opinions, tastes, and working-style preferences that apply broadly.

## Format:

Use section headers for each category. Within each category, list one entry per line, sorted by oldest date first. Format each line as:

[YYYY-MM-DD] - Entry content here.

If no date is known, use [unknown] instead.

## Output:
- Wrap the entire export in a single code block for easy copying.
- After the code block, state whether this is the complete set or if more remain.

I’m willing to bet that the information the AI gives you is right… but only to a point. AI does a terrible job of updating it’s own memory. So for example, it might not remember about your recent promotion, or that your kid isn’t 7 for the 3rd year in a row now… And that’s because you have to explicitly tell an AI when something is worth remembering permanently.

The Missing Manual

When you hire that "gem" of a candidate, you don't just hope they figure it out. You give them a Playbook. In the software world, that’s the Codebase. Don’t let the word “codebase” scare you. A codebase is just a Repo. And a repo is just a Folder. And by folder… I mean literally just any folder that you’ve designated on your computer. But when you designate this one… codebase <ahem> as your own personal “source of truth” this becomes your Context Layer.

If you want the AI to act like that high-caliber hire, you have to stop "chatting" and start "onboarding." You need to give it a permanent set of files that tell it:

• who it is and what it’s responsibilities are.

  • but maybe even more most importantly… what skills and tools you provide it with.

    • and in case you need a refresher…

      • a skill gives the AI the blueprint/framework for a plan.

      • a tool gives the AI the ability to execute the plan.

• who you are, and what matters most to you.

• what you sound like. what you say. what you would never say.

• what you do. what you sell. the product or service you deliver. your value proposition.

• your icp. what your ideal buyer actually looks like.

• where your product or service wins… and where it loses.

• who your key competitors are. how you differentiate against them.

• how you handle objections. how you sound when you’re doing your best work.

Without this, you don't have an AI assistant. You just have a very smart stranger who keeps forgetting about the promotion you received… last year.

Portability is the Point

The other, massive problem with relying on the AI’s "built-in" memory is that you’re a tenant on their land. If ChatGPT has a bad update, or your company decides to switch everyone over to Claude or Gemini tomorrow, your assistant’s brain stays behind.

A Sales Codebase is portable. It’s a set of simple markdown or text files. When a better AI comes out, you don't rebuild. You just pick up your codebase and move.

That’s what this series is about. We are building your permanent context layer.

The Sales Codebase Roadmap

Over the next several weeks, we are going to build this out together:

• Part 2: The Main Branch (The shared team brain: ICP, Product, Battlecards)

• Part 3: The Personal Branch (Your layer: Your accounts, your voice, your specific deals)

• Part 4: The Prompt Library (Prompts as assets, not one-time use)

• Part 5: The Merge (How individual wins become team infrastructure)

By the end, you’ll have a working Sales Codebase that makes every AI interaction faster, more specific, and—most importantly—actually useful.

Rep Action this week

Run that Memory Export Prompt I gave you above. Read what it spits out. Notice the gaps. Notice the outdated info. That mess is exactly why your AI output feels "almost right" but requires 10 minutes of editing.

That gap is what we’re going to fix.

The AI Playbook for Video Teams That Can't Slow Down

Wistia's new AI Video Marketing Trends report shows how marketers are using AI to move faster, improve quality, and extend the life of every video. See how leading teams are driving results without adding more work.

Download the Full Report

Millions of developers use Wispr Flow to dictate code, write docs, and give coding agents better context. 89% of messages sent with zero edits. 4x faster than typing. Free on Mac, Windows, and iPhone. Try Wispr Flow free.

A word from the author

Hi everyone. Thank you for your continued support!

This week’s issue is going to be a little more in depth than usual.

Yea, you read that right… this week you will get the full tool set, ready to plug and play with your own AI… or maybe even your own personal operating system…

And now…

A word from our advertisers: (click the ad to show your ❤️ love)

How Jennifer Aniston’s LolaVie brand grew sales 40% with CTV ads

The DTC beauty category is crowded. To break through, Jennifer Aniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.

Learn more

Stop Working Your Pipeline Top-Down. Start Working It by Signal.

Open your CRM right now.

Your pipeline is sorted by close date, or deal size, or whatever your manager told you to sort by last quarter.

None of those sorts tell you what to work today.

Signals do.

And most reps still aren't using them this way.

The problem with the default pipeline view

Every rep has the same problem on Monday morning.

Too many deals. Not enough hours.

You open the pipeline. 20, 30, 50 deals staring back at you. Every one of them looks "in progress." Every one wants your attention.

So you work it top-down. Biggest deal first. Or closest to close first. Or whichever name catches your eye.

That sort is easy. It's also wrong.

Deal size doesn't tell you which deal is moving. Close date doesn't tell you which buyer is paying attention. And your gut doesn't tell you anything you didn't already decide before you opened the tab.

The real question isn't which deal is biggest. It's which deal is alive this week.

Buying signals answer that question. Default sorts don't.

What a signal actually is

A signal is a change in buyer behavior or account context that tells you the deal state has shifted.

Not "the buyer said yes." That's a decision, not a signal. Signals are quieter.

Four categories worth filtering your pipeline on.

1. Engagement signals. The buyer is interacting with your stuff. Proposal views. Pricing page visits. Email opens after a week of silence. A forwarded deck to an address you didn't send it to. Anything that says "they're thinking about this right now."

2. Organizational signals. The buying committee changed. A new VP started at the account. Your champion got promoted. The CFO who blocked you in Q4 left. Any of those changes your path to close.

3. External signals. Something happened to the company that changes their urgency. A funding round. A missed earnings call. A product launch. A layoff. An acquisition. Public events create private pressure.

4. Silence signals. The ones most reps ignore. A deal that was active last week went quiet. A champion who usually replies same-day has now taken four days. These aren't red flags yet. They're the quieter kind. They tell you the deal is drifting and you have about a week to pull it back.

One more rule. A signal only matters if it's fresh. A funding round from eight months ago is not a signal. It's trivia. Anything older than 14 days stops telling you what to do today.

The shift: score deals by signal strength, not pipeline position

Here's the move.

Every Monday, run your pipeline through a signal filter. Not a qualification check. Not a deal review. A signal check.

For each deal, ask four questions.

1. What signal has fired on this deal in the last 14 days?

2. How strong is it?

3. Does it suggest urgency is going up, down, or flat?

4. What specific next move does that signal unlock?

You're not scoring the deal. You're scoring the moment.

A $500k deal with no signals this month is not your top priority. A $40k deal where a new CRO just started, your champion viewed the proposal three times, and the competitor just announced a bad quarter...that's your top priority. Even though it's smaller. Even though it closes later.

Because the signals say it's moving. The big one isn't.

The Pipeline Triage prompt

Paste this into Claude or ChatGPT every Monday, alongside a dump of your active deals and recent context for each.

You are a pipeline triage analyst. I'm going to give you a list of my active
deals and recent context for each. For every deal, evaluate whether a buying
signal has fired in the last 14 days.

Score each deal on:

1. SIGNAL PRESENT (Yes/No): Is there a specific, dated event from the last
   14 days that indicates a shift in buyer behavior or account context?

2. SIGNAL STRENGTH (1-5):
   1 = no signal
   2 = minor engagement (email open, page view)
   3 = clear engagement (proposal view, re-engagement after silence)
   4 = organizational change (new buyer, champion promoted, committee shift)
   5 = external pressure (funding, earnings, layoff, acquisition)

3. DIRECTION: Is urgency rising, falling, or flat?

4. NEXT MOVE: The single most valuable action I should take this week on
   this deal, tied to the signal.

Rank the deals from highest-priority to lowest based on signal strength and
direction. Flag any deal with no signal in the last 14 days as "cold this
week."

Be direct. If a deal has nothing happening, say so.

Here are my deals and context:

[PASTE]

What this looks like in practice

Sample triage output on five deals.

Ranked pipeline for this week

1. Apex Staffing ($40k)
   Signal: 4/5. New CRO started last Thursday. Champion viewed the proposal twice on Friday.
   Direction: Rising. New exec wants a win in their first 90 days.
   Next move: Short intro note to the new CRO. Reference the existing champion. Offer a 20-minute overview.

2. Nova Labs ($180k)
   Signal: 4/5. Company announced Series B last week.
   Direction: Rising. New capital means expanded initiatives.
   Next move: Send the CFO a one-page ROI model tied to their stated fundraising narrative.

3. Orbit Retail ($60k)
   Signal: 3/5. Procurement pulled the draft MSA into their system on Monday.
   Direction: Rising. This is the quietest but most reliable sign of real buying.
   Next move: Send redlines proactively. Ask for a 15-minute call with procurement.

4. Ridge Systems ($90k)
   Signal: 2/5. Champion opened the last three emails but hasn't replied.
   Direction: Flat, leaning down.
   Next move: Stop emailing. Send a specific yes/no question by LinkedIn DM. Break the pattern.

5. Delta Freight ($220k)
   Signal: 0/5. No activity in 18 days.
   Direction: Cold this week.
   Next move: Don't chase. Queue one breakup email for Thursday. Move on.

Look at what just happened.

Your biggest deal (Delta Freight, $220k) ranked fifth. Your smallest (Apex at $40k) ranked first because a new exec walked in the door. Orbit at $60k will probably close first because procurement is moving.

Size didn't matter. Signal did.

How to use this starting today

Don't overcomplicate it.

Step 1: Pick your active deals. Not all 50. The 10 to 15 in open motion.

Step 2: For each one, write three lines. Last touch. Last known buyer activity. Any external news you know about the account. If you don't know, check their LinkedIn company page and news tab for 60 seconds.

Step 3: Paste the whole thing into the Pipeline Triage prompt.

Step 4: Work the list top-down. But now top means "most live," not "biggest."

Step 5: Rerun it next Monday. Watch which deals hold the top of the list and which ones keep falling off. The ones that keep falling off aren't real pipeline. They're hope.

Why this works

Three reasons.

It replaces vibes with events. "This one feels hot" is not a plan. A dated signal is. You stop working the deals you like and start working the ones that are actually moving.

It kills the sunk-cost trap. Big deals that aren't moving eat your week because you've already invested in them. The triage doesn't care how much work you've put in. If nothing fired this week, it's not this week's problem.

It compounds your coverage. Run this weekly for a quarter and you'll spot patterns. Which industries generate signals you can actually act on. Which of your own deals you've been carrying for no reason. The triage doesn't just rank deals. It teaches you your pipeline.

The paid version goes deeper

What I shared today is the core triage. One prompt, four signal categories, a ranked list.

The full Pipeline Triage kit includes:

A pre-call prompt that takes your top-ranked deal and generates the first two minutes of your next conversation, tied to the specific signal that moved it up the list.

A weekly digest generator that summarizes your pipeline in five lines you can paste into a manager 1:1. No more "the deal feels good." You walk in with signals, direction, and moves.

A cold-deal recovery prompt that analyzes any deal flagged "cold this week" and tells you whether to keep it, kill it, or re-enter through a different contact.

Every rep has five deals this week they should be working on.

You probably can't name all five right now.

Run the triage this Monday. See which deals come up. Work those first.

Ignore the rest until they give you a reason not to.

Talk next week.

The AI Playbook for Video Teams That Can't Slow Down

Wistia's new AI Video Marketing Trends report shows how marketers are using AI to move faster, improve quality, and extend the life of every video. See how leading teams are driving results without adding more work.

Download the Full Report

Pipeline Triage
Quickstart Guide | sellingwithai.vip

What this is

The Pipeline Triage scores your active deals every Monday based on what's actually happened in the last 14 days. It tells you which five deals to work this week and which to leave alone.

Three extensions sit on top of the core triage:

• A pre-call opener that writes the first two minutes of your next conversation

• A weekly digest that compresses your pipeline into five lines for your manager 1:1

• A cold-deal recovery prompt that decides whether to keep, kill, or re-enter

You get access on two platforms. Use whichever you prefer.

ChatGPT Custom GPT: SELLING WITH AI PIPELINE TRIAGE GPT

Claude Project: Claude projects can't be shared the way GPTs can. If you want the Claude version, email me at jay@sellingwithai.vip and I'll send the copy/paste blueprint.

How to use it (10 minutes, every Monday)

1. Pull a list of 10 to 15 active deals. Not all 50. The ones in open motion.

2. Add three lines of context per deal. Last touch. Last buyer activity. One piece of external news about the account. If you don't have external news, check their LinkedIn company page and news tab for 60 seconds.

3. Paste into the Triage. You'll get a ranked list, a signal strength score, a direction, and a specific next move for each deal.

4. Work top-down. Top now means "most live," not "biggest."

5. Rerun next Monday. Watch which deals hold and which ones fall off.

The four signal categories

Reading your ranks

Commands you can use

• Triage my pipeline — Returns the ranked list

• Open call on [deal] — Writes the first two minutes of your next conversation

• Write my pipeline digest — Five-line summary for manager 1:1

• Recover [deal] — Keep, kill, or re-enter decision with a drafted message

• Re-rank (after pasting new activity) — Compares to last week and shows trajectory

Tip: The more context you paste, the sharper the signals. "Champion is engaged" is less useful than "champion opened the pricing proposal Tuesday and forwarded it to a name I haven't met."

Backup: the core triage prompt

If you want to run the triage without the GPT or Claude Project, paste this into any AI chat followed by your deal list.

You are a pipeline triage analyst. For every deal I give you, evaluate
whether a buying signal has fired in the last 14 days.

Score each deal on:

1. SIGNAL PRESENT (Yes/No): Specific, dated event from the last 14 days.

2. SIGNAL STRENGTH (1-5):
   1 = no signal
   2 = minor engagement
   3 = clear engagement / re-engagement after silence
   4 = organizational change
   5 = external pressure (funding, earnings, layoff, acquisition)

3. DIRECTION: rising, falling, or flat.

4. NEXT MOVE: The single most valuable action this week, tied to the signal.

Rank deals from highest priority to lowest. Flag anything with no signal
in the last 14 days as "cold this week." Be direct.

Here are my deals and context:

[PASTE]

Extension 1: Pre-call opener

You are a sales conversation coach. I'm about to get on a call with a deal
that ranked in my top three this week. Here is the deal context and the
signal that moved it up the list.

Write:
- The exact 20 to 30 second opening frame, tied to the signal
- The first question I should ask after the opener
- What I'm listening for in the buyer's response, with two possible paths
  based on what they say
- One trap to avoid (a tempting move that would burn the signal)

Keep it direct. No fluff. No pitch language.

Signal and context:
[PASTE]

Extension 2: Weekly digest for manager 1:1

You are preparing me for a pipeline review with my manager. I'm going to
paste my ranked pipeline triage from this Monday.

Summarize it in exactly 5 lines:

1. Active deal count and cold-this-week count
2. Top priority deal this week and the single reason it's top
3. Biggest risk in the pipeline and what I'm doing about it
4. Any committee shifts or external events my manager should know about
5. What I'm asking my manager for this week (intro, approval, coaching, nothing)

Must fit in a Slack message. No hedging. No "the deal feels good" language.
If there's nothing to ask for, say so.

Triage output:
[PASTE]

Extension 3: Cold-deal recovery

You are a deal recovery analyst. I'm going to give you a deal flagged
"cold this week" by my pipeline triage, plus its full recent history.

Decide one of three outcomes. Be direct.

KEEP: The deal is still real but quiet. Specify the exact trigger or
signal that would pull it back into active motion. Give me the one-line
watch I should set.

KILL: The deal is done. Write the breakup email. Two sentences. No guilt.

RE-ENTER: The primary path is dead but the account still matters. Name
the alternate stakeholder to approach, why they're a better angle now,
and draft the opening outreach message.

Pick one. Commit to it.

Deal history:
[PASTE]

Getting the most out of this

Run the triage every Monday at the same time. Consistency beats perfection. Ten minutes with the triage beats an hour of staring at the pipeline.

Track your cold-this-week count. After a month, you'll see the truth about your pipeline size. If you have 40 deals and 25 are cold every week, you don't have 40 deals. You have 15.

Use the weekly digest before your 1:1. Paste the five lines into Slack or your notes app. Walk into the meeting with signals instead of adjectives. Watch how the conversation with your manager changes.

Run cold-deal recovery on anything that's been cold three weeks in a row. Three strikes is not a stall. It's a decision. Make it before your manager makes it for you.

Pipeline Triage by Selling with AI | sellingwithai.vip

Questions? Reply to any newsletter issue or email jay@sellingwithai.vip

I am proud to be sponsored by Roku.

A click on the advertisement below goes a long way.

How Jennifer Aniston’s LolaVie brand grew sales 40% with CTV ads

The DTC beauty category is crowded. To break through, Jennifer Aniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.

Learn more

Hey there! Thanks for being here.

In part 1 of this 3 part mini-series, we set up Claude.

In part 2 (last week), we talked about how to actually live in it. Projects. Context. Using Claude as a thinking partner across your whole day, not just for one task at a time.

This begins part 3, the last issue in the series.

And it's the one that makes everything else compound.

The problem with prompting

Here's what most reps do once they find a prompt that works.

They run it. It works. They close the tab. Next time they need it, they're back to starting from scratch. Typing the same setup from memory. Getting slightly worse output. Wondering why AI feels inconsistent.

Prompts are not single-use. They are not disposable. They're assets.

When you find a prompt that reliably produces output you can use, that prompt is worth saving. Running again. Refining. Sharing with your team.

The best salespeople I know who use AI aren't better at prompting. They're better at keeping the prompts that work.

That's what a skills library is.

What a skills library actually is

A skills library is a collection of your best prompts, organized by task, that you can grab and run without thinking.

Not a folder of half-finished ideas. Not a dump of things you've tried. A short, curated set of prompts you actually use, for the tasks you do most often.

Some people store them in a Claude project as reference content. Some people keep it in Notion or Obsidian. Some in a pinned note. The format doesn't matter. What matters is that it exists, and you actually go back to it.

Here's what mine includes:

Discovery prep brief. Paste what I know about the company and contact. Returns a call brief and five questions worth asking.

Deal gap analysis. Paste my notes from any interaction. Returns a MEDDPICC gap analysis with specific next actions.

Prospecting outreach. Paste an account name and what I know. Returns three versions: LinkedIn message, DM, cold email.

Follow-up drafts. Paste the call notes and what the email needs to do. Returns a draft I edit, not write.

Stuck deal consult. Describe the situation. Ask Claude what's missing and what it would do next.

Five prompts. They cover 80% of my repetitive sales work.

Each one took about an hour to get right the first time. Now they run in seconds and come back usable.

That's why prompts compound.

How to build yours

You already know which tasks you do on repeat. The ones where you're always starting from scratch, always re-explaining the context, always feeling like the output is almost but not quite right.

Pick one. Just one (for now).

Write a prompt that's specific enough to get useful output. Not "write me a follow-up email." Something like this:

You're helping a B2B sales rep write a follow-up email after a discovery call. Here's what happened on the call: [notes]. Here's where we are in the process: [stage]. Here's what the buyer said they would do next: [their commitment]. Write an email that advances the decision without being pushy. Short sentences. No corporate language. End with a specific proposed next step.

Run it. See what comes back. If it's close, tweak the prompt, not the output. Keep tweaking until the first draft is 90% of the way there.

Save that prompt. Name it something obvious. That's your first skill.

Then build the next one.

In a month you'll have a personal prompt library that's worth more to you than any AI productivity course.

What running your sales process from inside Claude looks like

This is where the three issues of this series come together.

You have Claude set up. Projects loaded with your context. A skills library of prompts you actually use.

A morning looks like this.

You've got three calls today. Before each one, you open your project, paste what you know about the account, run your discovery prep brief. Three minutes. Three calls prepped.

After each call, you paste your notes and run your deal gap analysis. You see exactly what's missing. You know what to do next before you close your laptop.

Between calls, you're working a new outbound list. You run your prospecting outreach prompt for each account. You're not writing from scratch. You're editing from a strong start.

End of day, one deal has been sitting too quiet. You run your stuck deal consult. Claude identifies the gap you've been avoiding. You send the email before you shut down.

That's a full sales day, run almost entirely from inside one tool, with almost no starting-from-scratch.

The skill you're building isn't prompting. It's context-loading. The rep who can give Claude the right context, fast, is the rep who gets the most useful output.

The next level: building with Claude Code

Everything above runs inside claude.ai. No technical setup required. It works for anyone.

But there's a level beyond this that's worth knowing about.

It's called Claude Code.

Claude Code is not just a chat interface, its so much more. It's a tool that sits on your computer and lets you build things. You describe what you want in plain English. It writes the code, runs it, fixes the errors, and hands you a working tool.

Even though Claude Code is way more powerful, and can actually start building exactly what you are asking for… it still has a chat interface. Meaning, you don't need to know how to code. That's the point.

How to go from using prompts, to using a system

What salespeople are building with it:

A deal scoring tool that lives in the browser. Paste your call notes, it scores the deal on buyer confidence, internal alignment, process clarity, and decision ownership. Not a chat. An actual app you open and use.

A prospect research tool. Type in a company name and a LinkedIn URL. It generates a one-page prep brief. Faster than doing it manually. More complete than what you'd pull together under time pressure.

A follow-up email generator wired to your voice. Trained on emails you've actually sent. The output doesn't need editing, it needs a read.

The difference between these and the prompts in your skills library: these are tools you own. They don't live inside Claude. They run in your browser, on your computer. Nobody else has them. They compound in a different way because you can customize them indefinitely.

This is the part of the series I said was for developers. Let me correct that.

Claude Code is for anyone willing to describe what they want to someone who can build it in an afternoon.

If you've gotten value out of this series, if you've built your project, loaded your context, and started your skills library, Claude Code is the logical next step.

It's not required. Everything in this series works without it. But if you want to go deeper, that's where to start.

Where you go from here

This is the end of the three-part series.

Part one: what Claude is, why it's better for sales, how to get set up.

Part two: how to live in it. Projects. Context. Collaboration.

Part three (this one): skills library, running your process from inside Claude, and where to go if you want to build.

Three issues. Everything you need to start using Claude seriously for sales.

If you've built your project but haven't built your first skill yet, do that this week. Pick the task you do most often. Write a prompt for it. Run it five times. Refine it until you'd use it every day.

That's the move.

Talk next week.

The Skills Library Starter Pack

Copy/paste ready prompts for paid subscribers | sellingwithai.vip

Build Webinars That Keep Working After You Stop

Webinars drive major results when they're built to perform. The Wistia Webinar Guidebook breaks down how to plan, promote, and run webinars that actually convert. Get more sign-ups, increase engagement, and turn every session into a consistent source of pipeline.

Get the Free Guidebook

How Jennifer Aniston’s LolaVie brand grew sales 40% with CTV ads

The DTC beauty category is crowded. To break through, Jennifer Aniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.

Learn more

Housekeeping

Before I get into this week’s issue, I wanted to take a moment to thank you for being a reader of Selling with AI! I appreciate you going on this journey with me.

I also wanted to invite you to revisit sellingwithai.vip - as I have given the site a pretty drastic upgrade in appearance. Let me know what you think by clicking reply on this email.

Again, I am so very thankful for each and every one of you. Keep the content recommendations coming.

I have had a ton of requests come in asking me to cover specific tools… I even had one person asking me to help them set up their own Jarvis. I am working with this reader now and he is very excited to get started.

I read and reply to every single email that comes in from my subscribers.

Claude as a Co-worker: How to Collaborate with AI All Day

Most people open an AI chat… ask it something… then close the tab.

That's not collaboration. That’s a fancier Google search.

I used to do the same thing. Paste a prompt. Get output. Move on. Wonder why AI wasn't changing how I worked.

The shift happened when I stopped treating Claude like a tool and started treating it like a co-worker.

A co-worker who's already read everything about your accounts., who remembers your methodologies, who never needs context repeated, who's available at 6am before your first call and at 10pm when you're prepping for tomorrow.

That's what Claude can be. But only if you set it up that way.

The chat window problem

Here's what most reps do.

They open a new chat. Type something vague. Get generic output. Close the tab.

Next day they open a new chat. Explain who they are and what they sell. Again. Get more generic output. Close the tab.

Every session starts from zero.

That's the problem. Not Claude. Not the prompts. The setup.

A co-worker doesn't need you to re-introduce yourself every morning. They already know who you are, what you're working on, and how you like to communicate.

Claude can work the same way. Most reps never get there because they skip the setup.

What changes when you set it up properly

Last week I covered Projects. This is where it pays off.

When you load your Project with real context, your entire relationship with Claude changes.

I have mine loaded with my ICP, the problems I solve, my MEDDPICC framework, my writing style, and notes on my active accounts. Every chat I open inside that project starts with Claude already knowing all of it.

I don't explain what I sell. I don't re-explain what MEDDPICC is. I don't paste my tone guidelines every time I need an email.

I just ask. It already knows.

That's the difference between a chat window and a co-worker.

How I actually use Claude throughout the day

This is what it looks like in practice.

Before the first call. I open a chat inside my Sales Project. I paste in what I know about the company: recent news, hiring signals, what the contact said when they booked the meeting. I ask Claude to build a prep brief and give me the questions most likely to surface budget authority and urgency. Takes three minutes. I walk into the call knowing exactly what I'm going for.

During prospecting blocks. I don't write cold emails from scratch. I paste a quick summary of the account, tell Claude who I'm reaching out to, and ask for three versions of a first message: a LinkedIn connection request, a DM after they accept, and a cold email. It already knows my voice from the project context. The output is close enough to send after one read.

After calls. I paste my notes in and ask Claude to identify the MEDDPICC gaps. What did I learn? What's still missing? What should I focus on in the next conversation? It doesn't just summarize. It tells me where the deal is weak and what I need to find out.

When I'm stuck on a deal. I describe the situation. What's happened, who's involved, where it's stalled. I ask Claude what I'm missing and what it would do next. Not because I'm going to follow its advice blindly. Because talking through a problem out loud, even with AI, forces you to see it differently.

When I'm writing anything. Follow-up emails. LinkedIn posts. Call agendas. I don't start with a blank page. I start with Claude. Give it the context, tell it what the output needs to do, let it draft. Then I make it mine.

That's a full day of collaboration. Not one prompt. Not one use case. Claude woven into the work as it's happening.

The mindset shift that makes it click

There's a phrase I keep coming back to.

You wouldn't hire a brilliant assistant and only ask them one question per week.

That's what most people do with Claude.

The reps getting real leverage out of AI aren't using it for one task. They're using it as a thinking partner across everything. Research. Writing. Strategy. Deal review. Prep. Follow-up.

The more you bring it into your workflow, the more valuable it gets. Because it builds context. It learns how you think. It stops feeling like a tool and starts feeling like the smartest person in your corner.

That's what the subtitle of this issue means.

Claude is not a chat window.

It's a thinking partner, a research assistant, and a second brain. But only if you set it up properly.

Last issue you set it up. This issue is about how to actually live in it.

What's next

Last issue of the this 3-part Claude mini-series next week.

We're going deep on advanced workflows: building a skills library, automating your most repetitive tasks, and running your entire sales process from inside Claude without starting from scratch every time.

The paid section below has three copy/paste prompts that cover the core collaboration workflows: account research, prospecting outreach, and discovery prep. Each one is built around the four-part structure from this issue. Fill in the brackets, run it, see what comes back.

Members Section

Your ads ran overnight. Nobody was watching. Except Viktor.

One brand built 30+ landing pages through Viktor without a single developer.

Each page mapped to a specific ad group. All deployed within hours. Viktor wrote the code and shipped every one from a Slack message.

That same team has Viktor monitoring ad accounts across the portfolio and posting performance briefs before the day starts. One colleague. Always on. Across every account.

Start free. $100 in credits →