_{⏩ Forward this email to: a marketer who’s being told to do more with AI}

👋 Hi, I’m Amanda. I’m a fractional head of growth with over a decade of experience working with rapidly-growing B2C startups. I work with lean teams in a few ways:

✅ Fractional: You have an expertise gap on your team you need to fill with someone who can flex into both strategy and execution, for an open-ended time frame.

✅ Advisory: You’re either just starting out or need to validate your existing approach, and need someone with growth expertise to guide you.

If you’d like to explore working together, start here to tell me a bit about what you’re looking for. If you’re not quite ready yet and have growth questions you want to chat about, you can book office hours here.

Is the AI directive kind of a trap?

We’re all feeling the pressure of needing to “use AI.” Maybe it’s been a clear directive at your org, or maybe it’s that unrelenting feeling that everyone else is doing it, way better than you are.

Being told to “do more with AI” today is about as useful as being told to “do more with the internet” in the 90s. You can do incredible work with it, or you can waste equally incredible amounts of time with it, or you can stare at a blank screen with no idea where to start, scratching your head as to how everyone around you seems to be spinning gold with this thing.

To be clear - this is not an anti-AI post. I use it daily, and it’s made parts of my workflows drastically more efficient and some of my thinking meaningfully clearer and sharper. But there’s a trap I keep seeing really smart people fall into that could have some pretty serious implications for marketers in particular.

First of all, we’re not saving any time

The promise of efficiency may be true in pockets, but on the whole, is wildly exaggerated, as HBR found. A workflow that used to take four hours can now be done in 20 minutes, or a research synthesis maybe takes one well-versed person now instead of a team (the reality, even if we see the domino effect there).

But what most AI productivity evangelists aren’t talking about is the dynamic where as soon as you get one thing on your list running more smoothly, there are fifteen more things that could also benefit from the same treatment. The upfront work of building, connecting, prompting, testing, and tweaking these systems is pretty substantial, and the work of automating and improving more and more feels never ending.

If history is any guide, new technologies tend to free us up to…do more work, not give us lots of free time back. Productivity gains are real, but typically translate to fewer people needed, not free time granted to employees.

The marketing problem specifically

Here’s where I’m starting to see this get dicey for marketers.

When you talk to AI, there are no consequences for being sloppy. “Vibe”-whatever-ing, talking out your thoughts, iterating, and tweaking as you go are encouraged, in fact. You can forget something mid conversation and ask to be reminded. You can get your ask wrong four times and work through iterations until the output is what you wanted. You can give bits and pieces of all the context you should give, and still get something workable back. These tools are endlessly patient, and don’t acknowledge or care if you’re wasting their time.

Human communication obviously doesn’t work this way. Your stakeholders notice when you haven’t thought through your ask, your boss notices when you show up unprepared.

But most critically for marketers, your customers will notice - or, fatally, stop caring what you have to say - when it feels like your messaging was produced by someone who doesn’t know them.

If you’re spending hours a day in a frictionless communication environment with AI, your tolerance for the friction that real communication provides will quietly erode. Prompts and loose thoughts that eventually produce output implicitly train you that vague or half-baked is fine, but that approach doesn’t work when you’re marketing to humans.

Marketers know that messaging is the output of a deep understanding of who we’re talking to - what they care about, how they think, and what makes them trust us enough to act. You can’t really iterate your way to that understanding, because you’ll have lost that trust before you even have a chance to earn it. Marketers need to make sure the behaviors we’re learning by doing more with AI (which has significant benefits, to be clear) never extend into our relationship with the people we’re marketing to.

We all know AI is excellent

I’m not suggesting you use AI less. It’s an incredible tool with infinite use cases (which is part of why it can seem so overwhelming at first). Your job is to be thoughtful about what you bring to the start of an AI conversation or task, and where AI’s judgment ends and where yours begins.

The best applications of AI for marketers are the ones downstream of human judgment. It’s excellent at analysis and synthesis of information. Give it a large, messy data set, or a bunch of qualitative customer feedback, or six months of campaign performance, and ask it to surface patterns. It will do this instantly and probably do it better than a human.

It’s also a discerning thought partner (as long as you avoid simply confirming your own beliefs) for working through a strategic problem, thinking creatively when you’re stuck, pressure-testing an angle, or providing fresh eyes when you’ve been staring at the same doc for too long.

There are thousands of other great use cases I could insert here, but none of them should enable marketers to get lazy in their judgment.

It can’t replace your judgment

Your strategic brain, human judgment, and customer understanding shouldn’t be outsourced to even the best AI model.

You could write the world’s most detailed prompt and still not capture every contextual strategic nuance that lives in your head. My argument is that it should stay that way, no matter how advanced these models get, while acknowledging we’re really at the tip of the iceberg with this technology.

I don’t have a crystal ball, but I believe the marketers who are going to be irreplaceable in a few years are the ones who kept doing this level of work themselves, even though it could technically be outsourced to AI. The contextualizing of information to feed to AI, and the filtering of its outputs to determine what gets used, are already emerging as important skills. Don’t let a flawed promise of efficiency be the reason to stop building that irreplaceable judgment.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{⏩ Forward this email to: someone who’s been running the same ad copy for 8 months}

👋 Hi, I’m Amanda. I’m a fractional head of growth with a decade of experience working with rapidly-growing B2C startups. I work with lean teams in a few ways:

✅ Fractional: You have an expertise gap on your team you need to fill with someone who can flex into both strategy and execution, for an open-ended time frame.

✅ Advisory: You need to validate your existing approach or are just starting out, and need someone with growth expertise to guide you.

✅ Project-based: You need someone experienced to execute quickly and cleanly on a specific scope and deliverable you have in mind.

If you’d like to explore working together, start here to tell me a bit about what you’re looking for. If you’re not quite ready but have growth questions you want to chat about, you can book office hours here.

Good copy, bad copy

One of the quickest ways to torpedo alllll the hard work behind your growth efforts is with bad copy. Bad copy is hard to define, but it’s one of those things where you’ll know it when you see it. Maybe it’s unclear, superfluous, trying too hard, or making a joke that just doesn’t land.

Good performance copy, on the other hand:

• is clear and concise,

• gets you interested while remaining authentic and honest,

• doesn’t try to be clever or punny,

• and follows a logical visual hierarchy that saves the reader any extra work.

The job of performance copy is to drive scroll-interrupting action in seconds, so without the right balance of these four things, the rest of your strategy won’t even have a chance to shine.

If you go audit all your copy right now, chances are you might not even recognize some of it because it’s been running so long. Maybe you even identify some “bad” copy, even though you know what “good” looks like. This is incredibly common, because copy often falls through the cracks in an otherwise tight and cohesive growth strategy.

Where there’s a gap

A lot of early stage brands don’t have dedicated copywriters, which forces growth marketers to be the copywriters. This isn’t a bad thing – it requires you to get even closer to your customer’s mindset.

But oftentimes, this makes copy an afterthought, when it deserves so much better. Without well thought-out copy, you don’t have a way to actually speak to your customer, and the rest of your strategy becomes meaningless.

Some growing brands are lucky enough to have dedicated copywriters, who really understand the customer and what makes them tick. But they may be working across different business functions, and performance marketing copy, sales collateral copy, brand marketing copy, and event marketing copy are four completely different languages.

Messaging and angles may be consistent across customer-facing touchpoints, but specific copy must be tailored to meet the user where they’re at.

The job of some copy may be to tell a story, or provide a paragraph’s worth of useful details, but performance copy has to do more with less. Whether or not growth marketers the ones writing final copy, they’re responsible for making this distinction, to make sure “good copy” for an upcoming conference doesn’t become “bad copy” in a crowded social feed.

How to prioritize copy

Copy efforts tend to be missing from the detailed analyses, roadmaps, and creative strategy growth teams spend lots of time and effort on. Here’s where I’d start to make copy a core part of growth strategy.

Take these 4 exercises back to your team to bring copy to the forefront:

The clarifier

Take existing copy from your ads, your website, any OOH advertising you’ve done, internal brand books, etc. Try to compile 5-10 copy points.

Take each one, and edit it down to make it clearer, punchier, and more performance focused.

Now take each updated line, and do this again. You’ve now started to strengthen your performance copy muscle, and you have 5-10 new performance-focused copy angles ready to use.

The clean test

Take an ad that’s been running for a long time. Come up with two new copy variants for it, then hold all else constant and run your original against your two new versions. No matter which version wins, use results directionally (don’t obsess over stat sig in this case) to come up with 2 more versions. Repeat.

The repository

Build a repository of all your top performing copy. Compiling all these lines together should surface clear insights into what it is that’s working (and what’s not).

The audience split

Try copy angles that are distinctly emotional vs. distinctly feature/benefit focused. The idea isn’t to prove that one works better than the other – different members of your target audience will respond to different things.

All of these exercises together should pretty quickly illuminate how copy is a true growth lever, not just a creative detail. When you focus on copy as much as other components of your strategy, performance improves, because your message has a chance to land.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{⏩ Forward this email to: a media buyer who wants a promotion}

👋 Hi, I’m Amanda. I’m a fractional head of growth with a decade of experience working with rapidly-growing B2C startups. I work with lean teams in a few ways:

✅ Fractional: You have an expertise gap on your team you need to fill with someone who can flex into both strategy and execution, for an open-ended time frame.

✅ Advisory: You need to validate your existing approach or are just starting out, and need someone with growth expertise to guide you.

✅ Project-based: You need someone experienced to execute quickly and cleanly on a specific scope and deliverable you have in mind.

If you’d like to explore working together, start here to tell me a bit about what you’re looking for. If you’re not quite ready but have growth questions you want to chat about, you can book office hours here.

Shifting to a growth mindset

One of the misconceptions I hear over and over about growth marketing is that people assume growth marketing = paid ads. Of course paid ads are a huge part of most successful growth marketing strategies, especially if you need to learn and acquire customers more quickly than with “free”, but slow-burn, organic marketing efforts.

But growth marketing is an entire operational function that spans from messaging and positioning, to acquisition strategy, to conversion optimization, to analytics and modeling (and much more). This means a full stack growth marketer has many more tools in their toolbox and levers to pull than someone who is focused exclusively on media buying. Most importantly, operating with a growth mindset leads to effective prioritization of these levers for business impact, without getting distracted by noise.

The media buyer mindset

When you’re operating with a media buyer mindset, you’re deep in the weeds. You’re up to date on every new feature a platform releases, toggling and tweaking settings, working closely with account reps, and closely watching your CTRs and CPCs tick up or down. As a channel expert, you have your finger on the pulse of what Meta, Google, TikTok, Pinterest, etc. are doing to – theoretically – make your ads better.

To be clear, this level of detail is really important to a well-functioning growth org. But the most critical task is to be able to filter through all these features and settings to determine what drives true impact, or you may be wasting time forcing a square peg into a round hole. Most of these levers are optimizations at the edges, and probably won’t end up having meaningful business impact.

The growth marketer mindset

A full-stack growth marketing mindset involves zooming out. Growth marketers start with top-line KPIs, like revenue, or CAC, or LTV, and then work backwards into which levers across the business can actually drive those numbers.

Sometimes that lever is a paid acquisition test, but other times it’s restructuring a landing page to present information differently to a key audience segment. Or it’s running a test that changes the onboarding experience to improve conversion to activation. Or it’s building a new reporting view that drives action by providing new insight into which channels drive the most valuable retention, not just the most conversions.

Deciding where to focus your effort and what kinds of trade-offs to make is what separates a growth mindset from a media buyer mindset.

Tapping into a growth mindset requires being open to expanding the universe of possible tools in your toolbox you can use to drive the business forward.

Balancing short and long term

Where media buyers tend to focus on what the platform data is telling them right now (CPM is up, CPCs are down, ROAS looks strong), growth marketers think in longer time horizons. They know that paid ads can deliver quick insights and immediate customer acquisition, but they also know those wins won’t be sustainable if the business isn’t thinking about all stages of the customer experience. That might mean anything from investing in organic content or focusing on retention, to developing a user feedback loop and making meaningful product improvements. Growth requires balancing immediate traction with compounding, longer-term bets.

Rather than try to prove or disprove the value of every feature an ad platform gives them, growth marketers use ad platforms (and every other channel) as a tool to drive their goals.

Ruthless prioritization

This shift in thinking comes down to prioritization.

A media buyer mindset might lead to preserving campaigns even when they’re no longer serving their purpose, and hoping a tweak or toggle will unlock performance. A growth marketer mindset will instead lead to cutting losses quickly, reallocating budget and effort, and moving on to higher-leverage tests.

A growth marketer mindset also ignores a lot of noise. Growth marketers know that some metrics don’t matter when others aren’t where they need to be, and which opportunities are relevant while others aren’t worthwhile. Rather than getting distracted by vanity metrics or the newest features, they stay focused on what ladders up directly to business outcomes.

This discipline is what allows growth marketers to move fast, not because they’re running more tests, but because they’re ruthless about focusing on the right ones.

Reframes to try

Test yourself. If you’re ready to uplevel your thinking as a growth marketer, try these reframes:

• If you think: “My job is to drive the best performance on paid channels.” → Try thinking: “My job is to grow the business using paid marketing as one lever.”

• If you think: “This campaign has a good CPA, let’s scale.” → Try thinking: “Is this campaign driving high-quality customers at the right CAC?”

• If you think: “I want to test this new platform feature to see if it improves performance.” → Try thinking: “Knowing what we know about our customers, will this feature will help us better achieve our KPI?”

• If you think: “Spending more on our paid channels will drive more growth.” → Try thinking: “Optimizing our acquisition, conversion, and retention levers will drive more growth.”

The TL;DR

Where media buyers know the tools inside and out, growth marketers know how to use the tools to drive the business.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{⏩ Forward this email to: someone on a lean creative team}

👋 Hi, I’m Amanda. I’m a fractional head of growth with a decade of experience working with rapidly-growing B2C startups. I work with lean teams in a few ways:

✅ Fractional: You have an expertise gap on your team you need to fill with someone who can flex into both strategy and execution, for an open-ended time frame.

✅ Advisory: You’re either just starting out or need to validate your existing approach, and need someone with growth expertise to guide you.

✅ Project-based: You have a specific scope and deliverable in mind, and need someone experienced to execute quickly and cleanly.

If you’d like to explore working together, start here to tell me a bit about what you’re looking for. If you’re not quite ready but have growth questions you want to chat about, you can book office hours here.

This week, I spoke with Matthew Gattozzi, founder of Goodo Studios, a creative studio focused on creating ads for consumer brands. His approach is to connect creative strategy and production, using AI only where it makes sense for the business, to create ads that convert viewers into profit for brands.

We talked about:

• 🏆 When AI is the best tool for the job, versus when it’s not

• 🎨 How to think about using AI for creative, from an honest standpoint

• 🎥 How his team uses tools like ElevenLabs and Adobe Premiere in their development process

• ✏️ How a physical pen and paper (really!) are foundational to your AI tech stack

One of my favorite parts about this conversation is Matt’s clear framework for assessing when to use AI. Spoiler alert: “Because everyone else is” or “It just feels like I should” isn’t a good reason — instead, Matt asks a simple question: is this going to help get us the best possible outcome, more efficiently?

Our conversation, edited for clarity, is below.

Amanda Berg
Could you share a little bit about who you are, what you do, and what your journey into incorporating AI more into your day to day has looked like?

Matthew Gattozzi
I’m the founder of Goodo Studios, a content and creative studio that works with consumer brands on converting viewers into profit. Our whole thesis is that we want creative strategy and production living in one place and working together as closely as possible, to create the most effective content for consumer brands. The old way is having a big agency overseeing strategy, and a separate production company doing the creative work. And those two incentives are very different. 

The rise of social media opened up the ability for a creative person like myself to make a living with a camera, and that business like food trucks needed Instagram content just as much as massive brands like Nike.

I realized that I needed to bridge that gap between marketing and production. The pandemic and the rise of TikTok supercharged the need for creative for consumer brands, because these things changed where advertising is consumed — primarily in a video format.

Where AI comes in is that we operate in a performance landscape for our clients, so anytime we can leverage AI to get deeper insights to make the ads more effective, the better. For us, the use case for AI is more so in insights than it is for production.

What I care about is, how can we make the best ads? And if that’s using a cinema camera or using AI to generate an image, I don't care. I just care about outcomes for our clients, and I will use whatever tools we have at our disposal. I think a camera still works just fine, but I think if that changes, I'm okay with changing that, too.

Amanda
What did a typical day look like for you and your team before you started integrating AI into your workflow? When did you start making this shift?

Matthew
One of the biggest lifts that goes into producing these ads is the strategy behind them.

When we initially take on a client, we're doing everything from researching the product, their competitors, looking at reviews, to really trying to understand the psychographics of the consumers. There's a ton of research that has to get done when we first start, and as recently as two years ago, all of that was manual.

In between the steps of doing the research and documenting it all, all that info would live in a strategist’s head. The initial searching, building of tables, reviewing competitive ads, etcetera was done by someone and then synthesized into a doc. So that takes time and really limits the number of clients one person could take on.

Then, you’ve got this rapid feedback loop of data from Meta and Youtube ads, so every week we’re coming up with new ad iterations. We're trying to find new headlines, new statics, and testing on a very deep, detailed level. And in writing that brief each time, you’re starting from a blank page. So there was a lot of mental and manual load there.

Separately, because we own the production and editing process, there are situations where a creator would send video back to us and they’ve misread a line, or a brand name hasn’t been pronounced right. So we’d need to go back and forth, rebook talent, and get video re-recorded. You can see how all of this process limits what we could accomplish in a day.

Then around two years ago, we started using ChatGPT to analyze a CSV of an ecom site to pull out key themes, instead of reading the whole site manually. And slowly but surely, as these models have improved, we can use AI to search for things like competitors and UVPs. This just means you can get up to speed that much faster.

We were using Claude and some other tools to help us with writing and research last year, but it wasn't until 2025 where we were clear with everyone on the team that we are AI-first, because the tools are finally where we need them to be to give us an advantage. Two years ago there was still some hesitancy, but now it feels like everyone is on board.

Amanda
Aside from the upleveling that AI has been able to bring to the research and editing processes for your team, are you using AI in the actual creative development and content generation process? How do you think about the balance of what should be AI generated, versus what should start with a camera?

Matthew
One thing that we’re working on that’s not quite there yet is the use of AI greenscreen technology. Right now, a green screen can fall short when the background image doesn’t match the lighting of the person or elements you’re actually filming with. Today, when we need a person talking in a living room, we have to go rent that living room space. What we’re starting to build out is the ability to create that background space with lighting that matches the person you’re filming exactly.

Of course this technology already exists, but it requires a ton of CGI investment. If you can pull this off with AI, and can generate, for example, an upscale kitchen background for a high AOV customer without having to rent that space, that opens up a ton of possibilities. The future state is having a full “green screen studio” where you’re able to film entire videos without actually having to rent a house.

We’ve not yet done AI generated people or avatars or anything like that, because I don’t think the technology is there yet. Some people may say it is with tools like Google Veo3, which is a few hundred bucks a month. But it takes a really long time to generate those videos. And if you need to make edits and re-prompt, it can take a week to get the 5 seconds of video you need, so that’s not an efficient use case compared to a person who can do that work much more quickly.

So for us, using these tools is not so much a philosophical decision, but more about whether it’s actually efficient and helps us.

Amanda
As a creative yourself, do you have a POV on where to draw the line between what AI can create versus what humans should create?

Matthew
These are just tools we can use to create. Personally, I have digital cameras, but I still love shooting film —there's something beautiful about it. And film is a tool, it gives you a different aesthetic that digital just can't get. Where the line comes in is whether you’re being honest. You can't have an AI avatar say, “I lost 100 pounds using this supplement.” But an AI avatar can absolutely help create a new way to communicate your product of your brand.

So if these tools improve to allow people to be creative who otherwise may not be, and that barrier to entry doesn't exist anymore, I think that's a beautiful thing and we should embrace that.

Amanda
You talked a little bit about using AI to help with some post-production woes, like if somebody mispronounces a brand name. Can you share more about how you’re using AI in post-production to speed up the process?

Matthew
We use Eleven Labs to help us with voiceovers. If a client comes back to us and needs us to update or change a line, or they have a new idea they want us to incorporate, we don’t have to wait to get a voiceover actor, we can just generate a new voice.

So all of a sudden the speed from idea to execution is so much faster, because we already have all the footage we need. The barrier used to be the voiceover, whether we had an actor or were doing it ourselves, but now our editors can just input a script, make sure it sounds great, then export the content into Adobe Premiere and finalize the ad.

There are other ways AI in Premiere has helped us with sound design too, where we can make sure music sounds seamless without us needing to make the decisions and doing the technical work of where to cut. Auto-captioning is also so much stronger now, as is tagging clips, so our editors can just search for what they need rather than digging through 300 different clips to find what they’re looking for.

All these features Adobe is coming out with are crazy and have saved us so much time, which means ads can be more effective, and that gives us space to do better work. Most people think the time saved means we can produce more output, but the way we look at it is that we can significantly increase the quality of our work.

Amanda
Are you able to quantify how much time do you feel like your current process has saved your team? Tell me more about how you think about using that time to be more productive, whatever that means to your team.

Matthew
It’s so hard to quantify time, but it’s probably over a hundred hours.

In terms of how we think about it, it’s not like, okay, now people only work 10 hours a week. It's that instead of doing manual work that can be automated, now they get to work on more strategic things. I believe everybody in the world will become creative, because everything else is going to be automated at one point.

It’s going to be less about “can we get that data into an Excel sheet?” and more “how do we interpret this data and what actions do we take as a result?” Opening up that higher level, higher quality thinking is a way better use of a strategist’s time than putting data into a table to present to a client.

Amanda
From an organizational level, were there any aspects of AI adoption across your team that have been messy or haven't gone as smoothly as you would have hoped?

Matthew
I think that there’s this desire to automate things, but the reality is to automate something, you have to know your process first. When the SOP isn’t clear in the first place, that’s where I see people getting tripped up.

The other thing is that when people do get excited about automation, they want to automate everything. But maybe you don’t need that fancy Slack notification in the first place. So before we look to automate things, we look to eliminate things.

Also, I acknowledge that not everyone uses AI the exact same way I use AI, so we need to create flexibility with the tools that we use and how we’re using them. The hard part is that these tools are changing so fast, and it’s challenging for your tech stack to keep up. You used to have a bit more time to compare your options when buying SaaS, and now it feels like you have to move really quickly, but also avoid continually re-building processes you’ve already invested in automating, so we have to strike that balance.

Amanda
Everyone using AI in different ways is a great point. Most people are thinking about AI as this defined thing that exists in a box, when in reality AI is starting to have the ubiquity of “the internet.” And in the same way that how I use the internet is not going to be the exact same as the way any other person on earth uses the internet, the same can be said of AI. It sounds like that presents new problems organizationally to figure out what your AI tech stack looks like, at least in the current state with the number of tools out there.  

Even with the number of tools we have at our disposal now, I’m curious whether there have been any specific work streams or processes that you've tried to apply AI to that just haven't been a good fit for any reason.

Matthew
In terms of ad creation, I think there are a lot of tools that overpromise right now. Even if you have the perfect prompt and the output was exactly what you wanted, you end up with text baked onto an image. What if I want to change that text? I don’t have a file where I’m able do that, so there are some things that just don’t completely work yet. Sure there are ways around this, but you have to diverge from your existing workflow or add steps, which is inefficient.

I also think there’s a lack of good qualitative analysis from performance data, which is what really moves the needle. An AI agent can tell you which ad is your top performer, but they can’t watch the video so they don’t know why. They don’t know why you made this particular ad in the first place. So agents that can identify metrics for you aren’t super helpful, because we can just identify those metrics ourselves.

Amanda
How have you handled adoption of AI tools at your org such that the human elements of creative work, and focusing on the ability to do that deeper work, are still respected?

Matthew
Because our whole approach is to tie strategy and production so closely together, our strategists are the ones directing the shoot. That’s what differentiates us, and because they’re doing all the ideation, directing, and creation, they know that’s not going away in the near term. If your creative strategists are essentially just sitting behind a computer, writing and coordinating and reviewing content, a lot of that can be automated. For us, that automation is a good thing, because then we can focus on creating the best possible content and making it stand out and feel human.

Amanda
If you were advising a marketing team of one or a creative team of one, which is not super uncommon at small startups, is there a particular AI tech stack or starting point that you would recommend to them?

Matthew
Claude is the best writer, so I would start there. You could toggle between Claude, Perplexity, Grok, and ChatGPT for more searches and queries, it just depends how much you want to spend to get the best of everything. Where I’d really start is by getting used to using AI in everything you do. Next time you need to respond to an email, put that info into AI and get used to prompting it the right way.

I wouldn’t worry too much about automations and building out these crazy workflows at the start, because it’s so much more important to get comfortable with just using AI as a tool. Feedback is instant, so you can learn how to improve your prompting quickly.

And when you are ready to start automating, go back to pen and paper or post its and write out your process and your SOPs. You’ll either figure out what you can eliminate, or you’ll figure out what you can get ChatGPT or Claude to do — and there’s no cost to pen and paper plus an LLM being your tech stack.

In my case, I need to know what makes a great ad. So what I would do is spend some time creating a document that says, this is what a great static ad is, this is what a great video ad is. If you spend that time upfront to define what success looks like, the more time you’ll end up saving. If you get a bad output from AI, it’s because your input isn’t great. Focusing your time there instead of trying to be at the cutting edge of the latest automation will be much more valuable.

Amanda
Any closing thoughts that you want to share?

Matthew
Ultimately, the most important question to answer is: why use AI for any given task? If it’s FOMO or it’s just to be able to say you have this AI agentic workflow, you’ve probably missed the point for your business, which is to focus on driving the most revenue and profit in the most efficient way possible. Consider your main goals, ask whether AI allows you to do that, and only use it if the answer is yes.

I recently had a photoshoot where I was able to tell my photographer and model in real time how to adjust, and had tons of great content for the client to review. It would have taken so much longer for me to prompt all of that, the packaging wouldn’t have looked right, and that’s not what would have been best for our client. A real photo is actually the best output possible right now, so it’s not worth the tradeoff.

As a leader, I want to encourage innovation and encourage people to use these tools, but I also need to encourage that principled thinking so we don’t feel like we’re missing anything when we choose not to use AI. It’s all changing so quickly, and as long as we’re continuing to grow our business and our clients’ businesses, that’s what matters.

💡 Are you a marketer using (or figuring out how to use) AI in your work? I’d love to chat with you for the newsletter. Just hit reply and give me a 1 - 2 sentence summary of your current relationship to AI.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{⏩ Forward this email to: a new marketing AI enthusiast}

👋 Hi, I’m Amanda. I’m a fractional head of growth with a decade of experience working with rapidly-growing B2C startups. I work with lean teams in a few ways:

✅ Fractional: You have an expertise gap on your team you need to fill with someone who can flex into both strategy and execution, for an open-ended time frame.

✅ Advisory: You’re either just starting out or need to validate your existing approach, and need someone with growth expertise to guide you.

✅ Project-based: You have a specific scope and deliverable in mind, and need someone experienced to execute quickly and cleanly.

If you’d like to explore working together, start here to tell me a bit about what you’re looking for. If you’re not quite ready but have growth questions you want to chat about, you can book office hours here.

This week, I spoke with Sea-Jay Van der Ploeg about how she’s using AI as a marketer. Sea-Jay is a growth marketer and creative lead at Nourish, a telehealth dietitian startup. She’s led growth for consumer, healthcare, and wellness brands including Curology, Parade, and Julie.

She started earnestly integrating AI tools into her day to day at the beginning of this year, and we had a really fascinating conversation about how she’s tactically using AI after effectively teaching herself how to do so, some of the ethical implications of using AI creatively, and how she thinks about AI as a strategic augment to her work, rather than a replacement for critical job functions.

Some of my favorite parts of this conversation focused on how we both believe human thinking is going to remain critical to do good work — perhaps even more so — even when it can feel at times like AI tools are capable of doing almost too much for us.

I so enjoyed her sharp and measured perspective on this beast of a topic — our conversation, edited for clarity, is below.

P.S. — I’ve always been a proud em dash user and always will be, and none of this newsletter is edited by AI!

Amanda Berg
Could you tell me a little bit just in general about your current role, and what your day to day looks like?

Sea-Jay Van der Ploeg
I’m a growth marketing and performance creative lead at Nourish, a telehealth dietitian startup. I joined Nourish about a year ago as the second growth marketer, and in that time my day to day has transformed rapidly. In the earlier days, I was very much wearing all the hats like hands-on-keys media buying, setting strategy, writing all the briefs, setting creative strategy and influencer marketing operations.

Currently, my day to day is a bit more focused exclusively on that creative strategy component, so standing up the performance creative team, working with them on performance messaging, and standing up all the operations there.

Amanda 
Was there a time in this role before you implemented AI into your workflows?

Sea-Jay
AI integration for me has been quite recent. Starting in the new year is really when I started to ramp up more. Prior, I was doing a lot the executional marketing work by myself, and so thinking about how to scale that, I just turned naturally to AI.

Amanda
Where did you start? Are there any specific tools that you lean most heavily on?

Sea-Jay
The main tool I’m using the most is ChatGPT, including building custom GPTs, and I use Claude a bit as well. I don’t have a real preference for one or the other right now — it comes down to how these models integrate with some of the other tools I use. I've also experimented a little bit with building things in Lovable. For the coming quarter, there are some tools I’m keen to try after attending a Motion Creative Strategy webinar, like Poppy AI and some Figma Make AI plugins.

Amanda
Do you find that there are specific types of tasks where you will reach for ChatGPT over Claude or vice versa? Or do you like to try both and see which one gives you a better output?

Sea-Jay
For really robust visual analysis and image generation, I think ChatGPT is better right now. I've tried using Lovable and Claude to do some stuff that requires looking at visuals, and it was just less strong. Media buying and reporting is also something we’re leaning on AI to support us on — we still have dashboards and I pull a lot of things manually, but it’s been interesting to see how good ChatGPT is at coming to the same conclusions. 

We’ve tried feeding it raw performance data to see whether it can come to the same conclusions about where to toggle, where to turn off, where to put budget, and we’ve seen it can really help make a lot of those decisions. I’ve even built a custom GPT to help with naming ads in bulk uploads, since otherwise that’s such a manual process.

Creative strategy is another big one. I talk to a lot of folks who are like, how do I systematize learnings and understand what worked the last couple months, and then create briefs to reflect that? I've been building a knowledge base of all those performance learnings to utilize when I'm thinking about the next brief or next roadmap. I've saved a lot of time with brief writing that way. 

ChatGPT has also been useful for content and creative feedback. That said, I'm a little more picky about whether or not we want to use fully AI generated creative, depending on the use case. But I do think AI has been great for giving you a visual when you’re giving creative feedback on ads to designers. 

And then lastly, I’ve used it a lot for setting up scripts and automations in place of needing to rely on a growth engineer — Claude or ChatGPT can just teach me how to do it.

Amanda
How much effort does it take to get to a place where you feel like your custom GPTs are both usable and useful, and what’s that learning curve like? How much information do you need to provide?

Sea-Jay
I thought about the things I maybe would have had a virtual assistant or an intern do, and how I would have to train them to take on these tasks. And instead, I tried using that time to work on these custom GPTs a few hours a week. So I did this with the ad naming GPT that I built, and it probably took me a few hours after work every couple days over the course of a week or two to really get the hang of it. 

But once you figure out the best way to interact with ChatGPT, then it’s just a lot of bulk uploading information. And then you realize, if I can teach it how to do that thing, I can probably teach it how to do this other thing.

Amanda
Can you tell me a little bit more about how you are using ChatGPT to speed up the briefing process? Are you relying on another custom GPT with historical performance data, or a repertoire of assets that have already been built?

Sea-Jay
There’s a short term and a long term answer. In the short term, we’re using ChatGPT to do the actual writing, templatization, and putting all the information on one page. It's been a lot easier to just say to ChatGPT, here’s what I’m thinking in broad strokes, pull up the brief template, and fill it in. And that saves a lot of time versus me typing the whole thing up from scratch because then, I'm able to go in and just tweak a few things. 

Then the longer term approach that I'm trying to build this quarter is what you mentioned in terms of historical performance data. We have all of our top value props saved somewhere, we have another weekly log of which ads perform well, and all that knowledge is in somebody's head right now. If I give the GPT that knowledge, then tell it what direction we’re thinking of going in for next quarter, and have it help build a roadmap, it can give me five sample briefs it could write. And it shouldn’t supplant my work, but it helps a lot with the synthesis of it.

Amanda
I’ve always found building process and deriving insights from creative testing to be a bit of a struggle. Have you been able to use AI to come up with a process there that replaces the clunkiness of using spreadsheets or the need to pay for a separate creative performance tracking tool?

Sea-Jay
We have a loose system where we have a weekly readout on what's going on in the ad accounts, then we also do a monthly summary of higher level themes, and that is used to inform upcoming briefs but also give our creator and creative teams the ability to source and create stuff based off of those learnings.The first step is building that process internally to summarize those learnings in some capacity. For example, we’ve moved to using Notion to do a write-up, so I then take that and download it, and then feed it back to ChatGPT on a weekly or monthly basis.

Amanda
How much time are you now saving in a given week? 

Sea-Jay
For things like creative strategy, it’s cut down something that took up to five hours to probably one. For little tasks like ad naming, it’s cut down on some percentage of time. So it’s hard to add that all up, but I'm noticing that it just used to take so much longer.

Amanda
Aside from using that time to actually be able to think about the work, rather than to sit down and do all the manual stuff, are there other intangible benefits that you’ve found?

Sea-Jay
The meaningful higher leverage thinking time for myself and the team is really important. For example, we have designers who were working on making lo-fi static assets that were pretty easy but manual, but now we can use something like Figma Make for those and free them up to do more interesting work that requires more exploration. It gives us all back some brain space.

Amanda
Did you take it upon yourself to start using AI in your workflows, was it more of an organizational mandate or encouragement, or kind of somewhere in the middle?

Sea-Jay
Definitely in the middle. What's funny is before the new year, AI felt very “tech”, almost like crypto, and I was just like, I’m not into this. And then it started to be something popping up more in conversations with friends and I was like, okay, I could actually really use some resourcing help, so let me just try it out. And after seeing a friend of mine explain how they were using it, that gave me the confidence to start playing around.

From an organizational perspective, in the last month or so we've started a Slack channel where people are talking more about how they’re using AI. We’re all kind of playing around, and it’s been helpful to find folks to talk about it with.

Amanda
It sounds like from an organizational standpoint, there is encouragement and appetite to be using AI to make things more efficient.

Sea-Jay
There’s definitely cultural encouragement, and willingness to pay for access to these tools, which is amazing.

Amanda
I'm curious if there have been any messier aspects of AI adoption across your team, or processes you haven’t quite figured out the right way to utilize AI for yet.

Sea-Jay
It hasn't been a major trip up, but something worth discussing is the perspective around creative generation and creative guardrails. We’ve talked a lot about what feels honest and authentic to our brand and our storytelling, what looks good versus bad, and how we want to talk about all this with our creative teammates, for whom this is very much an existential question about their craft. 

Overall, I've tried to make it a very comfortable space to be able to share, and remain critical, so we don’t turn into this pro-AI cult. There's been a lot of open dialogue because there are still some ethical questions about the AI space. We’re making the space to discuss that in our monthly brainstorms, especially with members of our creative team. 

As for one example, we portray a lot of food imagery in our ads. And so we said, okay, AI can make a banana, that’s fine. We also include a lot of customer testimonials, and there it felt dishonest to use an AI-generated “customer.” So starting to have those conversations about how to honestly use AI in our creative has been important.

Amanda
You've already identified that there are things that AI is capable of doing that it probably, for moral or ethical reasons, shouldn't be doing right now for your specific job function. So it sounds like you’re starting to figure out where those lines are — how are you broaching those conversations with your team?

Sea-Jay 
We haven’t had super broad discussions about it, but it’s something I've personally been thinking about. We did have an initial conversation with our creative team members. I started with a slide of different creative POVs across the spectrum — for example, we had Hayao Miyazaki of Studio Ghibli, who is a strong critic of AI. And on the opposite end, there’s Kelly Wearstler, who uses AI to design beautiful spaces and is creating GPTs to do that. So we created this open space for people to talk about how they’re feeling, relative to these opposite points of view. 

People were pretty honest about their vulnerabilities, but also how they think they could explore using AI in a way that felt interesting and exciting, and that augmented their work. 

The other thing I'll mention, too, is that it's important to help set a perspective for the team, and my perspective has been that I don't think creatives will ever go away. And I really see AI as a tool to enhance your ability to create.

More and more we're going to see the importance of being a true creative director with AI as an augment, and creatives should feel empowered to get to that point. And I think most creatives want to be fluent in these tools, because it can be really empowering to be like, wait, actually, I can articulate my vision. Framing it like that has been useful because we all want to set a vision and a direction, and AI can help us do that. But it is all very existential, and I don’t have all the answers.

Amanda
Are there any processes or work streams that you’ve tried using AI for that haven’t been a good fit for whatever reason?

Sea-Jay
Two quick things that come to mind. As a marketer, I’ve found that some things just don’t need to be fully automated with AI, and instead would benefit from a robust automation flow. I can also have AI teach me how to do something, rather than creating a whole custom GPT around it. There are some problems that do need to be solved, but not necessarily with AI.

And the other thing that comes to mind is video generation. Right now, it’s still at the point where you watch it and immediately realize something is “off.” For us, it doesn’t feel quite perfect enough yet for us to be doing much with it, though that may change.

Amanda
Are there any other places where you feel like there have been gaps with the promise of AI versus the reality of what it can actually deliver?

Sea-Jay
My answer is slightly tangential, but I think a gap I’ve noticed is in prompt engineering as the thing that's going to separate people who can leverage the tool versus those who can't. That’s what’s going to differentiate the next class of growth marketers.

The quality of the output for all these tools is very much limited by your ability to articulate what you want. For me, using more AI tools has actually inspired me to think more about how much more important creative direction and clarity of thought is, because if you just can't do those things, those tools will never help you accomplish what you need them to do.

If you don't exercise those muscles, then your tool will just inherently be limited by the quality of your prompting.

Amanda
That’s a really, really good point. Prompting feels to me like a skill we all need to get fluent in, and is distinct from other communication skills because you're not working with other people, you're working with a machine that the output is going to be based on your input. 

It’s such an important reminder that at the same time, human thinking will continue to be important, but perhaps in different ways. You still need to come from a place of giving clear direction and being articulate to your point. I wonder if that ability to articulate is something that you've had to work on to get better output from these tools, or is that an inherent skill you feel like you’ve always had?

Sea-Jay
I've been fortunate to feel like I've had this skill, especially when I’ve needed to give creative direction and feedback on assets, which I'm sure every marketer has done. Obviously, you can't just be like, “I don't like it.” And if you just told a GPT “I don’t like it,” it wouldn’t know what to do next, versus giving clear and specific instruction. 

I also think at a higher level, being a good prompt engineer is very similar to someone who is very good at giving feedback. If you have those skills and just are able to be specific, it lends itself to prompt engineering pretty closely.

Amanda
How do you and your team keep up when there are new tools popping up every day and LLMs are evolving every week? And how do you assess what tools or updates are worthwhile to explore, when it feels like a total free-for-all and your options are expanding exponentially right now?

Sea-Jay
I see a lot on LinkedIn, I see a lot on TikTok that I’ve been saving and trying for myself, I try to sit in on webinars where new tools are being demoed, and I try to just talk with other people in the space. The other day a friend of a friend said they built an AI agent, and I was like, can we hang out? Can you show me? I’m totally non apologetic about it, and I’m just trying to learn and trial new tools as much as I can. 

The other thing that’s been helpful is to actually just brainstorm all the things you want to solve. It's very easy to go in and be like I'm going to make some pictures, versus thinking about what are some actual problems that I'm facing, that I think could expedite or enhance, and particularly broadening it beyond work. I’m using ChatGPT to help me get organized for a vacation, so figuring out ways to use these tools to solve other problems in my life has been helpful too. 

Amanda
Regarding some of the bigger existential questions, I’m curious about your POV on where the lines should get drawn around when we should and shouldn’t be using AI in our work.

Sea-Jay
Our conversation makes me think that like, number one, humans should be responsible for determining what those guardrails are that you're creating as an organization. AI shouldn’t be left to make those rules. 

As it relates to marketing, I think AI can get very close to replicating and simulating creative taste, but I don't think it can quite achieve emotional resonance. I don't think it'll ever fully get there. We're marketing things that other humans are going to see or read or watch and be like, wow, that's hilarious, or like, that's so moving. We still need a human in the room to like, just gut check that and make sure that it all feels right. I think it’s more intangible. 

And lastly, I think it’s so important that we still have human touch points. I’ve read about people getting interviewed by an AI. That feels dystopian and makes me pretty skeptical — these things require human touchpoints. We can’t just never talk to a human again. 

Amanda
Are there specific elements of marketing as a job function that you feel concerned about AI replacing? 

Sea-Jay
I feel pretty concerned about entry level marketers. I was so excited to get my first job in marketing and not know anything, yet there was still work for me to do, like the manual stuff that no one else wanted to take on. But I was so excited, whereas now, the market for new grads feels less promising, especially if companies are going to turn more to AI to do more of that work.

And then I do think we’ll always still have creative colleagues. The core of marketing is creative, even though there’s growth and data and analytics, but I think AI can only enable, rather than fully replace, that creative core. 

Amanda
I’ve thought a lot about that creative element specifically. If humans stop creating, then we’ll only be able to continue to create based on AI and existing knowledge. 

It’s like that prompt that was making the rounds a couple of weeks ago where it was like, take this image and duplicate it a hundred times, and you end up with something totally unrecognizable at the end because it's just continuous AI generated content that it's using to create the next thing. 

The magnitude is of course different, but if humans stop producing the inputs, then our outputs will feel un-human too. So I think there is actually incentive for people to continue to be the creators and to use AI as augmentation rather than replacement, especially since to your point, humans are the ones consuming and being marketed to, but I also have concerns about seeing where that line falls.

Sea-Jay 
Yeah, we could have a whole other conversation about that. Taste is changing. Public ability to discern what's true or false is changing. To your point, creativity is rooted in life experience. So I’m very curious to see whether you can output something that feels resonant if it's not based on a life experience. That's like a broader question I have. Like this robot, did this robot live through like X, Y and Z or can it replicate an output that looks like it did that? I don't know!

Amanda
Any closing thoughts regarding the use of AI in marketing and how it’s been helpful in your work?

Sea-Jay
It’s definitely generated a lot of questions that have led us to all have very interesting discussions. 

I think as a marketer, I’d love to find other marketers to talk about this with. There are so many bright and creative marketers who I’d love to share perspective and community with, in a way that doesn't feel like it needs to be super pro-AI. Just a space to have these kinds of conversations.

Marketing typically has an early adopter culture, and I’d love to see more marketers playing around with AI because it feels like we could really lead the charge here. 

💡 Are you a marketer using (or figuring out how to use) AI in your work? I’d love to chat with you for the newsletter. Just hit reply and give me a 1 - 2 sentence summary of your current relationship to AI.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{⏩ Forward this email to: a new founder}

👋 Hi, I’m Amanda. I’m a fractional head of growth with a decade of experience working with rapidly-growing B2C startups. If you’d like to explore working together, start here to tell me a bit about what you’re looking for. If you’re not quite ready yet and have growth questions you want to chat about, you can book office hours here.

I recently spoke with a new founder about marketing validation testing. She’s working on launching a tech hardware product, and wanted to see how to best validate demand for her idea before investing in it further. Specifically, she wanted to understand how to forecast her realistic revenue opportunity and how to message her product to the market, and had a small budget to work with.

If you’re in a similar boat, there are a bunch of ways to approach this, depending on the lift you’re willing to take on.

The “goldilocks” test approach

Here are the main tactics you could use to size and message to your market, ranked by level of investment, how quickly you can get answers, and how accurately those early signals will translate when you go live.

Surveys and interviews: Surveys are as low-lift as it gets — you can create a Google form for free and share it with your network. Interviews will require a bit more of your time, but will get you richer data. However, your results might be biased if most responses come from your own network, and it might take awhile to get enough input to inform any decisions.

Meta ads: Running a meta ads test requires some budget and some light technical set up (a landing page, some sort of conversion event you can use as a directional signal, and campaign setup), but will yield you a clear, data-backed signal within days. You’ll also be able to quickly gather directional signal on all your different messaging.

Kickstarter or preorders: Seeing what a conversion rate for actual purchases might look like is of course the most “accurate” test, but there’s probably a significant marketing lift here if you’re this early in your journey. Presumably, you also probably intend to ship product at the end of your pre-order campaign so as not to risk reputational damage to your brand, which means you need product ready to go.

Given all your options, the versatility of Meta as a channel makes it the “just right” balance of lift, speed, and accuracy for most founders at the earliest stages of their journey.

How to run this test on Meta

Before you get started, there are a few elements you’ll need to have ready to go. If you’re at the earliest stages of your journey, this might be the first time you set these things up:

• A Meta ads account

• Messaging and ad creative to test

• A landing page with some sort of action a user can take (ex: submitting their email to get on the waitlist)

• Conversion tracking set up on Meta for the action you decide to track

Once you have your ad account set up, you’ll want to think about your messaging and creative. This is a good time to start clearly defining the persona(s) that make up your audience, and the unique value props that speak to them. From there, you can build a few iterations of your marketing messaging. See more about building value props, and why they matter, here.

Then, you’ll need a landing page. I typically recommend the landing page builder Unbounce because it’s drag and drop, affordable, and has built in A/B testing functionality (that I’d recommend using later, after you gather your first signals). But depending how your site is or will be built, you may be fine with something like a Squarespace, or a Shopify landing page builder app.

On your landing page, you’ll need to have some sort of conversion action that a user can take, that you’ll use to validate demand. This can be something like a preorder or deposit, an email capture to get on the waitlist, or watching a product explainer video. Of course, the more similar this event is to your ultimate purchase, the more accurate your test results will be, especially if your product is niche or has a high price point.

How to interpret data from your validation test

Once all your elements are set up and you’re ready to launch your test, here are the data points you’ll want to pay attention to.

To understand the right messaging approach: Directionally assess click through and conversion rates per ad to determine the most engaging messaging. Your ads should be set up so each one speaks to a distinct persona in your audience, with a distinct message (Marketing mistake #1: Trying to speak to EVERYONE who could possibly be interested in your product. Your messaging will end up feeling bland and generic, and you’ll end up reaching no one). 

To understand your potential market: The funnel from Meta impression to click to conversion can help you approximate different layers of your market:

• Your total addressable market (TAM) can be very roughly approximated by the age, gender, and locations of the users Meta serves your ads to

• The portion of your TAM that makes up your serviceable market (aka the users you could realistically serve) can be roughly approximated by your click-through rate

• The portion of your TAM that makes up your obtainable market (aka the users you could realistically win) can be roughly approximated by your conversion rate, depending how close your test action is to your ultimate conversion event

Once you’ve run a test like this, you’ll have a really good idea how your product will be received by your audience, and where the strong and weak points are in your initial funnel. Then you can continue iterating on your audience definition, value props, marketing messaging, landing pages, and product development, so you can launch with a bang 💥

P.S. — Want to set up a marketing validation test of your own? Just respond to this email if you’d like to work together!

✨ One marketing thing: Will OpenAI’s decision to stay a nonprofit mean less likelihood of ads in Chat GPT? Discuss!

✨ One fun thing: The second episode of the new Seth Rogen show on Apple TV is one of my favorite episodes of TV I’ve seen in a long time.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{Forward this email to your paid search marketer}

Every so often, Google changes their ads UI and quietly introduces new settings. If you haven’t been deeply hands on with your search ads over the past few years, you may not be familiar with all of the settings that could be hiding in your account, wasting budget, and hindering scale. Google does a pretty good job of burying these, so here are some things you should look for, and where to find them, as of the current iteration of the ads UI in April 2025.

Auto-apply recommendations

You may notice when you log into your ads account that Google has “recommendations” for your campaigns. More often than not, these recommendations are designed to get you to spend more money, and aren’t going to help you achieve efficient performance. Google might be auto-adding new keywords, auto-duplicating exact match terms to broad match, or creating and launching ads for you if you haven’t checked on this setting in a while.

To review and adjust these settings:

➡️ Select “Recommendations” from the left hand menu

➡️ Select the “Auto-Apply” button on the top right, above your “optimization score”

➡️ Review all the options that are checked off, and un-check any that give Google too much control over your campaigns. Typically, the only options I like to keep checked are “Use optimized ad rotation” and whatever bidding settings are relevant.

Bidding for new customers

Google gives you the option to optimize bidding for new customers, which sounds great if your goals are focused on customer acquisition. The problem is that Google still isn’t great at definitively identifying which impressions come from new customers, despite all their data from Gmail and Chrome browsers. This means the vast majority of traffic is not able to be definitively identified as “new”, and so is not bid on by default. Campaigns with this setting may seriously struggle to scale.

To review and adjust this setting:

➡️ From the campaigns view, go into each campaign’s settings by hovering over the campaign name and clicking the gear button

➡️ Under the “Customer Acquisition” setting, check whether you’re bidding equally for new and existing customers, or optimizing for new customers only

Automatically created assets

Google gives you the option to allow it to create ads for you based on your landing page, your domain, and your existing ads. There are few instances I can think of where it makes sense to allow Google to write ad copy for you, especially because copy can be created based on anything that exists on your domain. Most campaigns are organized to advertise specific products or services, and giving Google free rein of all the copy on your site to create ads could be an organizational, messaging, and performance disaster.

To review and adjust this setting:

➡️ From the campaigns view, go into each campaign’s settings by hovering over the campaign name and clicking the gear button

➡️ Under the “Automatically created assets” setting, check whether you’ve toggled this on or off

Broad match expansion

The broad match expansion setting allows Google to bid on broad match versions of all the keywords in your campaign, even if you’ve set them up as exact or phrase match. This is a surefire way to show up for irrelevant searches and waste a whole bunch of money, especially if you don’t have an incredibly buttoned-up negative keyword list.

To review and adjust this setting:

➡️ From the campaigns view, go into each campaign’s settings by hovering over the campaign name and clicking the gear button

➡️ Under the “Broad match keywords” setting, check whether you’ve toggled this on or off

Networks

You might assume your Google ads will only show when someone searches on Google, but that’s only true if you’ve specified this in your networks settings. Aside from the Google SERP, ads may show up on sites within Google’s Search Partner Network or Display Network, without much visibility into performance specifics beyond these broad buckets. My typical recommendation to avoid low quality traffic is to start by including the partner network, but not the display network, and keep an eye on your performance breakdown.

To review and adjust this setting:

➡️ From the campaigns view, go into each campaign’s settings by hovering over the campaign name and clicking the gear button

➡️ Under the “Networks” setting, check whether you’re including Search Partners and / or the Display Network

Negative keywords

Robust negative keyword lists are critical to a scaled-but-efficient search strategy — I wouldn’t use broad match keywords without them. This tells Google which search terms NOT to surface your ads for; “free” is a basic example if you’re selling a product or service. Google has buried this setting with its most recent iteration of the ads UI, but here’s how to find it.

To review and adjust this setting:

➡️ Hit “Tools” in the far left menu

➡️ Open the “Shared Library” drop down

➡️ Click “Exclusion lists”

➡️ Hit the plus sign to create a new negative keywords lists, or click into existing lists to add keywords and ensure lists are applied to all relevant campaigns

Conversion columns

You should have conversions as a column in your campaigns, ad sets, ads, and keywords views. But depending on the conversion actions you have set up at the account level, you might want to double-check the specific columns you’re using. If you have non-KPI events set up as conversions and are using the “all conversions” column, you’re going to be over-counting the number of “true” KPI events.

To review and adjust these settings:

➡️ Click on “Columns” and “Modify columns” from your campaign (or ad set, ad, or keyword) view

➡️ Expand the “All columns” dropdown

➡️ Within the “All columns” dropdown, expand the “Conversions” dropdown, and review which metrics are checked off. “All conversions” is likely going to count every instance of events you have set up as a “conversion”, whereas “Conversions” will count those set as primary actions only.

Now what?

Reviewed all these settings and still not happy with performance? Just reply to this email and we can schedule a deep dive.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{⏩ Forward this email to a marketer who needs a vacation}

We all deserve to be here right about now.

Being a growth marketer in Q4 is not for the faint of heart. Every year, Q4 sneaks up on us again, and it’s like we completely forgot everything we learned last year about the importance of working ahead. Maybe you can’t teach an old dog new tricks…

That said, there are plenty of things you can do now to work ahead, especially now that BFCM is behind us.

Here are eight of the best things you can do before the end of the year to make sure Q1 feels like less of a doozy than Q4.

1. Review the assumptions going into your growth model or finance forecast

If you have both a growth model and a finance forecast that pulls from your growth model, you’re already in great shape. Even if you have just one or the other, it’s important to validate that the assumptions being built in make sense.

If you expect Meta to be your primary source of scale next year, what’s the plan for getting there and how does that plan translate into the assumptions you’ll build into your growth model? If you’re expanding into a handful of new geographies in 2025, what assumptions is finance making about marketing reach in those geos, knowing not all cities, states, or countries are equal from a population standpoint?

Assumptions are just that, and they should be built in a way that forecasts growth on the whole. But they should also be viewed as collaborative stretch goals rather than totally outside the realm of possibility. Nothing wrecks morale and motivation like goals you know are entirely unobtainable.

2. Take stock of influencer and creator content, and whitelisting agreements

What influencer content performed the best for you this year? It’s worth revisiting agreements with those creators, and continuing or expanding on those partnerships. It’s also important to try to standardize your influencer agreement as much as possible so your team doesn’t have to manually keep track of individual contractual quirks.

While you’re at it, make sure you have whitelisting rights for an extended period of time for content that performs well (it’s worth the cost). You don’t want TikTok ads going dark on you unexpectedly, or creators revoking access to their Meta biz managers because you forgot to extend your rights.

And wherever you can, diversify creator content so you’re not reliant on just one or two long-standing ads.

3. Audit your search ad copy and search term reports

Paid search specifics often get overlooked — there’s often an “if it ain’t broke don’t fix it” approach to search since it’s intent based and not as “fun” to work on as paid social.

But there’s so much efficiency and scale you could be leaving on the table by ignoring your ad copy and search term reports.

Most brands hardly ever touch their ad copy. This could be for good reason — it’s hard to beat winning ads, especially as character limits on Google have become more restrictive over time. But checking on top ads to ensure both copy and extensions are still relevant and optimal for your business is something worth doing a couple times a year.

Your search term reports, on the other hand, are where you’re going to uncover some immediate opportunities, especially if it’s been a while since you perused these reports. It’s a mistake not to look at these regularly to add or negate keywords. You’ll probably find there are search terms you’re not bidding on that are converting efficiently, as well as search terms matching to your keywords (particularly if you over-index on broad match) that are not relevant, don’t drive quality conversions, and are costing you money.

4. Take a look at your top paid social creatives from the year

Grab all your top performing paid social creatives from the last year and drop screenshots and videos into some slides or a Figma board. What do they have in common in terms of hierarchy of information, messaging, and style or visuals? This is the time to use your qualitative brain, and create more of what’s working for 2025.

Bonus points: Do the same thing for your worst performing paid social creatives — both the ones that never got traction, and the ones that spent plenty but converted inefficiently.

5. Compare your source of truth to platform data

Most brands are using a CRM, BI tool, or some other database as their source of truth, instead of relying on Meta or Google conversion tracking as gospel. That said, in-platform conversion data is critical for real-time creative and budgeting decisions, as well as directional performance insights.

Compare your in-platform conversions to source-of-truth conversions by month. Keep in mind that platform attribution settings are likely to be more generous than any last-click attribution your source of truth is using.

There will almost always be a discrepancy, but if that discrepancy has grown over time, or is consistently larger than, say, 30%, you’ll want to investigate and fix what’s going wrong so you can make sure you trust platform data for those real-time optimizations.

6. Organize all your backlogs and lists of crazy ideas, to prioritize them for 2025

You’ve probably sat in meetings this year that turned into brainstorms or problem solving sessions, came up with some great ideas of data to analyze or ideas to test out, and then never got back to those ideas because the day-to-day of your job got in the way.

Hopefully you jotted those down in a note somewhere, or maybe you even have a formal backlog. Run through your notes and organize them into a clean list, decide what’s still relevant and worth doing, and prioritize them accordingly to get ahead for 2025 planning.

7. Review learnings from tests from the year

You’re probably running a lot of tests, all the time. And you probably don’t remember off the top of your head what you learned from that one landing page test or ad copy test from last February.

First, make sure your test results are organized into a central tracking document. Then, go back through all the tests from the past year and remind yourself what you tested and how your hypotheses held up. Maybe some of the results feel really obvious to you now, while others still feel like they should have panned out differently.

In any case, compiling all learnings from the year should surface some patterns, and give you a great jumping off point for planning a testing backlog for 2025.

8. Review how you’re documenting learnings, change logs, etc for repeatable growth

One of the themes here is good documentation, organization, and note-taking. If you don’t have any formalized system in place for documenting learnings, test results, or change logs, start with the one that will be most impactful based on upcoming priorities in 2025.

If you do have some centralized tracking in place, think about both adoption and ease of use across your team. Is everyone updating trackers regularly? Are there duplicative docs that could stand to be combined? Is the information within them useful? Are key insights and timelines easily distilled? Now’s a great time to think about process improvements for the upcoming year.

If you missed it earlier this year, here’s the framework I always start with to build a testing tracker.

✨ One marketing thing: The best article I’ve found so far on the pretty scary changes coming to Meta for advertisers in the healthcare, financial services, and political categories.

✨ One fun thing: This mini-documentary (~23 mins) about what Japan teaches its kids from NYT will make you feel all the feels.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{Forward this email to someone who needs to test new messaging.}

Burning growth marketing questions? Let’s chat — book office hours here.

Why value propositions matter

Clear marketing messaging is one of the fundamentals you cannot pass go without. But if your product has many features and benefits, or has use cases for many different types of customers, it can be really challenging to nail down how to speak about it in a way that gets people to care.

If you’re spinning your wheels trying to solidify a clear marketing messaging strategy, a value proposition building exercise may be in order.

Value propositions are meant to distill all the innovative ways your product solves a problem into a few clear and punchy phrases, but they’re not a one-to-one swap for marketing copy. Think of value props as a high-level way to organize the benefits of your product, whereas ad, site, or email copy is much more specific.

A good value prop effortlessly communicates two things to a potential customer: “why you [the company]?” and “why me [the customer]?” 

The trick is that you need to have an understanding of who your ideal customers are and how to segment them, so you can speak to what really makes each group tick.

Fletch has a fabulous visual that walks through the value prop building process for B2B SaaS startups, so let’s break it down step by step, and translate it into a process you can use for B2C.

Fletch’s value prop builder. Credit: fletchpmm.com

Part 1: Customer profile definition

Identifying and defining your customer profile comes first. You could have two distinct customer profile segments, or you could have many, depending on the complexity of your product. It’s unlikely that you only have one.

Successfully running this value prop exercise requires an understanding of, or at least some educated guesses as to, your main customer profiles, as they’ll all have different needs and drivers.

This portion of the exercise corresponds with the “market segment” row of the Value Proposition Builder visual above. To translate this process for B2C, let’s take a fictional flavored electrolyte powder brand that has two distinct customer segments:

• Fitness junkies

• Average person who wants to make healthier choices

As this brand, you’d start by running through the value props exercise twice — once for each customer segment.

Let’s flesh out an example for the “healthy lifestyle changes” segment. Here are the questions you’ll need to answer:

• What: What are these healthy lifestyle seekers trying to do? Meaning, what problem are they trying to solve?

• How: How are they doing it? Meaning, what are these people doing currently to solve their problem, if they’re not aware of your product?

• Limitation: What limitations exist with their current approach? Meaning, what’s making their ultimate goal harder, less efficient, inconvenient, or too costly to achieve?

• Problem: What problem do these limitations create? Meaning, how do these limitations act as a blocker toward the goal defined in the “What” step?

Here’s how answering this first block of questions might look for the healthier lifestyle choices segment:

Part 2: Product and solution explanation

Once you’ve defined your customer segment and the problem they’re having, in our case, healthier lifestyle seekers who are trying to stay more hydrated, you get to introduce your product as the solution.

This corresponds with with “Product” row of the Value Proposition Builder. At this phase, the questions you need to answer are:

• Capability: What’s my product capable of doing? Meaning, what does my product do such that the the user segment’s limitation is removed?

• Feature: What specific feature of my product removes that limitation?

• Benefit: What’s the ultimate benefit to the customer of this feature?

You may discover there are multiple capabilities, features, or benefits worth mentioning. You’d then go through this exercise two or three times per customer segment, which would give you four to six value props total for two unique customer segments. This is how you unblock that messaging rut!

Here’s an example of how answering this second block of questions might look for our healthier lifestyle segment:

Putting it together

When you combine your customer profile definition with your product and solution explanation, the output is the ability to speak directly to a customer’s unique problems, needs, and desires.

Even if you’re incredibly early in your customer profile definition journey, I’d challenge you to segment out at least two distinct customer segments and run through this exercise. You’ll find unique ways of speaking to each segment, and demonstrate a true understanding of your customer, which is meaningfully more effective than trying to talk to all prospects with the same generic messaging.

This week’s edition of Growth Therapy is brought to you by AdQuick. If you’re ready to explore out of home ads, but need a quicker and easier solution than the traditional way, check them out at the link below 🔽

Run IRL ads as easily as PPC

AdQuick unlocks the benefits of Out Of Home (OOH) advertising in a way no one else has. Approaching the problem with eyes to performance, created for marketers with the engineering excellence you’ve come to expect for the internet.

Marketers agree OOH is one of the best ways for building brand awareness, reaching new customers, and reinforcing your brand message. It’s just been difficult to scale. But with AdQuick, you can plan, deploy and measure campaigns as easily as digital ads, making them a no-brainer to add to your team’s toolbox.

You can learn more at AdQuick.com

✨ One marketing thing: This is just your reminder that it’s officially Q4, it’s an election year, and holiday season is coming. Good luck out there!

✨ One fun thing: You’ve heard of Moo Deng, but have you been following Pesto the penguin?

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{Forward this email to someone with lots on their testing roadmap}

Burning growth marketing questions? Let’s chat — book office hours here.

Experimentation as part of your growth strategy is all fun and games (not to mention critical) until you end up with some really unclear results.

Maybe that landing page headline test you were SURE would increase conversion is looking flat to the control, or the new creative you cycled in that your team worked on for a week isn’t getting any traction. What the heck do you do next?!

Great marketers need to be data-driven, but they can’t be exclusively data driven, especially in a dilemma like this.

Here are three steps to help you triangulate how to proceed in a situation where your results aren’t crystal clear. If you’re super used to relying exclusively on data, each step will feel a bit more uncomfortable, but I promise these are all important skills to hone.

😀 The most comfortable: Use data first, and practice data honesty

Even if it turns out data isn’t the only way you get to an answer, it should be the first line of defense.

This means making sure your experiment is set up for success from an analysis perspective from the jump. You should have a hypothesis as to the metric you’re going to impact, and a pretty solid idea how you’re going to understand that impact, before you make changes.

Aside from knowing how you’re going to be able to determine success, you should also keep in mind the concept of data honesty: looking at the whole picture, without any bias.

It’s easy to want the tests we come up with to be winners, and to look at the data a certain way to try to confirm our hypotheses. This phenomenon tends to happen more frequently on smaller, scrappier marketing teams, where the same people devising the testing roadmap are working to execute tests and analyze results. Zooming out here is key and allows you to review the full impact of your results without inherent bias for wanting your hard work to pay off.

Make sure you can understand how your test impacts all the relevant points in your funnel, and that you can clearly convey the “so what,” so your first line of defense can be a clear reading of the data at your disposal.

🙂 Somewhat less comfortable: Look for directional signals

If you’ve started with the data-first approach above, but haven’t reached statistical significance or otherwise don’t have a clear result, giving your test more time is probably not the answer. The “right amount of time” for a test to run is entirely dependent on your budget and traffic levels, but if you’re totally unsure, aim for two weeks and see what volume looks like.

If two weeks, or whatever amount of time you’ve dedicated to your test, has garnered a good enough volume of data, but not a clear answer, look for directional signals in the data rather than definitive results. If you’re confident that you have enough data to normalize for any noise, and test results are fairly close to your control, more time isn’t going to get you closer to stat sig.

Instead, understanding what sorts of directional insights you can gather allows you to remain nimble. One example:

• You run a creative test meant to drive a CVR improvement

• The test has a much better CTR than the control, but a virtually flat CVR

• The test may still be directionally better than the control, presuming the CTR increase offsets costs, and you have other levers to impact CVR

Another simplified example, for the visual learners:

Super simplified landing page test data.

It looks like the test performed slightly worse, but these results are close, and not stat sig. Run your results through a stat sig calculator like this one to see whether or not you can rely on these results to hold. If not, you can say that directionally, there’s no lift from this test, and if the changes you made in your test version aren’t necessary, you can safely scrap them and move on.

Keep in mind that while you may not get a clear result with every test, you may not be taking big enough swings with your testing if none of your tests give you a definitive answer.

A good exercise to clear this “testing block,” which often happens when you’ve been so close to your product for a long time, is to:

• re-orient yourself with your customer personas, and

• revisit which marketing psychological principles are most likely to resonate with them.

Then, go back to the drawing board and design your tests to get back to to these psychological basics.

😬 Uncomfy, but so important: Don’t discount logic and intuition

Most growth marketers are so used to relying on data that “going with your gut” can feel like sacrilege. But once you’ve analyzed all the data and looked for directional insights without a clear answer, this is your final step.

Presuming you know your product, audience, and the dynamics of your business well, you should feel comfortable using logic and intuition to make decisions that lean more qualitative. Sometimes you just don’t have the data to support the results of a change that was made, and NOT making a decision on how to proceed is the worst possible outcome.

Let’s say you run an A/B test where you add some social proof to a landing page that doesn’t disrupt your hero or main CTA, and find that conversion rate of your test is slightly lower than your control. Logically, adding social proof where there was none should only help conversion. Perhaps one conclusion is there’s a more impactful way to convey this information. At the same time, you can also be confident in your own judgment that it’s unlikely that the addition of social proof is the reason for a slight CVR decrease, and what you’re noticing is more coincidental correlation than causation.

It’s also possible that while your top-of-funnel CVR declines slightly, your LTV could increase down the line, but it’s unrealistic to wait around to prove that data while your test continues running in the background.

When you get comfortable making decisions based on logic in the absence of clear data, not only will you be able to move more swiftly through your testing roadmap, but you’ll become a much more confident marketer.

✨ One marketing thing: Meta is no longer allowing detailed targeting exclusions, and will stop serving ad sets using these as of Jan 2025.

✨ One fun thing: I’m almost always in the “if I want dessert I’m going to eat a real dessert” camp, but this “healthier” chocolate mousse is changing all that…

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{Forward this email to someone whose paid search and paid social need to be more} 🤝

Burning growth marketing questions? Let’s chat — book office hours here.

Why Google + Meta is a 1 + 1 = 3 situation

You’ve probably heard the advice that consumer brands should start out by focusing their paid marketing on Google and Meta, before expanding to other channels. This approach allows you to do two key things:

1. Nail the fundamentals. These two channels have the most advanced ad-serving algorithms, and afford you the most scale — by far. For most consumer products, it doesn’t make sense to focus on other channels until you feel comfortable with your performance on these two.

2. Create a growth feedback loop. Meta is the channel that allows you to create demand by spreading your message to new audiences — the “push” channel — whereas Google helps you capture that demand by grabbing the folks actively looking — the “pull” channel.

More on nailing the fundamentals can be found at the top of this post, but for today, we’ll focus on the growth feedback loop portion.

A push and a pull

Meta

Meta is a “push” channel — it enables you to push your message out to as many eyeballs and potential customers as you’re willing to pay for. While granular targeting on the platform has been rendered virtually useless, the Meta algorithm has gotten incredibly smart at helping advertisers balance scale with getting in front of “the right” audience.

But there’s one big catch to the virtually unlimited scale you can achieve on Meta — none of the people seeing your ad have specifically raised their hand to hear from you.

You also might be catching them at a bad time. Marketers are probably the only people who open Facebook or Instagram to look for ads — everyone else is checking their notifications, scrolling through the latest updates, or looking for entertainment.

In this way, Meta ads are like the modern-day billboards of the advertising world, albeit way more sophisticated and with slightly better attribution. You can absolutely drive conversions directly from Meta (and you should always be setting up your campaigns to optimize for a desired action, rather than a goal like awareness), but making sure you have Google — your pull channel — set up properly will prevent leaky points in your funnel.

Google

Google, on the other hand, is a “pull” channel — it pulls in the people that are specifically looking for either your brand, or exactly what it is you offer, even if they don’t yet know your brand.

Before you think that sounds to good to be true, know that there are a few caveats:

For one, your targeted and blocked keywords need to be super buttoned up. Do not let Google auto-apply any recommendations, especially not the one to run all broad-match keywords, or you’ll waste a bunch of your budget on searches that just aren’t relevant to you.

Secondly, since Google is pulling in demand, that demand needs to come from somewhere. Either you’re in an industry where plenty of demand already exists and you’ll be facing competition and high costs, or there’s not much existing demand, but you can harness it from a channel like Meta.

(Side note: I don’t really believe in the concept of “demand generation.” Your product, and the way you tell the story around it, both need to be good enough to harness demand, but trying to generate demand where there is none is a losing game.)

Tying it all together

Running Google and Meta together is a situation where the outcome is greater than the sum of its parts.

Because of the massive reach you can achieve on Meta, you start to build awareness as a bonus byproduct of pushing out your message to new, but passive audiences. Not all of these people are going to remember your brand or product, and not all of the ones that do will be ready to learn more or buy.

But some will — and because you’re constantly increasing the pool of interested potential customers on Meta, you’ll have more search volume to pull in than if you were to rely on Google alone.

Rather than thinking of Meta as an exclusively upper-funnel channel and Google as an exclusively lower-funnel channel, think of the two as complementary parts of the non-linear user journey.

Ultimately, the best converters are the ones that raise their hand for your product, and the push-and-pull approach ensures you capture the largest number of hand raisers possible.

This week’s edition of Growth Therapy is brought to you by the 1440 newsletter. They truly take the bias out of news — and fun fact, they’re named after the year the printing press was invented! Check them out at the link below ⬇️

Seeking impartial news? Meet 1440.

Every day, 3.5 million readers turn to 1440 for their factual news. We sift through 100+ sources to bring you a complete summary of politics, global events, business, and culture, all in a brief 5-minute email. Enjoy an impartial news experience.

Join for free today!

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{Forward this email to a marketer who loves pizza but hates content planning}

Burning growth marketing questions? Let’s chat — book office hours here.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

🍕🍕🍕

If you’re not familiar with Gozney, they’re an outdoor pizza oven brand with an inspiring origin story, streamlined product mix, and brilliant content strategy.

Their founder, Tom Gozney, grew up in the UK, ended up addicted to drugs in his teens, and thankfully got into treatment to get sober in 2007. After treatment, he knew he needed to find hobbies that were healthier but still allowed him to be social, and threw himself into cooking. He eventually built a pizza oven in his backyard, and realized he was on to something.

It was such a hit that his friends started asking him to build pizza ovens for them, and the Gozney brand was born.

What are they selling, and who’s it for?

Gozney’s pizza ovens

Gozney’s core product is its pizza oven, which has four models. Their entry-level oven is $500, whereas the top-of-the-line oven costs $2,000. This isn’t an impulse purchase — their customer does plenty of research both across brands and between products before they make a buying decision (ask me how I know!).

Gozney describes their ovens online as “restaurant-grade outdoor pizza ovens for your backyard,” appealing to a well-defined audience — the advanced home cook with disposable income and at least enough patio space for something the size of a small grill.

They also sell accessories as add-ons, from utilitarian items like pizza cutters and pizza peels (aka the giant pizza spatulas you need to safely move a pizza in and out of an 800 degree oven) to cookbooks, co-branded aprons, and their own dough mixes.

Gozney’s Youtube channel

Gozney has a popular Youtube channel that plays a big role in its status as a beloved brand. They’ve mastered the art of a strong content strategy by:

• delivering value to viewers,

• solidifying their credibility, primarily through chef collaborations, and

• making you want their product, without them needing to sell it to you.

If your brand is considering going on a video content journey, read on to see how Gozney does it — and take note.

The stats

The top videos on the channel aren’t even about pizza!

As of this writing, Gozney’s Youtube has…

• 265k subscribers

• Almost 500 videos uploaded, going back 8 years — that’s an average of about a video a week

• Engagement rates between 2% - 4% on the latest content, at or above YouTube’s average

• Views over 1 million on their top 5 videos — none of which are even specific to pizza!

The strategy

There are two notable elements about Gozney’s Youtube strategy that make it a winner:

1. While they sell a product that could be for anyone — meaning it’s not limited to a demographic or geographic subset, or a customer that must meet certain conditions — they don’t try to speak to everyone.

   Even with unique content buckets, it’s clear they know their audience, and are speaking to the advanced, into-food, home chef.

2. There’s very little self-promotional “here’s how our pizza oven works” or “why our pizza ovens are the best” content. There are a small handful of videos like this that are meant to help you compare models or understand the technology, but these tend to have lower views.

   Content is focused on actually utilizing the product and its versatility in a fun and attainable way, without being salesy.

The content

Content on the channel varies in length and style, but is all high production value. They have some content specific to their entry-level pizza oven, but tend to feature their highest-end model, creating a subtle-but-smart upsell by showing you just how good their best product is.

There are a few key content pillars, organized into clear playlists:

• GozneyTV: 10-15 minute videos, featuring mini-series with different food personalities that, in their words, “bring to life people, places, and inspiring food.”

• Pizza Recipes: How-to content featuring all kinds of recipes from all kinds of people. There’s everything from classic margherita pizza with authority Chris Bianco, to frutti di mare pizza with Bangkok-based Instagram food personality Feng Chen.

• Outdoor Recipes: More recipe content that shows the versatility of Gozney ovens, featuring everything from camping-style grilling, to scratch-made bagels, to whole fish.

• Cook Different: These videos highlight chefs using Gozney ovens to make their own creative recipes.

• Pizza for Beginners: Short, 2-3 minute, tactical content. Videos here are high production quality like the rest of the channel, but are focused on cut footage and on-screen text rather than people, making production of this type of content a much lower lift.

A clear theme across most content on the channel is consistent collaboration with well-known chefs and food personalities. Folks like Matty Matheson of The Bear, Frank Pinello of Best Pizza in Williamsburg (highly recommend if you’re in the area), Thomas Straker of social media food fame, Brad Leone of Bon Appetit, and Chris Bianco of Pizzeria Bianco and Bianco diNapoli tomatoes all make regular appearances.

Gozney is able to build deep trust and credibility with viewers with this approach — after all, respected folks in the food world, and specifically, the pizza world, enjoy these pizza ovens enough that they’re willing to regularly collaborate.

It’s worth noting that these are not A-list, Hollywood celebs, and they don’t have to be. They’re just personalities that Gozney knows their audience will get really excited to see.

The TL;DR

Gozney’s content strategy is ultimately pretty simple:

• They show their product being utilized in a fun and aspirational, yet attainable way.

• They build deep trust in their brand by focusing on featuring recognizable personalities.

• And the common thread through all their content is that they don’t push their products on you, but you’re left really wanting to experience them for yourself.

✨ One marketing thing: The Olympics may be over, but the advertiser drama is not. (Semi-relatedly, how do we think laid-off employees feel when they see their ex-employer as a major Olympic sponsor? 😶)

✨ One fun thing: Here’s a pizza pic from my own backyard pizza oven experimentation to close us out. This dough recipe wasn’t my favorite, but I’ve since tweaked it:

Red pie with pepperoni, ricotta, and hot honey 💯

Questions? Comments? Topic requests? Just hit reply ↩

_{Forward this email to someone who’s tracking a whole mess of marketing initiatives}

Burning growth marketing questions? Let’s chat — book office hours here.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Track the wrong metrics and some are bound to end up in the KPI Graveyard. Image credit: Corporater.com

Brand marketing vs. growth marketing

There’s tons written about the differences between brand and growth marketing, and why both are important.

Put simply, brand marketing is how you effectively communicate the heart, soul, and personality of your brand, and growth marketing is how you effectively capture and retain revenue. Neither can succeed without the other.

But these definitions are abstract and apply to every brand differently, and sometimes marketing teams just need to clearly understand how to decide what makes their brand and growth marketing “effective,” beyond just “awareness” or “revenue.”

Here are some KPIs that will help you make sense of performance. Not all of them will be relevant to your brand, and that’s ok — focusing on a small handful of the right KPIs is just as important as identifying the wrong ones.

Brand marketing KPIs

Many brand marketing initiatives are simply less measurable than growth marketing initiatives, particularly in real time — but that doesn’t mean NOT measurable. As tempting as it may be to measure brand on vibes alone, there are a bunch of ways to validate your efforts are paying off, especially when tracking these metrics over time.

Aided and unaided awareness:

• Aided awareness measures the number of consumers that recognize a brand, when prompted with its name or logo (Ex: Which of the following brands have you heard of?)

• Unaided awareness measures how many consumers can recall a brand without prompting (Ex: Which brands do you think of when you need xyz?)

• Running these kinds of awareness studies usually requires a third-party tool

How to use it: Track these metrics over time to understand how awareness is growing as your brand marketing efforts increase in scale, or to assess lift after a particularly big brand push.

Organic search volume:

• The volume of searches your brand terms get, as measured by your search ads accounts and / or your SEO tools

How to use it: Track volume over time to understand things like seasonality, growing brand awareness, and any spikes that can be tied back to your brand marketing efforts.

Increasing quality of new customer acquisition:

• Measured by tracking contribution margins, you can make sure the quality of new customers entering the fold continues to increase over time

How to use it: Tie brand efforts and messaging back to steady margin improvements, and identify which brand efforts are the most valuable and worth continuing to focus on.

Increasing quality of repeat behavior:

• Measured by tracking LTV, you can make sure quality of retained customers increases over time

How to use it: Tie brand efforts and messaging back to steady LTV improvements, and identify which brand efforts are the most valuable and worth continuing to focus on.

Organic social benchmarks:

• Follower growth rate on your accounts

• Content engagement rates (like likes, comments, shares)

• Content performance grouped by theme or bucket

How to use it: Make sense of the less tangible on social and use growth in these metrics over time to validate what’s resonating with your audience.

Social listening:

• The monitoring and analyzing of online conversations and mentions of a brand and its industry to understand sentiment and market penetration

• Running social listening usually requires a third-party tool

How to use it: Capitalize on trends, sentiment, and conversation about your brand and competitors.

NPS score or similar:

• A quantitative metric that scores sentiment about your brand by splitting customers into promoters and detractors

How to use it: Track how customers, both in aggregate and within cohorts, are feeling about your brand over time. You can run all kinds of analyses on this data, like what promoters have in common vs. what detractors have in common, and optimize from there

Growth marketing KPIs

Data and growth marketing initiatives go hand in hand 🤝, so there’s a bias to try to measure EVERYTHING you possibly can. But too much data can be just as confusing as too little data, so focus on the KPIs that are relevant and enlightening for your business. Here are a bunch to choose from:

LTV/CAC:

• The ratio of the lifetime value (LTV) of your customer to the cost of acquiring them (CAC)

• There are differing opinions on what a sustainable LTV/CAC should be, but around 3:1 is a solid benchmark to indicate a healthy balance between growth and profitability

How to use it: Understand the efficiency of your acquisition and retention efforts.

Full funnel conversion rates:

• The conversion rates of each step in your funnel, beyond a conversion from an ad, particularly if your funnel is long or complex

How to use it: Full funnel conversion rates help you understand quality and CAC, beyond something like a landing page or a form conversion. While your growth efforts may not have a huge impact on later-funnel steps, particularly for long funnels, your CAC will still be impacted (it’s a team effort).

Qualified conversions and growth rate:

• The number of good conversions you’re driving

• The rate at which these good conversions increase over time

How to use it: What “qualified” means will be different for every brand, but this is another way of looking at full funnel conversion rates. Simply, growth IS the rate and volume of your qualified conversions increasing.

Activation rate:

• Another way of looking at quality, this is the percentage of your users that activate — typically into a paid tier after something like a free trial

How to use it: Tracking the portion of your users that are willing to pay for your service will help you understand how well you’re delivering on expectations with your marketing.

Landing page CVR:

• The percentage of folks that take a desired action after landing on your landing page

How to use it: Whether or not you’re running a robust landing page testing program, tracking your landing page CVR allows you to understand what messaging and visuals resonate with your users.

Directional CPA:

• The CPA of different ads or campaigns relative to other ads or campaigns

How to use it: While relying on in-platform metrics alone will only ever give you a small piece of the big picture, looking at your CPAs directionally can tell you what ad concepts resonate and drive conversions most efficiently. Then you can double down on your creative testing.

Revenue per session or per user: 

• Backing out total revenue across a total number of sessions in a time period or users in a cohort

How to use it: Typically used to compare the short-term dollar value of different user experiences or user cohorts, but is more limited than a longer-tail revenue metric like LTV.

Retention rates or churn: 

• Retention is the percentage of customers that stick around after an initial conversion

• Alternatively, churn is the inverse — the percentage of customers that don’t make it to an ultimate step after beginning their journey

How to use it: No growth metric gives you a full picture of performance if you don’t take retention into account. Retention rates or churn rates are a good diagnostic to ensure the customers you’re acquiring — especially if you’re paying for them — stick around.

This week’s edition of Growth Therapy is brought to you by AdQuick. They’ve made buying and tracking out of home inventory as easy as running Meta ads. Check them out at the link below ⬇️

Modernize Out Of Home with AdQuick

AdQuick unlocks the benefits of Out Of Home (OOH) advertising in a way no one else has. Approaching the problem with eyes to performance, created for marketers and creatives with the engineering excellence you’ve come to expect for the internet.

You can learn more at www.AdQuick.com

✨ One marketing thing: The immediate success of Kamala Harris’ campaign with Gen Z, in stark contrast with Joe Biden’s failure to galvanize Gen Z, shows exactly why you can’t try to force virality.

✨ One fun thing: People are going bananas (or carrots?) over this new Trader Joe’s product…I have yet to see it, but I’m on the hunt.

📣 This week’s shoutouts 📣

This week’s shoutout goes to Noah Stambovsky, Creator of Touchpoints Newsletter!

Want to get your name & business in front of the Growth Therapy audience? Share the newsletter 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{Forward this email to someone with a lofty goal to hit this quarter}

Burning growth marketing questions? Let’s chat — book office hours here.

The OKR → Hypothesis → Initiative framework

OKR setting, growth planning, quarterly prioritization….irksome but critical work that even the best marketers dread.

Without the right approach to figuring out how the hell you’re going to increase LTV by 15% or reduce churn by 30%, these planning exercises are impossible at worst, and overwhelming at best.

Often times, we’re guilty of trying to approach planning and prioritization bottoms-up, listing out all the things we could do this quarter that seem like good ideas, or that have been sitting on our to-do list for a while.

There’s a simple framework that flips this process on its head by taking a top-down approach instead. Approaching planning this way prevents you from spinning your wheels, and forces you to identify tactics that are actually impactful in hitting your goals. Here’s how it works.

Step 1: Start with your OKRs

This step is intuitive — you’re ultimately going to be mapping tactics to each OKR, so pick one to start with. Let’s use this example:

Of course, all OKRs should have an objective — “increase the number of customers served” — and a key result — “improving our conversion rate from 5% to 7%.” Typically, you’ll have a few key results associated with each objective, but we’ll keep it simple for this example.

The OKR portion of this framework describes your business problem. Delineating business problems from user problems is important, because while your job is to solve both, your ability to do so relies on the understanding that your users will only ever care about their problems.

Think about the last time you walked into Whole Foods — did you beeline for the ice cream aisle or spend time worrying about how they’re trying to increase same-store sales growth?

Understanding user problems comes in next.

Step 2: Build your hypotheses

Hypothesis building is the meat of the process, and is the part that takes the most getting used to if you typically plan bottoms-up. But it’s worth getting comfortable with — it’ll force you to think differently about how to solve problems.

The goal of this step is to come up with reasonably evidence-based theories that explain existing user behavior. The “user” part is critical here — too often we make the mistake of jumping right to solutions to solve business problems, without understanding the user’s problems.

Our hypotheses are our opportunity to dive into the user’s problems, and thus should be theories about how our users are currently experiencing our business.

If you think about it, the only thing you really CAN hypothesize about is user behavior. You KNOW your business, how it works, what information you have on your landing pages, what the product experience is like, etc. Here’s how to put this into practice and write a good hypothesis.

How to write a good hypothesis

Some examples of the below.

A good hypothesis meets a handful of criteria:

It SHOULD be user-first.

In practice, this means your hypothesis should generally start something like…”We believe our users are confused by our…” rather than “We believe we don’t currently do a good job of…” This is because hypotheses are your opportunity to explain user behavior, not business problems.

It SHOULD theorize about user behavior and psychology.

In practice, this means your hypothesis might sound something like “We believe potential customers don’t understand the outcomes they can expect from using our product, and therefore are hesitant to convert.”

It SHOULDN’T predict an outcome.

In practice, this means your hypothesis shouldn’t include something like “We believe we can improve our conversion rate from 5% to 7% by…” Instead of predicting an outcome within your hypothesis, you’ll actually measure the outcome of your initiative by assigning and tracking a singular KPI.

It SHOULDN’T include the tactic or experiment itself.

In practice, this means your hypothesis shouldn’t read something like “We believe that by including more testimonials on our landing page, we can improve conversion.” This would be putting your tactic first — BEFORE you’ve defined the problem the user is facing that would warrant this tactic.

It’s worth noting that each one of your OKRs can, and probably should, have a handful of hypotheses associated with it — maybe two or three. Once you have your OKRs, which represent your business problem, and your hypotheses about how your users’ problems contribute to your business problems, then, and only then, can you start thinking about how you’re going to tackle your OKRs.

Step 3: Determine the right tactics to march toward your goals

Determining initiatives and specific tactics is the step that folks often jump to right after reviewing their OKRs.

But until you spend time on that critical middle step of trying to understand how your customers are behaving and why, you’re attempting to come up with solutions to problems you haven’t defined.

Let’s take our OKR…

…and one of our strong hypotheses…

“We believe our customers don’t understand the outcomes they can expect from using our product, and therefore are hesitant to convert.”

…and talk about how to determine the best tactics.

Now that we know what we want to achieve — a measurable conversion rate increase — and have a solid guess as to why we’re not achieving that metric today — customers aren’t clear on outcomes — it’s MUCH easier to come up with a whole host of ways to try to close this gap.

Some ideas are:

• Include engaging customer testimonial content on product pages

• Ensure clear, desirable outcomes are prominent features on product pages

• Introduce before and after visuals that demonstrate outcomes

But without the focus of attempting to improve user understanding of outcomes, the list could be all over the map and you’ll be attempting to boil the ocean.

Other notes about your tactics

• Each tactic should have one measurable KPI that’s relevant to its OKR

• You should be able to track this metric easily

• As often as possible, tactics should be metric-based, rather than binary (e.g., “how much did we move our KPI”, instead of “did we or did we not get this thing done”)

Test yourself

Write down a few of the projects you want to prioritize before the end of Q4.

Now, consider one of your relevant OKRs. Spend a few minutes thinking through a strong, user-focused hypothesis that’s relevant to that OKR.

Then come up with a few tactics that will clearly help you solve both the user problem you’ve defined, and moves you toward the key result you need to achieve.

That list of tactics will probably be different, and more focused, than the list of projects you started with.

✨ One marketing thing: Non-alcoholic beer brand Athletic Brewing just raised another $50m to expand operations. I say this is a marketing thing because they positioned themselves differently than non-alcoholic incumbents when they launched, and their success speaks for itself.

✨ One fun thing: A clever summertime reminder.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{Forward this email to someone who’s trying to optimize their funnel}

☀️

Burning growth marketing questions? Let’s chat — book office hours here.

Marketers often default to wanting to reduce friction in their funnels as much as possible. We remove steps in our signup process, obsess over add-to-cart and checkout rates, and create lead magnets with big, bold signup buttons to drive as many leads as we can.

But, the goal isn’t really to drive as many leads as we can. It’s to drive as many good, high-intent, high-value leads as we can. Friction should be thought of as your friend, not your enemy, in accomplishing this.

Caveat: If your product is something that could be considered an impulse purchase, then you should absolutely focus on removing any sources of friction. Make it as easy as possible for consumers to take that desired action.

But in virtually all other cases, using strategically-placed friction to build intent and quality will help you activate, not just cultivate, users. Even still, some types friction should be avoided altogether, so let’s dive in to how and when to best use friction to your advantage.

Three types of friction

Not all friction is created equal. While some types of friction can be used strategically, others should be avoided entirely. Here’s how to tell them apart.

Misplaced friction

Misplaced friction should always be avoided. This is friction that prevents your user from taking your desired action, without building intent or quality along the way.

Think about scrolling Instagram and seeing an ad for a chic AND comfortable looking pair of summer sandals — the jackpot. You weren’t browsing for shoes in the moment, but you want to see what these cost and whether the reviews are good. Well, you clicked the ad and got dropped on a collections page of hundreds of shoes. There’s no way you’re going to spend time sifting through this whole mess to find the sandals from the ad and you abandon ship. This is misplaced friction.

Other common examples are broken experiences like dead links or 2FA sign-in where codes never get delivered, or disjointed and unexpected experiences, like unclear and confusing UX, or being prompted to give a credit card number before you know enough about what you’re buying.

You’re probably too close to your site and user experiences to properly audit them for misplaced friction. Instead, have a friend who’s as close to your ICP as possible review your different user journeys, and ask them to note places that something unexpected or frustrating happened in the process.

Positive friction

Positive friction, which encourages users to take the desired action, can be strategically used to your benefit when done right. Positive friction allows you to do a few things:

Build intent through the user journey

When I was leading growth marketing at SmartAsset, a $1B fintech startup that matches people with financial advisors, I saw firsthand how a funnel full of “friction” actually solved for intent and quality.

We dropped users into a long questionnaire that asked them all about their personal finances, net worth, retirement plans, and personal information, like where they were located and their contact info. At the time it was somewhere close to 30 questions, which might sound crazy, but we learned there were two phenomena at play here:

First, the friction we were creating with such a long lead gen flow actually did a great job at weeding out low-intent users that never wanted to work with a financial advisor in the first place.

Second, the sunk cost fallacy was strongly at play, meaning folks that had invested the time to answer all these questions didn’t want to abandon all their progress partway through. Of course, good UX is necessary to avoid any misplaced friction in an experience like this.

Lead the user to your desired action in an optimal way

IKEA is the classic example of using positive friction to nudge users toward a desired action — a big purchase. Stores are designed with friction in mind — a one-way maze across an average 300,000 square feet of furniture, lighting, homewares, and even entire kitchens. But they make it easy and fun, with engaging displays, arrows to point you in the right direction, wide aisles that are easy to maneuver, and exit points if you need.

Of course, this design is very intentionally thought out. By introducing friction that makes your shopping trip more of an exciting journey than a chore, IKEA achieves a larger basket size and more time spent in store from its shoppers.

Take advantage of user expectations at each stage in your journey

Cookie consent banners are a good digital example of using expectations to your advantage. While the first sites to put these up for compliance reasons probably dreaded doing so, consumers are now so used to simply clicking “accept” to get these annoying banners out of the way that there’s no misplaced friction here.

Sure, it’s one extra step, but most users are going to take the one-click, easy way out, rather than customizing their cookie preferences one by one on each site they visit. Of course there’s good intent behind giving consumers control over their data, but by making the desired outcome the easy choice, brands get to maintain the ability to anonymously track their users.

Put this into practice by browsing around on a bunch of different sites, and observe where your expectations of how to get from step to step are met or not. Was a button or login experience placed somewhere you didn’t initially expect? Was it totally seamless to access what you needed to? Each incremental step in a user journey is inherently a friction point, but some are so expected that there's not going to be any drop off.

Preventative friction

Preventative friction discourages an undesirable behavior, and should be used sparingly.

The most ubiquitous example of preventative friction is the experience you encounter when you try to cancel a subscription. Subscribers are often a brand’s most valuable users, because they drive repeatable revenue and a high LTV. A subscriber cancelling their subscription is therefore a highly undesirable behavior. Most brands will take advantage of this, and put a slew of steps in front of you before they actually let you cancel.

The most egregious example of this that still sticks with me to this day is when I tried to cancel a clothing subscription I’d purchased unknowingly, when I was fresh out of college and making virtually no money. I thought I was just buying clothes online, but apparently I’d missed some fine print where I was signing up for a mystery box to be delivered to me each month. A recurring charge like this obviously didn’t fit into my tight budget, so I went to cancel, and ultimately had to take and pass a quiz on this brand’s cancellation policies in order for my cancellation to be processed. Don’t be this brand.

When considering adding preventative friction to your user experience, give it something I like to call the “eye roll test.” Consumers are savvier than ever and can tell when they’re being bamboozled. If your typical user is going to roll their eyes at your preventative friction, or, worst case, get legitimately angry, leave it out. It’s not worth it.

The TL;DR

Friction is not the enemy! 

While misplaced friction that prevents users from achieving desired outcomes should always be avoided, positive friction and preventative friction can be used to your advantage for any product that’s not an impulse purchase. Positive friction can help you build intent and drive a qualified, activated user base, where preventative friction can reduce unwanted behaviors.

This week’s edition of Growth Therapy is brought to you by AdQuick. They make browsing, buying, and measuring performance of out-of-home inventory as easy as running Meta ads. If you’re ready to dabble in OOH, check them out at the link below ⬇️

Run ads IRL with AdQuick

With AdQuick, you can now easily plan, deploy and measure campaigns just as easily as digital ads, making them a no-brainer to add to your team’s toolbox.

You can learn more at www.AdQuick.com

✨ One marketing thing: The latest on the TikTok ban.

✨ One fun thing: The /FindTheSniper subreddit has become one of my favorite internet corners recently.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{Forward this email to someone who spends most of their day refreshing Shopify.}

Up first, special announcement — introducing ✨office hours!✨ 

A few times a month, I’ll open up a handful of 15-minute “brain-picking” slots. They’re absolutely free, so if you have some marketing questions you want to chat through, or want to see if it could be a fit to work together, you can grab some time with me here.

There used to be a clear-cut ecom playbook. At its most reductive, it read something like “Find manufacturer, stand up Shopify storefront, spend a ton on Facebook ads, print money.”

While it’s no longer this easy, the principles hold. Find a manufacturer once you’ve identified a quality product that solves a real pain point, set up your Shopify storefront with UX and brand principles in mind, and acquire customers responsibly — and retain them.

Truthfully, I’ve learned a lot of this the way most of us learn important lessons — the hard way. But what are our mistakes if we can’t grow from them?

Here are the 9 biggest mistakes I’ve made running marketing in ecom, and what to do instead.

Focusing on your products instead of your offer

A few years ago, I was brought in as the first in-house marketing lead at a women’s health product startup. I was excited about the mission of women helping women to easily and discreetly bring healthcare to your door, for products that otherwise might be embarrassing to buy.

This peace of mind should have been the core offer.

Instead, I focused too much on marketing the individual products that made up this offer, talking about their features and benefits, and why people should buy them. I learned quickly that when acquisition marketing that runs on limited budgets, and focuses on too many individual products, it’ll never be profitable.

The problem is, most potential customers are NOT ready to buy at any given point. This is even more true with healthcare products that are only needed on occasion. But if your marketing is offer-focused (creating and communicating value) rather than product-focused (buy buy buy!), you can gain a potential customer’s trust.

A clear, core offer — in this case, being the place to go for peace of mind when it comes to women’s health — reinforces your brand, so customers come right to you when they’re ready.

Forcing subscriptions instead of offering trials

At the women’s health startup, there were a handful of products that made sense as subscriptions, like supplements, and others that did not, like UTI tests. Slightly discounted subscriptions were offered for every product regardless, and when the time came to figure out how to increase our subscriber base (gotta love that LTV), my bright idea was to make our supplement products only available via a subscription.

Well, subscriptions certainly increased. Then cancellations followed. Turns out, no one likes being forced into subscribing to a physical good before they can try it out (whereas of course a subscription model is core to services, software, and media).

Particularly for products that claim to drive some sort of improvement, why wouldn’t you let someone trial the product before locking them into a subscription? If the product doesn’t work as advertised, then you have bigger problems to solve.

And a cancelled subscriber is way worse than a one-time purchaser — the cancelled subscriber is never coming back. You can still heavily incentivize subscriptions where it makes sense, but giving customers a choice will ensure those that subscribe stick around for the long haul.

Adding products that seem similar, but are for distinct user groups

Understanding your users, what they want, and the utility and value they’re getting out of your product is paramount before you start adding new products and marketing them.

Expanding your product line is often a great way to increase LTV and bring in new customers, but there’s a fine line to walk. When we added new products for distinct health conditions to the existing line, my mistake was to try to market new products to the user we already understood — here’s why that doesn’t work.

“Women’s healthcare” is an extensively broad category. Even within the same product — like pregnancy tests, one of this brand’s core products — there are distinct user groups purchasing for totally different reasons (planning pregnancy vs. avoiding pregnancy). Speaking the same way to everyone who purchases this product, let alone different products, is a recipe for failure.

This is the same reason that if you think your total addressable market is “everyone”, just because anyone could in theory benefit from your product, you’re playing a losing game from the start.

Working with the wrong 3PL

3PL selection doesn’t typically fall under marketing’s purview, but if you have consistent pick-and-pack and delivery issues, you can kiss retaining customers goodbye.

If you sell a physical product, the 3PL you work with has massive implications for your business’ success and your brand’s perception. Customers who receive a damaged product in a torn up box, or receive their order later than anticipated, don’t care if this is the packer or carrier’s fault. These experiences reflect poorly on your brand, regardless who’s responsible at which step.

I’ve seen brands work with everyone from well known-global 3PLs to boutique, white-glove 3PLs. Logistics will never be perfect, but interviewing multiple 3PLs to understand both price and level and service is critical.

Customer complaints I’ve seen range from shipping product in other brands’ packaging (🤯), packages arriving waterlogged, dirty, and damaged, packing inventory in incorrect boxes such that merch gets jostled around in transit, you name it. The best thing you can do is figure out who’s going to treat your products as carefully as you would.

Not paying attention to margins

Margins are just as much of a marketing problem as they are an ops and logistics problem.

Your margins are critical to pricing your products, but they’re just as important in how you market them. Your business probably has an aggregate margin you need to remain above, but margin on each individual product likely differs.

If marketing costs start shifting toward moving lower-margin products, it’s going to be hard to maintain that healthy aggregate. I’ve made the mistake of not building this view into my regular marketing tracking docs, and not realizing until much later that last month’s spend was not as profitable as it seemed.

Understand the margins on your products, and use this knowledge to your advantage when budgeting.

Not having a cohesive content and promo strategy

Consumers are savvier than ever, and yet we still think we can get away with content and promotions calendars centered around made-up holidays (today is purportedly National Peanut Butter Cookie Day, and Pride apparently shares June with National Turkey Lovers Month and National Headache and Migraine Awareness Month).

Even if there’s a made up holiday that seems relevant to your brand, customers see right through it when you email them and say “it’s national x month, so we’re doing a sale!” Sure, maybe you’ll incentivize them to buy — everyone loves a deal. But your promos should be a part of building your brand, not cheapening it.

I’ve learned how important it is to build your content calendar from a place of authenticity, and to use promos sparingly so as not to train your customers to wait for them.

Ignoring the power of social proof

If you go to your site right now and can’t pinpoint at least 3 examples of social proof — like reviews, awards, press logos — within a few seconds, stop what you’re doing and figure out how to get it.

Everyone constantly talks about social proof, because it really is that psychologically powerful. Most consumers know that brands have total control over the reviews they show on their site and the press logos they choose to share, but they still work.

When our shopping options are virtually endless, we need to see that someone else has found this product valuable, or that the brand is associated with a publication we trust, before we hand over our credit card number.

If you’re struggling with garnering social proof, take a step back and ensure you’ve started with an excellent customer experience layered on top of an excellent product. I’ve found that if you try to incentivize reviews without these things, those reviews will be lukewarm.

Not getting buy-in on conversion rate optimization

Conversion rate optimization, or CRO, can get a bad rap. Sometimes founders aren’t convinced that some tweaks to the user experience can make a massive difference in performance. And UX should really be at the forefront of CRO — the intention is never to deceive customers into making a purchase, it’s to make their journey as frictionless as possible (note: friction is a good thing in intent-based conversion paths, but not in ecom).

I’ve learned that focusing on the human element is one way to get buy-in to opening up resources for CRO. Another is to demonstrate how CRO is one lever in your efficiency equation — and it’s often one of the last to get pulled.

If your messaging is dialed in, your paid channels are optimized, your post-purchase experience is top notch, and you haven’t challenged your assumptions about your site experience, you’re probably sitting on a goldmine.

Not understanding that alternative forms of problem solving are your real competitors

I’ve fallen into the trap of beginning to work with a brand and immediately looking to see what all their direct competitors are doing. It seems logical — especially in ecom, where for almost every brand, there are a handful of virtually indistinguishable copycats.

While keeping a pulse on what competitors are doing is worthwhile to a point, arguably more important is to stay on top of the alternative ways someone could tackle the problem your product exists to solve.

Selling DTC women’s healthcare products? A potential customer could simply go to the pharmacy if they need something ASAP, or to their doctor if they have a more pressing concern.

Selling soda that claims to boost gut health? A potential customer could simply add probiotics to their vitamin regimen, or consider making changes to their diet if they have real health concerns.

Selling high end fitness equipment? A potential customer may prefer to spend less cash at a time for a gym or studio membership, where they also get some social interaction.

You could argue, perhaps correctly, that folks seeking out these alternatives are not “your customer,” but being honest about what you’re up against can only help you build stronger positioning.

✨ One marketing thing: Brilliant ad to announce a new product launch, particularly for anyone who loves both Succession and peanut butter (it’s me, I’m the target audience)

✨ One fun thing: I went to see the band Goose this past weekend (they’re from my hometown!) and they played this cover. The original is one of those songs that you know, but don’t know why you know it, and then are pleasantly surprised to hear at a jam band’s show.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

_{Forward this email to someone with a lean marketing budget.}

I made you a nifty flowchart!

The brand search argument

If you look, you’ll find countless internet debates about whether or not it makes sense to spend money on brand search. You’ll surely find an answer that confirms your existing opinion (see: internet debates), but I’m here to tell you there are 4 distinct scenarios in which it makes sense to spend on brand search.

Otherwise, you’re probably wasting money that could be better utilized elsewhere in your marketing budget.

A lot of the arguments you hear in favor of spending on brand search seem to make sense. Typically, you’ll hear things like:

“Search is a high-intent channel that harvests demand (as opposed to social, which creates demand), and folks searching for your brand are the HIGHEST intent!”

• My counterargument: All 100% true — but this still isn’t a strong argument for paid search. If you show up at the top of the page in organic listings, you’re still going to harvest that high-intent demand.

“You have to play defense! If you don’t bid on your own brand name, a competitor will!”

• My counterargument: Also entirely possible. But maybe you spend 30 seconds each week checking to see if this is happening, before you start spending money on brand search.

“You want to be able to control the narrative! If you have bad press or reviews, brand search ensures you decide what people see about your brand!”

• My counterargument: If this is the case, a search ad is the flimsiest of bandaid fixes. Address the root problem here and you’ll be a hundred times better off.

Instead, here are the 4 times it does make sense to spend money on brand search.

If your competitors are

The “playing defense” argument IS a sound one, if you need to. Google your brand’s name right now. Is a competitor’s ad showing up first? That means they’re bidding on your brand name, and it would make sense for you to come in and “defend” that turf by bidding on your own brand name.

As soon as you start doing this, your competitor’s ads should all but disappear from the top slot. Google will see that your brand name is significantly more relevant to your site than your competitors’.

If competitors aren’t bidding on your brand name, then you really don’t have to, either. But just because a competitor isn’t bidding on your name today doesn’t mean they won’t get the bright idea to do so next month. Like I said above, this could be something you check once a week (which is free and virtually zero effort) so you can jump in when needed.

If your brand name is generic

Maybe your brand name is a made-up word or phrase that’s unique to your brand alone, like “Oura” or “Misfits Market”. Someone searching for these things could only be looking for information about your brand.

Or maybe it’s a regular word or phrase, like “Mode” (the analytics platform) or “Calm” (the meditation app).

In cases like this, you want to be sure that folks that are actually searching for your company can find it easily, at the top of the page, without needing to scroll through a bunch of irrelevant results.

You might be thinking, “What if I’m Mode, the analytics company, but someone is googling ‘mode’ for help on their math homework?” Well, that just means you might have a lower click-through rate on your brand search ads than you’d otherwise expect, because the folks that aren’t looking for you aren’t going to click your ad.

CTR is mostly a vanity metric you needn’t worry too much about, though it is an ad quality signal, and therefore can impact how much you pay per click. If you’re focused on improving your CTR, it should be for ad quality and conversion quality purposes.

Unless you’re in a situation where your brand name is highly generic and there are multiple companies with the same name, your brand CTR should still be the highest in your account by far.

If you need to geographically target

Some of the startups I work with started out by operating in just a few markets, and then expand their footprints as they scale.

If you’re not operating on a national scale, bidding on brand search in your markets of operation can be a helpful way to ensure you’re showing up in search results within the markets you serve. SEO alone may not be enough to surface your brand in the markets you want when you’re new or small.

Bidding on brand search immediately as you expand into new markets can also help you measure awareness of your brand in new markets over time, if you keep good track of search volume.

If you want another easy way to measure impact from other initiatives

Assessing the impact of other marketing initiatives on brand search is a smart way to measure the less measurable — think awareness and brand plays like out of home, audio, TV and the like.

To do this effectively, you do need to have a solid baseline of brand search volume and performance as a basis for comparison.

Take this example:

I work with a healthtech client that has a detailed listing about their services on a third-party directory. Clients who come in after viewing this listing should be much higher intent than those that come in cold, so we’ve tested sending some traffic to this third-party page.

While there’s tons of rich explainer content on this page, the page is not at all conversion-optimized. Therefore, our expectation is not necessarily to see many direct conversions from the traffic we send to the third-party page, but perhaps a correlated increase in brand search volume during the time this test is live.

We’re running brand search already, because we fall into the generic name bucket, so establishing that baseline won’t be a problem. If none of the prior three situations apply to you, but you think you may want to use brand search to measure impact from other initiatives in the future, think about how much baseline data you’ll need to be comfortable using it to make decisions.

The bottom line

If you have a healthy marketing budget, maybe you’ve decided you want brand search to be always-on, regardless of the cost. But if your budget is limited, or if you just want to get smarter about how you’re spending, brand search may be an easy line item to cut.

Ultimately, the best test is to pause brand search for a week or two to understand how incremental it really is. Were these folks going to find you anyway? And if so, do you need to keep paying for them?

This week’s edition of Growth Therapy is brought to you by Stori AI. They’ve won a bunch of Product Hunt awards by building a sweet AI-powered social content generation tool that ensures your brand look and tone is consistent, and even nudges you when you’ve been posting too much promotional content (❗️ We’ve all been there). Check them out at the link below ⬇️

We needed STORI. So does your startup.

Like many startups we had a problem - need for consistent content marketing and social media presence. Then we thought - what if AI could help us? The rest is STORI. It creates, edits, publishes and schedules content within minutes.

See it yourself.

✨ One marketing thing: A great article came out in the NYT this week about the decline of made-for-Instagram cookware. I’ve said it before and I’ll say it again — you can’t market your way out of a bad product. (If you’re actually interested in cookware that could outlast you on this earth, All-Clad, Le Creuset, and even newcomer Made In make pieces worth investing in).

✨ One fun thing: This week’s newsletter was written with a view in Jackson, WY 🤠 

Not a bad place to sit and type

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

Credit: marketoonist.com

There comes a time in every startup’s lifecycle when it feels like something just isn’t working. This is normal — presumably there are quite a few things you’re doing for the very first time.

What separates the successful startups from the rest is the ability to identify these problems ego-free, swiftly prioritize them, and solve or squash them.

If marketing is what’s “not working”, it’s important to figure out what specifically is going wrong — marketing is broad and encompasses everything your org is doing to bring more people to your product.

More often than not, one or more of these nine things is happening when a founder tells me, “Marketing’s not working!”

You’re chasing the big swings before the fundamentals are in place

I once worked with a founder who was eager to stand up billboards and TV commercials. The problem? Existing paid acquisition on digital channels wasn’t even close to breakeven, and scale was limited given the few markets the brand was operating in.

Poor performance from paid acquisition can stem from many things — some of which I’ll touch on more — like:

• Subpar paid channel strategy

• Messaging & positioning not resonating

• Unclear definition of target audience

• Conversion funnel not optimized

• Product issues

• Revenue model issues

The point is, you need to be completely confident in these fundamentals before you can go spend a boatload of money on bigger swings. Focus on dialing these in, and more importantly, understanding which you need to dial in, and the bigger opportunities will be no-brainers.

You hired the wrong skillset

Hiring at an early startup can be one of the most difficult things to get right, particularly when you don’t know what you don’t know! (P.S. see my marketing hiring guide for founders here.)

The most common mistake I see early startups make is that they hire someone, whether an agency or a full-time marketer, without ensuring that hire’s skillset is a match for what they need.

To avoid this trap, go through the following exercise:

• Clearly lay out concrete goals for the next 1 - 2 years

• Take stock of the marketing skills you already have in-house, if any

• Determine where the skill and experience gaps are

• Define the problem you’re looking for this hire to solve: Decide what combination of skills and experience will get you to your 1 - 2 year goals

It’s ok if you’re not sure how to determine this yourself. This is a great time to talk to an advisor who can help you think through how you’ve defined the problem, and what type of experience, specific skillset, and support level (e.g., consultant vs. full time hire) you need to get the job done.

Making the wrong hire is an expensive mistake, from both a financial and time perspective. It’s always in your best interest to find marketing support that understands the stage your business is at, where it needs to get to next, and the tools they have to make it happen.

You’re being prescriptive to your marketer or marketing team

It is a win-win when founders are involved in marketing. It’s also important to let marketing own and direct their strategy once you’ve brought in an expert you trust (otherwise, why make the hire?!)

Often times, founders have very specific testing ideas they want to run, which is great! Prioritizing these tests accordingly within marketing’s roadmap should be a collaboration between leadership and marketing, rather than a directive. Nothing derails a marketing team’s progress like unclear direction and diminished autonomy.

You think you just need to “get Facebook ads working”

This is one that my fellow marketers and I hear all. the. time…and we can’t blame you!

Facebook aka Meta is an incredibly powerful advertising platform, to which many brands attribute unlocking early scale.

But “getting Facebook ads working” isn’t a strategy in itself. This is another case where the priority should be nailing the fundamentals, without which you probably won’t see much improvement.

That said, a scalable paid channel like Meta is a great medium for putting up a bit of budget to get some quick feedback, iterating based on that feedback, and so on.

You haven’t properly prioritized or budgeted for creative resources

Creative development is one of the most overlooked pieces of the marketing puzzle, but it pays dividends for brands who prioritize it.

And this doesn’t mean expecting your growth person to be in charge of creating designs from scratch in Canva — scrappy marketers can often throw an ad together in a pinch, but design is an entirely different skillset than the analytical marketing approach taken by a strong growth person.

A good growth marketer will have an eye for creative that performs well, and will understand required specs. That said, they’ll need budget to work with a strong freelance designer or agency to continually produce the necessary assets.

Creative has always been an important marketing lever, but it’s become even more critical in digital channels in recent years. These days, creative IS your targeting on a channel like Meta. And if you’re quickly scaling on paid social channels, you’ll need new assets constantly to continue running your campaigns efficiently.

Once you start thinking of creative like a key component of the fuel in your growth engine, it becomes much easier to prioritize accordingly.

Your messaging isn’t resonating

Maybe you have awesome creative, the right marketer, a clear understanding of your audience, and an optimized funnel in place, but you’re just not seeing the pickup you’d expect from marketing.

Now’s the time to diagnose your messaging, and make sure it’s compelling and clear to your audience. Clear is ALWAYS better than clever, and please don’t try to bring humor into your marketing until you’re 100% certain your key messaging is resonating with an audience you understand inside and out. Here’s a classic example:

The infamous tweet from Burger King that *royally* backfired (no pun intended)

If you suspect your messaging just isn’t compelling enough, going through a positioning and messaging exercise like this one is a good place to start. Then, find a freelancer copywriter who specializes in your industry and have them take a look.

You don’t have product channel fit

Brian Balfour, founder of Reforge, explains this concept best: “Products are built to fit with channels. Channels do not mold to products…You control your product, you do not control the channel.”

Don’t devote resources to trying to make your product go viral if nothing about it lends itself naturally to virality. Don’t spend your entire budget on paid marketing if your value proposition is incredibly niche, or your time from consideration to purchase is very high. And don’t think you NEED to be on [channel x] just because you see other brands there.

You know your product, and hopefully your audience, so you know best what mediums to reach them. A local business in your hometown would never run billboards three states away — if you’re confident your audience isn’t on Snapchat, why advertise there? Don’t let the noise distract you.

You don’t have product market fit

You already know you have to build a product for your market (e.g., to solve a problem they’re facing) rather than for yourself (e.g., to sell the business and make some money).

Understanding that market and their problem, why your solution is the best way to solve their problem compared to existing alternatives, and how to get the word out about your solution in a way that speaks to your market’s motivations is just as critical.

A quick Google will show you there are endless PMF surveys out there, but here are two quick tests. You may not have found PMF yet if:

• Your audience doesn’t believe their problem would be way more painful if your solution went away, or

• You have way more supply than demand for your product after marketing it for a long time

Your budget is spread too thin (or maybe you’re just not spending enough)

There are two key mistakes I see when it comes to early marketing budgeting decisions:

1. A relatively healthy budget, that’s just spread too thin

2. A budget that’s just not big enough to show you results

The first problem happens when you’re appropriately prioritized things like ad spend, headcount, creative resourcing, marketing materials, etc., but you’re still trying to do too many things with your budget. Take a look line by line at where you’re spending, and try to consolidate your budget across what’s most impactful.

The second problem happens when your budget is just too low to produce meaningful results, or to give you enough data to clearly understand what’s working. There’s no real magic threshold, but this is where it’s important to understand your unit economics, model out growth, and determine what you need to spend to bring in new customers and retain them.

This week’s Growth Therapy is brought to you by the American Marketing Association. They offer certifications, training, webinars, you name it! Find your local chapter at the link below ⬇️

Join the Essential Community for Marketers

Stay ahead of the curve and join the thousands of marketers shaping the future of marketing with the American Marketing Association. As the largest community-based marketing organization, you’ll find award-winning content, professional certifications (PCM®), industry-leading training events, and vibrant local chapters. Members get access to an abundance of resources, on-demand courses, and the most essential community in marketing.

Get exclusive benefits when you become a member today!

✨ One marketing thing: I’m always trying to learn more about SEO, as it’s not my area of expertise. Here are 10 things every blog post needs, according to this recent post from Demand Curve.

✨ One fun thing: I might plan a trip to Miami just to eat at this new restaurant.

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

Sorry, this is a requirement, I don’t make the rules

Happy May! 🌼

This week I’m continuing to pound the pavement about how important it is to actually talk to your customers.

Without hearing directly from your customers, you’re really just guessing what they want. Maybe you’ve taken it a step further and have run enough proper tests to have some data to back those assumptions, but there’s still no substitute for a proper feedback loop.

I’ve heard founders say that because they are their customer, further feedback isn’t necessary. While a founder experiencing the same problem their company aims to solve is often how they come up with their initial business idea, once you’re too close to your product, you’re just not experiencing it the same way your users are.

So what’s a startup to do?!

Luckily, I chatted all about how to implement customer feedback loops with fractional COO Lidiya Becker.

• Lidiya is an operator with 15 years of experience across e-commerce, logistics, CPG, industrial, and IoT products

• She scaled Happiest Baby (maker of SNOO) from $0 to $50M in annual revenue, with a hyper-focus on the customer experience

• Currently, she advises and consults for high-growth brands at Build It Labs

Here’s what you’ll learn about:

1. Why feedback loops are so important if you care about your customer

2. The different types of data you should be considering

3. The right time for companies to implement feedback loops

4. Getting cross-functional buy-in when implementing a new process like this

5. Surprising insights that have come out of customer feedback loops

_{Our conversation below has been edited for length and clarity.}

Amanda Berg: Can you explain what a customer feedback loop is, and how you've used them?

Lidiya Becker: We’ve all heard that it's important to listen to your customers. But interestingly enough, I think few companies are doing this systematically.

This means systematically listening to a customer's feedback and pain points, to understand whether the product or service is solving a real problem and providing value. I think it's dangerous to build assumptions about what customers want without truly listening to them. 

Rather than coming up with some internal POV on what customers are saying, it's about going directly to the source and limiting bias. So it's not a big surprise that the best companies really understand their customers’ pain points, and build solutions to address them. Feedback loops enable us to get close to a customer's voice systematically, and without investing an inordinate amount of time.

• Step one is to listen to a customer's voice, using both metrics and anecdotes, on a regular cadence.

• Step two is to turn those datapoints into insights, and take action on those metrics or anecdotes, and then repeat the whole process again — listen, learn on a regular basis, derive insights, and take action.

AB: Can you talk about how you've used these feedback loops, and how you've set them up?

LB: Generally speaking, you'll have some sort of measurement, like a dashboard or report that's shared on a regular cadence via email or a meeting or in a collaborative tool like Notion. The owner could be, for example, a customer experience or customer care lead.

Most importantly, there's an insight and POV provided by that owner. Facilitating some sort of dialogue, and ensuring there’s an action plan, means this can be built into a repeatable process and cycle of continuous improvement.

That said, this process shouldn’t just include quantifiable metrics. Anecdotes, and listening to the customer’s voice in their own words, are incredibly important and complementary to the data.

AB: Something I see a lot of earlier stage companies struggling with is figuring out that “how” piece of implementing a process like this. They’re asking whether they need a tool to do this, if they should send out a survey, how to incentivize folks to get on the phone with them, etc. How do we actually get from “we know we need to do this,” to actually actioning on it?

LB: I don't think there's necessarily a secret, but it's important to experiment with a lot of different approaches and determine a way that you get some sort of reasonable sample size. It also has to be repeatable. 

There are a few different approaches to hearing exactly what the customer is saying:

• Go into your social channels and see what customers are writing in your ad comments, in your messenger. You’ll want to do what’s called sampling — sample a set number of transactions, every set number of weeks, and focus on finding themes.

  • You might just focus on social and see if you can observe any patterns, like what customers are or are not happy about. This approach may not be scalable, but it's important because you're getting really close to the customer.

• And of course, there's your customer service system, like a Zendesk, where customers are emailing or chatting or calling you. The leader of a customer care organization should be reviewing those transactions on a regular cadence, and using a rubric to understand and measure whether you’re serving your customer well. 

• Lastly, from a marketing perspective, you can send out surveys and you can do focus groups and those can be quite interesting as well, but those require a whole different set of resources.

  • For surveys, I’ve recommend thinking strategically about the questions you ask in order to limit bias. Talking to Humans is a great read that addresses this.

AB: At what stage should companies think about implementing these feedback loops? Is it ever too early? And secondly, what sorts of resources, whether it's budget or tools or headcount, are really needed to do this properly?

LB: The right time is yesterday! As early as possible is my answer, with 15 exclamation points! Before you launch, you want to understand if your product is solving a true pain point, and that customers want it. So gaining feedback is incredibly critical from that point on. Even if you don't have a product yet, you can have potential customers react to something that's close to your product.

In terms of systems, you don’t need anything fancy or a lot of investment to do this. Things can get more complex as you scale, but early on, having some high level metrics, or even proxy metrics, is enough. It's easy enough to track customer satisfaction (CSAT), to send out an NPS survey, to collect reviews. There's a lot of tools that do these things pretty out of the box, making it easy for you to review them and take action.

AB: How do you think about assigning metrics, even if they're softer metrics, to the more anecdotal side of feedback?

LB: I love the idea of rubrics. We look at an interaction with a customer, and we evaluate it and codify it. A rubric allows you to systematically define what good looks like, with checks like — are we providing a good experience? Are we making it easy to do business with us? Are we talking to customers in their own language and avoiding jargon? Things that aren’t necessarily quantifiable.

AB: I love what you said about how these metrics aren’t always quantifiable, because supplementing things like your CSAT and NPS scores, your star ratings, with more anecdotal and less quantifiable metrics will give you a really full picture, that looking at just one or the other wouldn’t give you.

LB: In companies of all sizes, I see this tension between anecdotes and metrics all the time. We have a CSAT score of 95%. But you see really negative reviews, so you know there's some sort of disconnect. 

Jeff Bezos actually has a really good quote on this. It’s something like “when the anecdotes and the metrics disagree, usually the anecdote is right.” 

The point is that the anecdote is pushing you to dig deeper. Maybe either the metric or the anecdote is a one off that represents a tiny portion of your customer population, and you should do nothing about it. But it could also be that we're missing something. We're missing a high risk situation. It could be that we're measuring things incorrectly. We could not be thinking about things the right way.

So it's definitely worthwhile to push the organization to really understand what's going on, to ultimately serve a customer better.

AB: When it comes to actually building these feedback loops into an operational process, how do you ensure that there's cross functional buy-in and that teams actually adopt these frameworks?

LB: Buy-in is so important, because customer experience is a cross functional effort. It's never just a finance problem, or operations problems, or a marketing problem. We have to be very aligned and work well together to support the customer.

There are a few ground rules that enable some companies to be very effective at serving the customer, without which it’s going to be challenging.

First, a growth oriented culture is important. Feedback loops of all kinds work exceptionally well in a culture that values learning and improvement, and minimizes finger pointing and blame. It's such an important example to set from the top down.

As soon as there's blaming and shaming, people will be much less willing to speak up, offer novel ideas, expose problems, and this is such an important thing to protect. I can't stress this enough, because anything else you build will probably crumble if you don't have this kind of culture.

AB: Just to level-set a bit, I think companies need to know that no matter how great their business is, customer feedback is never going to be 100% positive.

There's always going to be a person who's not happy, and maybe that person just isn’t your customer and that's okay. Or maybe there’s actually something that went wrong. And it’s so important for companies to be open to feedback, and accept that improvement is a requirement.

LB: 100% agree. It’s always hard to hear negative feedback, but having this attitude of curiosity and openness and understanding is key. Like you said, there’s always going to be negative feedback — but these are really opportunities. And it might be an opportunity to further define who your ideal target customer is and who is not. It’s okay to make some customers not happy, because they might not be your target customer, and you're not going to optimize your business for them.

Back to ensuring adoption of a process like this, the second thing you need is setting up the right level of visibility to enable collaboration. So we need to show the right information to the right people in an easy to understand way, so it can be actionable. That means translating insights into a hypothesis, or a rubric score, so teams can learn and react. Again, feedback loops are all about learning. 

The third thing is that ultimately people want to be part of a winning team, and are motivated when they see progress. When I led operations at Happiest Baby, this was key to implementing a system of a variety of feedback loops, not only customer-driven. Being a part of the process means you see improvements over time, which is really motivating and creates an ongoing flywheel.

AB: Surfacing what doesn't work is important for improvement, but surfacing what does work is just as important so you do those things consistently. Identifying those wins so you know what NOT to change, especially when you’re early, is so important. 

LB: It's so important to share wins because I think just as humans, positive feedback is so meaningful for the people on the team doing the hard work. I've seen this drive very quick and meaningful progress, over and over.

AB: Are there any common pitfalls you've seen when companies try to implement these feedback loops?

LB: A few:

• Focusing solely on metrics and ignoring qualitative feedback, like customer anecdotes. Quantitative and qualitative data are complementary, and then anecdotes give valuable context and often expose blind spots. 

• I’d also say lack of follow through. I've seen this happen time and again in various companies I’ve advised. They’ll build this beautiful dashboard and then forget all about it, or even if they look at it every month, there's no insight that drives an action. Our process is only as good as our commitment to follow through. If we observe a trend or an issue, we have to make a hypothesis and take action to test something.

  • Let’s say we have a higher abandoned cart rate than average in the industry. We might make a hypothesis that we think there are too many steps in the checkout process, so we're going to try to streamline it. And then I've seen a lot of companies take action, but forget about what happened as a result. There’s a level of discipline needed to track experiments, monitor their results, and to mitigate that risk. It’s important to assign a single accountable party that oversees testing initiatives, and then setting a concrete time to follow up.

AB: What’s one of the most surprising or valuable insights that you've learned from customer feedback, that you wouldn't otherwise have uncovered?

LB: So many — customers are smart and savvy and helpful. I believe that customer trust must be earned over and over. 

One of my most valuable experiences in my career has been working at a company called McMaster-Carr. They're a hundred-plus year old, billion dollar company. They operate at an incredibly high profit margin, and are hyper focused on the customer. I’ve seen that companies who operate like this with integrity and with empathy for their customer really thrive. The lesson that doing the right thing for the customer tends to lead to a good outcome for the company too is one I’ve taken and applied to other rapidly growing startups. 

In terms of a specific insight, one of the most surprising, that I’ve since seen over and over again, is that McMaster customers were more satisfied overall after we made a mistake, but recovered well. So if you drop the ball during a customer experience, and if you really care and do a good job recovering, their satisfaction scores are even higher on average, which to me was so surprising.

AB: Wow, they’re more satisfied than if everything just goes right from the start? That’s so interesting.

LB: Exactly. So we have to have a soft landing, we have to recover gracefully. There's still such a huge opportunity to win the trust of your customer, even if there’s been a poor initial interaction. This has proven true time and time again.

AB: Any last closing thoughts on feedback loops, or anything else that's super important for companies to know if they want to implement these things or improve their processes?

LB: I think the most important points we've gone over are:

• The marriage of quantitative and qualitative is so important.

• The commitment to growth and learning is so important.

• The principle of companies being customer obsessed ends up serving the business well.

• And overall, I think the process is an Agile-inspired operational process. It aligns with a lot of Agile principles, like transparency, collaboration and continuous improvement.

This week’s edition of Growth Therapy is brought to you by Black Crow AI. They help make sure you can identify ALL the traffic that comes to your Shopify store so you don’t lose out on customers. Shopify brands, get $100 when you book a demo at the link below! ⬇️

How much is anonymous traffic costing your business?

For DTC brands, unidentified shoppers = untapped revenue.

Black Crow AI identifies 100% of returning users & predicts shopping behavior patterns so you can effortlessly acquire more customers.

Shopify brands with $2M+ in GMV get a free $100 gift card to book a demo.

✨ One marketing thing: Three excellent retention insights from Eli’s Newsletter.

✨ One fun thing: Is it still fun if it makes you cry? 🥲 See why elephants are the GOAT

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩

How to visualize different types of attribution models. Image credit: Impact.com

Attribution modeling has always been a hot-button marketing issue. Simply put, attribution is your ability to credit your marketing initiatives with success or failure in driving key outcomes.

But because different attribution models assign credit to touchpoints in the consumer journey in different ways, each one — even the most complex and expensive to implement — will give you a different version of the “truth”.

Why the concept of attribution matters

The more you understand the success of your marketing initiatives, the better you can understand:

• What your customer path to conversion looks like

• How to allocate resources across marketing initiatives

The more insight you have into your customer journey and how the dollars you’re spending are driving your desired outcome, the better and more effective your marketing strategy can be. Simple, right? 😉

Of course, it’s never quite this straightforward. Think about your typical decision making process and journey to making a purchase, and how that compares with each of the different attribution models below.

Different attribution models, defined

At a high level, there are two main buckets an attribution model can fall into: single-touch attribution and multi-touch attribution (MTA).

Each have their merits, and some models are better than others depending on the complexity of your customer journey.

For example, a single touch attribution model may work just fine for an ecom product that’s made as an impulse purchase, whereas a multi-touch attribution model may be more appropriate for a purchase that has a longer sales cycle, requires more research, or has an omni-channel component.

Single touch attribution models

The basis for single-touch attribution models is that there’s a single touchpoint in the user journey that’s the most influential in driving a conversion.

• First touch attribution: The customer’s first touchpoint gets credit for driving the conversion

• Last touch attribution: The customer’s last touchpoint before a conversion occurs gets credit for driving that conversion. This is probably the most commonly used, lowest-lift attribution model, particularly if you use UTM parameters for ad and link tracking

Honorable mention:

• “How did you hear about us” or HDYHAU: Not so much an attribution model as an additional datapoint, this is a question asked immediately post-conversion with preset options for the user to choose from. A few notes:

  • Since this data is self-reported by the user, it’s sometimes undeservedly seen as “less true” than UTM-based tracking. But it’s actually pretty trustworthy — if you’re skeptical, add some dummy options (like TV, if you’re not advertising on TV) and you’ll see how virtually no one thinks they heard about you this way.

  • It’s precisely because this information is user-reported that it’s so valuable in helping you understand what initiatives are memorable to your users

Multi-touch attribution models

The basis for multi-touch attribution models is that there are multiple touchpoints in the user journey that influence a conversion.

• Linear attribution: Equal credit for the conversion is given to all touchpoints. E.g., if you have 4 touchpoints, each is responsible for 25% of a conversion

• Position-based attribution: More credit is given to specific touchpoints at certain parts of the user journey, depending which are most important

• Time-decay attribution: More credit is given to touchpoints the closer they are to a conversion

• Data-driven attribution: Credit is assigned to the most influential touchpoints, according to machine learning

  • These will typically be touted as the “best”, especially with all the advancements in ML made available to marketers in the last year or so, but the downside is they’re a total black box

A note on marketing mix models (or media mix models)

Marketing mix modeling / media mix modeling / MMM, as I’ll refer to it from here, is more of a statistical model than an attribution model, but it’s still worth mentioning here.

MMM takes into account all potential factors that lead to a conversion — not just touchpoints in the user journey — and assigns relative credit to them.

Funnel has a great explainer video on MMM that includes a helpful sports analogy:

Think about trying to assign credit to individual players in a soccer game based on the outcome of that game. Obviously the player who scores a goal doesn’t get 100% credit for that goal, or for the win. This calculus is what an MMM tries to solve for, taking into account as many factors as possible to appropriately credit them for the “win”.

MMMs can be built by in-house data scientists, specialized third party agencies, or leveraged through external SaaS platforms.

Note that while MMM is an option within the total universe of attribution possibilities, it’s not a practical option for a small startup. For sake of comparison, The Knot recently published a great article about building an MMM practice, and their annual revenue is around $400m. You need a LOT of data, and money, to build an MMM, and I bring it up to demonstrate how deep and complex you can go on attempting (operative word) to understand attribution if you wanted to.

Taking it back to reality

This past week I saw an influencer ad for a pair of pants I decided I absolutely must have. I clicked on the ad, and immediately determined, based on the gimmicky site (compare-to pricing! countdown timers! spin to win!) and lack of fabric composition info on the product page, that this particular pair of pants was going to be low-quality and I would look elsewhere.

I was then absolutely bombarded with ads from competing brands, quickly realized this was all the same pair of dropshipped pants at varying low price points, and resolved to look for something similar from other brands I trust. I narrowed it down to a few pairs and ultimately ordered from a brand I’ve purchased from before.

In the interim, the brand I purchased from had sent me a few emails, and I’d been to their site directly a couple of times to browse for the pants I was looking for. But those touchpoints — the emails and my direct visits to the site — were certainly NOT the driving forces behind my purchase. See where this gets complicated?

Maybe I’m more neurotic than most in my purchasing journey, but the fact is there are MANY factors that influence a user’s decision to convert, and, most importantly, those factors don’t happen in a trackable vacuum.

No single attribution model can take into account every external factor, particularly those that happen inside your customers’ brains! MMMs, as they’re custom built, can do a better job of incorporating the macro, but you have to have LOTS of data, and $, to make building and employing an MMM worthwhile.

Know your size

If you’re not doing at least $10m in annual revenue, I wouldn’t recommend spending any of your budget on advanced attribution modeling, and certainly not on building an MMM.

You should be able to get 95% of the way there with a combination of data sources, like your ad channels, your source of truth database, UTM tracking, and post purchase surveys. The exact combination is where the art and the science comes in.

Marketing to humans

While it’s always worthwhile to understand the world of possibilities when it comes to attribution, and have something to aspire to ($450m in annual revenue? Yes please), it’s also important not to lose sight of the fundamentals.

Reminding ourselves that we’re marketing to our fellow humans is a theme I always like to come back to.

Some attribution models are going to be better than others depending on the nature of your business, but none alone will ever provide a complete version of the truth. That’s because the humans on the other side of those models are making countless decisions and consuming tons of information on a daily basis. Think back to my pants example, and for such a simple purchase!

Supplementing your quantitative attribution data with qualitative customer insights will always paint a richer picture than even the most advanced attribution model can.

My tactical advice

To do this, the approach I typically take with small startups (under $10m annual revenue) combines a few different methods to triangulate a plausible version of the truth. We get a clearer picture of performance this way, when we know in our heart of hearts that last-touch alone isn’t enough.

This method is basically free, and gets you 95% of the way there, even if you’ve started to incorporate less trackable marketing, like audio / video / out of home:

1. Track last touch conversions by channel based on UTMs

2. Implement a HDYHAU (How did you hear about us?) survey immediately post purchase

3. Track per-channel conversions according to both UTMs and your HDYHAU survey

4. Calculate your per-channel CPA for both UTMs and HDYHAU

5. Compare your per channel CPAs between UTM and HDYHAU to understand how channels are being over- or under-credited on a UTM basis (paid search is almost always over-credited, while paid and organic social are almost always under-credited)

6. Use these insights to inform your channel investment

7. Bonus: Track how your percentage of conversions by channel from your HDYHAU results changes over time to understand what initiatives resonated with your users

Ultimately, the art-and-science of it all can be uncomfortable for very data-driven marketers. Getting comfortable with understanding signals, even when your data isn’t and will never be perfect, allows you to move faster and make even better decisions.

This week’s edition of Growth Therapy is brought to you by Hubspot. They just released a guide with 100 prompts you can give ChatGPT to make your work day way easier. Grab it at the link below! ⬇️

Want to get the most out of ChatGPT?

Revolutionize your workday with the power of ChatGPT! Dive into HubSpot’s guide to discover how AI can elevate your productivity and creativity. Learn to automate tasks, enhance decision-making, and foster innovation, all through the capabilities of ChatGPT.

Download the offer for actionable insights and tips on harnessing AI for a more efficient, impactful professional life. 

📊 One marketing thing: A newsletter recommendation for the B2B marketers in the house

✨ One fun thing: If you like a quirky feel-good story, this one is complete with a beloved and irreverent Facebook group, a movie-worthy love story, and a wedding officiated over margaritas at a local joint (by none other than my dad!)

Was this message forwarded to you? Need a little Growth Therapy? Sign up below 👇

Questions? Comments? Topic requests? Just hit reply ↩