WEBVTT 1 00:00:00.100 --> 00:00:09.720 [upbeat music] Welcome to a special Cannes edition of the Rebooting show. I am Brian Morrissey. Over the next few weeks, I'll have conversations that I had during the week in the south of France. 2 00:00:10.240 --> 00:00:18.540 In the following conversation, I ditch my usual recording situation of a darkened room for a poolside villa in the south of France. 3 00:00:18.780 --> 00:00:31.490 I am joined by Dotdash Meredith CEO Neil Vogel, and Axios senior media reporter Sarah Fisher, for a wide-ranging discussion of the implication of AI's collision with the publishing industry. 4 00:00:31.820 --> 00:00:42.160 Neil discusses the deal Dotdash Meredith cut with OpenAI recently, as well as his contention that many companies are using AI as a handy excuse for their own poor performance. 5 00:00:42.700 --> 00:00:55.780 Sarah provides a good technical point on this, that for many publishers, the nuances of copyright law weaken their hands in negotiations, which make deals an easy choice as opposed to the uncertainty and, uh, definite expense of litigation. 6 00:00:56.280 --> 00:01:05.340 I hope you enjoy this conversation. If you do, please be sure to like and review this podcast if you have a chance, and send me your feedback to bmorrissey at therebooting dot com. 7 00:01:05.820 --> 00:01:19.360 Now on to my conversation with Neil and Sarah. [upbeat music] Thanks for joining. This is the second annual, Neil. Is this gonna be, like, a tradition? Uh, if I don't get fired. [laughing] Yes. Okay. 8 00:01:19.840 --> 00:01:29.280 Well, let's get into whether you're gonna get fired or not. Thank you all for coming. I really like Cannes for these little, little intimate events. Thank you for having us at your lovely home, Neil. 9 00:01:29.420 --> 00:01:39.020 For those who are listening to the podcast, this is-- we're doing this at the Dotdash Villa, which is in Cannes. It's a little off the Croisette. There's a nice pool here. We've got some rosé. 10 00:01:39.100 --> 00:01:44.020 I have not been drinking any. I can't, can't speak for my guests. Neil did not respond. Not today. 11 00:01:44.140 --> 00:01:53.020 So joined by Neil Vogel, the CEO of Dotdash Meredith, and Sarah Fisher, the senior media reporter at Axios, who I will say must-read... 12 00:01:53.100 --> 00:02:01.570 Everyone reads Sarah, but, like, you are, I think, the, like, sort of must-read of doing [clapping] daily reporting in this industry. Ditto to you, Brian. You are also a must-read. Yeah, but we do different stuff. 13 00:02:01.570 --> 00:02:09.580 And a must-listen, and a must-listen. I do different stuff, but you're like, you know, you, you, you know exactly what is going on. And so I wanna actually start with you, right? 14 00:02:09.720 --> 00:02:16.519 Because some- Can we touch on the must-read thing for a second? Yeah, go ahead. Because I, I... You guys are a must read. We're gonna object. [laughs] We also have here- This is gonna get awkward. 15 00:02:16.640 --> 00:02:24.320 No, no, no, this is important because this is gonna be the premise of the whole thing. [laughs] We pretty much, for our people, also have, like, a don't read list. Like- Who's on it? These are the... 16 00:02:24.380 --> 00:02:33.660 I can't tell you that, obviously. [laughs] That's an hour now. But, like, you should not read this because it is nonsense, and it is not fact-based, and it is telling you the wrong things. 17 00:02:33.740 --> 00:02:42.430 And, like, there's a woman named Leah Wire, who works for us, who runs people, who always says, "The math is not mathing." Meaning, like, what you're saying makes no sense. 18 00:02:43.060 --> 00:02:49.320 And there are so many people who write about this industry without doing the work and without understanding what's going on. 19 00:02:49.760 --> 00:02:59.320 We end up with these upside-down narratives that, like, the world is ending and everything is horrible, and, like, everything is not horrible. The world is definitely wiggling, but it's not ending. Okay. 20 00:02:59.380 --> 00:03:05.240 Well, I wanna get into that. Agree. Because, like, and this was my first question, Neil. Rant one of what will be like ten. Okay. Yeah, that's my nerd. 21 00:03:05.260 --> 00:03:14.700 But 'cause someone was-- I was talking with someone, and they said, "Oh, you know, I, I, I, I went to this, this session and, and Sarah was moderating it, and it was such a downer. 22 00:03:15.160 --> 00:03:23.820 She kept going down all of this list- What?... of all of the problems- Wait, when was this?... facing the publishing industry." I don't know which one, because you've done a lot of sessions this week. 23 00:03:23.880 --> 00:03:34.020 Yeah, but I don't remember. How many? Twenty-one. Twenty-one sessions. Hard. [laughs] That's insane. You didn't do prep calls for all twenty-one, did you? All of them. Oh, my God. What is going on in your view? 24 00:03:34.060 --> 00:03:43.600 You're, you're reporting every day on this, okay? There is, there is, I think... I know what Neil is going for here. There is this negative narrative, and, and, you know, Neil, I don't... 25 00:03:43.660 --> 00:03:52.660 I think I can expose Neil as the person who called me a negative dude. And I say, "I'm just a weather forecaster, and it's gonna rain, and we're in England." 26 00:03:53.040 --> 00:04:03.340 What, what do you see going on right now overall in the publishing industry? Because we keep hearing about extinction narratives, and Neil's gonna push back on that. I will too. 27 00:04:03.800 --> 00:04:15.300 So I think what's going on is between, like, twenty fifteen and twenty nineteen, double-digit digital growth, that's obviously not sustainable, but we only had that for advertising. 28 00:04:15.320 --> 00:04:26.480 We only had that because linear mediums were still converting to digital, so you had expedited growth. Then the pandemic happens. We have volatility for, like, you know, three to four years. 29 00:04:26.969 --> 00:04:37.820 And this is the first year when I got a lot of analyst forecasts that they said, "You know, we're kind of stable now." The difference, though, is that we're at, like, single mid-digit growth. 30 00:04:38.360 --> 00:04:46.159 So what happened was we all built these businesses anticipating growth that was not sustainable, and so then we had to rightsize our businesses. 31 00:04:46.350 --> 00:04:51.540 So in the beginning of the year, it felt doom and gloomy because we were doing a ton of layoffs and restructurings. 32 00:04:51.560 --> 00:05:04.540 And also part of that was because digital ad business is slowed growth, linear is negative growth, we had to diversify, and so everyone had to come up with new products, new teams, and that just felt messy. 33 00:05:04.600 --> 00:05:16.240 Does it mean that we are extinct? Absolutely not. The challenge is there will and there was some publishers who were ill-equipped to ride some of these changes, but they would've gone out of business anyway. 34 00:05:16.730 --> 00:05:27.900 So I actually think the publishing landscape is relatively healthy, but if you're a, like, a traditional publication, obviously it's existential because it's really hard to m-move your business over. 35 00:05:27.940 --> 00:05:40.000 And also the CPMs in the digital business might not be as high as they were. If you're, like, a cable network, like, your best days might be behind you unless you have a really smart digital strategy. 36 00:05:40.080 --> 00:05:48.400 So I think it-- the publishing landscape has a lot of bright spots, but, you know, there's challenges. Yeah. Neil, talk to me about the math, because you're not in the news industry. 37 00:05:48.440 --> 00:05:58.900 I think a part of the narrative issue is a lot of parts of the n-news publishing industry are extraordinarily challenged for a bunch of different reasons. 38 00:05:58.910 --> 00:06:08.020 I think, I think for whatever reason, people confuse publishing and media with news. Agree. Because the people writing it are in the news industry. Correct. Agree. Correct. Generally speaking. And- Yes... 39 00:06:08.040 --> 00:06:15.240 there's, right, there's a lot written about, and, and we very, we, we say over and over again, "We don't do news. We don't do politics. We don't do sports. It's not what we do." 40 00:06:15.360 --> 00:06:23.592 We do-We've been through this a million times, like home, food, tech, travel, health, finance, beauty, style, entertainment. Like, that's what we do. Areas of passion and interest. 41 00:06:23.672 --> 00:06:29.392 We are double-digit revenue growers the last two quarters, right? Which is an anomaly, though. Yeah, right. 42 00:06:29.572 --> 00:06:40.052 But my, my view is, and it's an extension of what you just said, everyone is confusing challenged businesses and challenged business models with something wrong with the industry. 43 00:06:40.512 --> 00:06:43.632 There is nothing wrong with the industry. Like, I, I think... 44 00:06:44.072 --> 00:06:53.672 And again, look, past performance is no guarantee of future results, but I think we've worked really hard to build loyal audiences around really strong brands. Everybody always wants to know what our trick is. 45 00:06:53.692 --> 00:07:03.092 Our trick is nothing. Our trick is we just work really hard. We make really good stuff, and we're very smart about how we monetize it, and we're, we're very consistent, and we are boring as can be. 46 00:07:03.612 --> 00:07:15.672 And, like, every trend that comes up and every article all the don't-read people write about, like, "Oh my god, scale is terrible, this is wrong, it's..." No. Pivot to events. P- nonsense. No, no, no. 47 00:07:16.272 --> 00:07:26.722 Get very high-quality scale under high-quality brands, deliver real value to advertisers and consumers, and you're good. These have been the rules of media for a hundred years, right? 48 00:07:27.092 --> 00:07:37.372 Media companies as far back as you could see, if they're doing a good job, should have twenty, thirty percent EBITDA margins and should grow between five and twenty percent a year. And those rules are still the same. 49 00:07:37.432 --> 00:07:49.452 If you're sentimental and-- or don't understand the new way to do it, you can't blame your inability to change or your organizational inertia on something existential in the space. 50 00:07:49.512 --> 00:07:53.192 Now, there are existential things going on in the space, which I'm sure we're gonna talk about, but we're not talking about that. 51 00:07:53.212 --> 00:08:01.862 We're just talking about, like, zoom out at how people run these businesses, and you, you invest money, you expect return. It's very simple. I totally agree with that. Yeah. 52 00:08:01.892 --> 00:08:09.972 And just a thing to add about the distinction between news and non-news, there's two things impacting the news industry that are not necessarily impacting the broader publishing industry. 53 00:08:10.132 --> 00:08:20.292 One is there is not a massive trust deficit in, for the most part, non-news content. There is a massive trust def-deficit in news content, and there's massive... 54 00:08:20.752 --> 00:08:31.732 People argue me-- with me on this, but I believe that there is real news avoidance happening in the wake of two major wars. So because of that, you have a little bit of a different challenge in attention. 55 00:08:31.752 --> 00:08:42.432 And then the other challenge is from a margins perspective, like, the news industry has so many opportunities to be optimized with things like AI, but it's still, for the most part, like, really human-intensive. 56 00:08:42.472 --> 00:08:52.512 Like, good investigative, good reporting is really human-intensive. Whereas I think with a lot of other content creation, you can do so much more optimizing with things like AI and all this stuff. 57 00:08:52.532 --> 00:09:04.092 We are gonna get there at the news industry, but for now, we're kind of being a little cautious until, like, the copyright laws ensure that our IP is protected. Okay. We've got an awning moving. We had this last year. 58 00:09:04.162 --> 00:09:12.442 Am I having, like, a flashback? I swear to God we did-- that happened last year. We also got rain though last year too a little bit. Okay. So speaking of rain, AI. Like, I think when, when... 59 00:09:12.452 --> 00:09:18.052 Well, not AI, I'm sorry, Search. Search. Yes. Because a lot of times it's conflated. There's a lot of changes going on in Search. 60 00:09:18.232 --> 00:09:27.542 I can remember years ago, I was like, "Neil, explain to me, you're dependent on an algorithm. We're seeing" [laughs] It's, it's gonna keep going. 61 00:09:28.092 --> 00:09:36.032 And you said it's better to be- It's a very good, it's like, metaphor for Search, right? [laughs] It's, it's better, it's better to be attached to a very stable algorithm. 62 00:09:36.092 --> 00:09:40.592 Now, your business has diversified with traffic sources. This was back when you were a dot dash. 63 00:09:40.672 --> 00:09:55.032 There was not all of the, the Meredith properties that came over, but I would think when, when Search changes, it, it throws your business into a bit of a disarray. It does and it doesn't. I, I think, look, if you are... 64 00:09:55.092 --> 00:10:04.012 We've, we've talked about this for, like, going on a decade now. If you're a publisher and you blame algorithms for your problems, you're a very bad publisher. 65 00:10:04.032 --> 00:10:11.451 And part of how we've built our business is a lot of diversity. We're a lot of diversity across industries. We're a lot of diversity within industries. 66 00:10:12.012 --> 00:10:19.892 So we try to be set up so any one algorithm, we're not overly exposed to in any way. So if there's... 67 00:10:19.932 --> 00:10:29.152 I mean, we've been dealing with Google algorithm changes for a decade, and if you look at our audience, and we disclose all this publicly, our audience just keeps growing and growing and growing and growing. 68 00:10:29.172 --> 00:10:34.372 And that's obviously a, a sizable fair part of it, as it is for, like, everybody on the internet. 69 00:10:34.872 --> 00:10:42.212 That being said, I, I, I don't think I'm, like, divulging any state secrets to say Google has gotten significantly less predictable in what they do. 70 00:10:42.412 --> 00:10:51.902 Like, the rise in Reddit out of nowhere, like, I don't know who made that decision. Like, the AI showing up at the top of the page, which is effectively like a giant answer box, but more, but this and that. 71 00:10:51.972 --> 00:10:58.462 And the shuffling of the page and all of this and all of the nonsense that they just decide that, like, "Oh, you're a publisher that does this. 72 00:10:58.492 --> 00:11:10.772 Well, we don't like this, so you can't do it anymore," without any sort of warning. It feels somewhat arbitrary. It's not particularly affecting us because I think we're very prepared for ups and downs. 73 00:11:10.812 --> 00:11:20.192 There are also parts of the algorithm... No one ever talks about when the algorithm benefits them, right? 'Cause it do- it does that too, right? So we're not complaining about Google. 74 00:11:20.272 --> 00:11:31.172 We can, we can talk about Google in the context of AI and their behavior versus other people with premium publishers and people like Axios, but it's just the state of the state. Things are moving faster. 75 00:11:31.232 --> 00:11:39.892 I would love it if no algorithm and no user ever changed their habits in any way and locked in today, we would just crush it, but that's just not the way of the world. 76 00:11:40.432 --> 00:11:51.332 And we just- But how much of your distribution of the people arriving are through s- me- intermediated by an algorithm? Oh, I'm not supposed to answer that for, for, like... 'cause we're partners. 77 00:11:51.372 --> 00:11:57.182 Okay, fifty percent is Google- No, no, I can- Another one is seventy-five percent. I don't know. Like, like significantly more than half. Okay. Okay. Significantly more than half. 78 00:11:57.182 --> 00:12:03.382 So, like, you're, no matter what, your business is gonna be at, at, at the mercy of some kind of algorithm. Almost every business on the internet- Yeah... 79 00:12:03.382 --> 00:12:13.202 whether you're a publisher, whether you're a travel site, or whether you're a retailer, is fifty percent, seventy-five percent algorithm in between you and there. And, and can I just make a note on that? Yeah, of course. 80 00:12:13.252 --> 00:12:21.172 Even for, like, a news-based business, which is much less dependent right now on social, I think we're still highly dependent on Search. 81 00:12:21.252 --> 00:12:27.852 Even other sort of interme- intermediaries where you're going direct to consumer, we're still being impacted by algorithms. Yeah. 82 00:12:28.052 --> 00:12:33.566 I mean, I can't tell you how many times we deal with spam filters in Gmail.We're an email business, you know? Yeah. 83 00:12:33.596 --> 00:12:41.996 It's like you're always gonna be at the whim of other people's platforms to an extent, but if the audience has a direct relationship with you you'll still... They'll figure out how to get to you. 84 00:12:42.246 --> 00:12:51.356 And think that-- and th-that's the key, right? Uh, so which brings me to AI, because AI is, is gonna affect email too. I mean, it's summarizing emails- Yes...and that's inevitable. 85 00:12:52.126 --> 00:13:03.876 And Jim VandeHei has, has, has written about, um, this as a looming threat at least, and, and adapting the business to, to address it in a, in a bunch of different ways. But let's talk about the AI. 86 00:13:03.925 --> 00:13:10.206 I mean, uh, you, you made a, a... You did... You were one of the publishers that really was one of the early deals with OpenAI. 87 00:13:10.696 --> 00:13:18.976 So you wanna just, like, tell us all the terms of it, uh- I mean, I can tell you what we've said publicly, and it's, it's- [laughing] But look, we, we believe, and, and we've been very upfront about this. 88 00:13:19.016 --> 00:13:25.196 Like, I work for Barry Diller. He's been very, very upfront about this. We believe our content is incredibly valuable, which it is. 89 00:13:25.656 --> 00:13:36.736 Evergreen content, researched, it's health, it's written by a doctor, blah, blah, blah, all that stuff. We need to be paid for training, and if you're going to use us in your results, we need to be cited. 90 00:13:36.916 --> 00:13:43.956 We need credit, and the credit has to be visible, and that has to... the credit has to benefit us. And then we need other stuff, like we gotta escalate as things grow. 91 00:13:44.496 --> 00:13:58.586 And what, what I can say about OpenAI is I think they have a pretty good understanding of the ecosystem. I think they've been an excellent partner. I think they put forth a deal to us that was... 92 00:13:58.616 --> 00:14:03.376 Look, it made Barry Diller happy, and it made me happy, and it made our whole team happy, so we, we did it. 93 00:14:03.936 --> 00:14:15.596 Again, I would say as a company, we've been fairly vocal on the need for fair consideration for all the work we do, and it's a model for things we can do going forward. Now, everyone's... 94 00:14:15.676 --> 00:14:26.196 You said like, "Oh, this, this is like the Netflix... This is like Netflix, where Netflix gave... The studios gave Netflix all their movies, and then they lost it." No. They already have our movies. 95 00:14:26.576 --> 00:14:34.046 Like, they already have our stuff. Yeah, but Barry was talking about suing, right? Wasn't he, Sarah? Like- Well, he was talking about creating a coalition of publishers- Yeah. 96 00:14:34.256 --> 00:14:41.976 And-...to work together- Which never works...to create a label framework. Yeah. They... Well- [laughs]...and part of the reason it didn't work is because publishers just generally have such varied interests. 97 00:14:42.416 --> 00:14:52.956 Like, for example, a publisher that has, you know, really heavily re-- that's really heavily reliant on search is totally different than a publisher that is really heavily reliant on podcasts. 98 00:14:53.016 --> 00:14:56.225 So, like, why would there be a carried interest amongst all of them? 99 00:14:56.716 --> 00:15:09.736 Like, I think the way that w-I'm thinking about it, and I would welcome your pushback on this, Neil, is if you're thinking about working with an AI company, it's like, okay, do we sue? Do we strike a deal? 100 00:15:09.756 --> 00:15:16.356 Do we do nothing? Generally speaking. If you're gonna sue, how much leverage do you have? There's like two ways you can think about this. 101 00:15:16.676 --> 00:15:27.616 There's statutory damages, so I filed registrations with the copyright office, and if you ripped off that work, I actually have a stronger legal case to be compensated for the damages. 102 00:15:27.976 --> 00:15:38.276 Then there's actual damages, which I didn't file any sort of, you know, registration that I can point to that you stole the work, but I'm arguing for that theoretically that these damages are there. 103 00:15:38.336 --> 00:15:48.996 That's a much harder le-legal case. So if you're a publisher, particularly digital publishers, that have not filed as many copyright registrations, you just don't have as much leverage for a deal, or for a lawsuit. 104 00:15:49.036 --> 00:15:56.096 Like, a deal might look interesting. And then what does the size of that deal look like, right? Like, why would OpenAI or any other company want to strike a deal with you? 105 00:15:56.136 --> 00:16:00.696 It could be dependent on what they would predict the size of a settlement could be. 106 00:16:00.716 --> 00:16:07.296 But I think the third option of just sitting it out completely, I think a lot of people are starting to realize is a little bit of a scary option. 107 00:16:07.356 --> 00:16:13.756 Because if you sit it out while all of your competitors are experimenting, you know, suddenly you're at a disadvantage. 108 00:16:14.196 --> 00:16:23.416 So it's a, it's a complicated decision for every sort of publisher, but everyone, everyone has to consider at least one of the three options. You can paint a fairly optimistic picture for publishers. 109 00:16:23.456 --> 00:16:34.516 It says if you make quality content that is worthwhile, and you can do these deals like we did across a number of people, that looks like a very good outcome for us. 110 00:16:34.576 --> 00:16:41.156 Look, I would much rather- Well, as long as people don't go to-...an economic solution to this problem than a legislative solution...you know, just get the answers instead of going to the page. 111 00:16:41.216 --> 00:16:51.896 Isn't there an obvious- Well, again, again, like, let's... You're, you're making a, a, a long leap that says they're gonna just take our content, and that's not really how it works, right? I... 112 00:16:52.255 --> 00:17:00.036 What we want is we look at the situation, we're like, "What do we want the outcome to be?" We want a very favorable economic solution with people who have already taken and are using our content, essentially. 113 00:17:00.576 --> 00:17:05.636 If we can get to an economic solution, great. We've got to an economic solution with OpenAI. We're talking to a few other people. 114 00:17:06.036 --> 00:17:14.326 If we cannot get to a economic solution with you, then we have to weigh all of our options. And I think you're gonna find exactly what Sarah just said. 115 00:17:14.936 --> 00:17:27.776 I, I don't think you're going to find publishers sitting on the sidelines and waiting too long to actively do something. That has not historically been a good play. Like we say all the time, we want to happen to things. 116 00:17:27.796 --> 00:17:36.416 We do not want things to happen to us. You have to be out front of things, and you have to, or you're just gonna get left behind. Yeah. Yeah. A-and just a few nuances to consider. 117 00:17:36.836 --> 00:17:46.546 So one, Neil's business is very much built on evergreen content. That can be really good for building your... training your LLM on backlog catalogs or, you know, back catalogs. 118 00:17:46.976 --> 00:17:54.486 If you are mostly real-time content, like live news and sports, that's gotta be a very different type of consideration and deal because the, uh, the... 119 00:17:54.516 --> 00:18:03.796 what you're then doing is you're striking a deal for your information to be used in an output, not necessarily training the LLM, okay? This is totally different deals. 120 00:18:04.336 --> 00:18:08.796 Then you have to consider the type of content that's valuable for an LLM. I'll give you a weird example. 121 00:18:09.276 --> 00:18:25.416 A lot of video companies in news have some interesting leverage because when you are Sora or any other kind of generative AI video company, you need to train your LLM off of videos of things that sometimes only uniquely exist in news content. 122 00:18:25.596 --> 00:18:30.956 Great example of this is a tornado. Any movie you are watching is not a real tornado, okay? 123 00:18:30.976 --> 00:18:42.648 That's a fake tornado.But only footage of real tornadoes is really mostly in news archives so they want access to that kind of stuff because they need to train their LLMs on the real thing. It's very strange, right? 124 00:18:42.888 --> 00:18:54.258 So like suddenly-- Oh, another thing with news publishers and video, these archives tend to be offline. They can't be scraped. Yeah. So people have weird leverage than Spanish language broadcasters. 125 00:18:54.808 --> 00:18:58.628 Interesting le-leverage. They need Spanish language content to train off the LLMs. 126 00:18:58.988 --> 00:19:08.068 So each model, each publisher has such wildly different advantages and disadvantages, incentives and disincentives, which is why the Barry Diller coalition was hard to do. 127 00:19:08.548 --> 00:19:19.888 And I think that's why you're gonna see these deals evolve so dramatically over time, because we're still in real time figuring out what the parameters here are. A good example, the AP struck a deal with OpenAI. 128 00:19:19.948 --> 00:19:28.317 The AP is a licensing business for decades. They're so used to licensing content, it's not as weird... Sorry, guys. Everyone's looking at me because the awnings are going back and forth. Oh, yeah. 129 00:19:28.408 --> 00:19:37.588 No, there's a battle of the awnings below. [laughs] But we're good. We're good. But the AP, like, they've never been in a, a traffic ad-supported business. They've always been about licensing. 130 00:19:37.628 --> 00:19:40.408 So striking a licensing deal- Yeah... is, like, more natural to them. 131 00:19:40.448 --> 00:19:49.378 But, you know, as part of that deal, if another publisher brokers a similar deal with better terms, like, I think their deal says that they can go back to the drawing board and renegotiate. 132 00:19:49.868 --> 00:19:58.328 So I think you're gonna have a lot of these, like, optimization clauses to protect people, and no one deal is gonna look like the other, and some deals will work out great, and other deals won't work out great. 133 00:19:58.588 --> 00:20:06.188 But there's no, like- So what is, what, what do you hear in your reporting from, about why New York Times is, is really the one prominent that's, that's suing? I mean, there's- It's the statutory damages. Okay. 134 00:20:06.208 --> 00:20:16.628 It's because the w- this is my understanding. The way that we file copyright claims in the United States is very cumbersome for digital publications. We do it. We copyright everything. Yes. 135 00:20:16.648 --> 00:20:22.858 But we're big, and we have a lot of capital. Yes, and also you have a lot of print. Yeah. Print is easy. You just mail it to the copyright office. 136 00:20:23.348 --> 00:20:35.228 So w-what the Times has done is they've filed millions and millions and millions of registrations. When you go to court, you have something that you can actually point to to say, "They ripped that off of me." 137 00:20:35.728 --> 00:20:40.968 And that's a much stronger legal case than trying to argue that for fair value. 138 00:20:41.468 --> 00:20:50.848 So the Times, I think, believes that they have a very strong legal case to win damages, statutory damages claims in court, which could amount to billions of dollars. Yeah. And that's why I think they chose to sue. 139 00:20:51.328 --> 00:21:02.548 S-so- Can I add one thing? Yeah, go ahead. And, and this is like all these conversations that we have, and I have them, and she has them, and you have them, like, everyone speaks in such definitives. I do. We all do. 140 00:21:02.608 --> 00:21:12.657 I-- But I can't emphasize enough how little anybody knows about anything, and we're, we're very close to all these guys making these products and making these properties. Mm-hmm. 141 00:21:12.668 --> 00:21:20.028 And we believe they're honest brokers being hon-- Like, every three weeks, what is happening there is completely different. 142 00:21:20.068 --> 00:21:33.788 Like, n- the, the guys running the biggest AI things have no idea what they're gonna be doing in six weeks. They say they do, just like I say I do. I don't either. Like, we're gonna have to figure this out on the fly. 143 00:21:33.828 --> 00:21:45.508 You're going to have to have relationships with these businesses. You're gonna have to believe the incentives, the contracts, the humans are somewhat good actors, or- Yes... it's gonna be bad. 144 00:21:46.048 --> 00:21:50.488 So how is your relationship with OpenAI different than, like, say, your relationship with Google? 145 00:21:51.248 --> 00:21:58.668 Uh, I'm not sure how to answer that other than saying, like, we don't have a relationship with Google like this- That's what I mean... in, in any way. [laughs] So, so- So I mean, is this a positive? 146 00:21:58.678 --> 00:22:05.108 'Cause I mean, I think one of the interesting things is- I mean, our relationship with Google is, like... It's not even a zero. It's like a null set on this- You're a partner of Google in some ways. 147 00:22:05.168 --> 00:22:13.448 I mean, not, like, directly, right? Because, like, you and Google... I mean, Google is distributing your content. You are providing... There is a value exchange there. 148 00:22:13.458 --> 00:22:22.028 Well, the historical value exchange was Google can have your content- Yeah... to build a search page that they can sell ads on, and as, and as a result, we get traffic. Yeah. That is the historical bargain. 149 00:22:22.228 --> 00:22:28.478 A-and, and, and they said, they would say- We're fine with that bargain. Yeah, and we're gonna have a church and state, and we're gonna say an algorithm and a black box. We're fine with that bargain. Okay. 150 00:22:28.478 --> 00:22:39.608 The, the bargain that we're not fine with is they've changed the bargain. Now they're taking all of our content to compete with us, right? Eat what we eat, then eat us. Like, that... 151 00:22:39.768 --> 00:22:51.108 If that is what happens, that's the risk. That's what we have to protect against. That's what, why we're doing deals with OpenAI, so that doesn't happen. Now, that being said, I go back to, like, the math is mathing. 152 00:22:51.688 --> 00:23:00.668 We haven't seen any impact to OpenAI on our traffic at all. Again, they, they do stuff above the search page. Maybe we're, 'cause we're in the links in the box because our stuff's really good. 153 00:23:00.748 --> 00:23:10.318 Like, uh, like, there's a lot of people blaming AI for things that are not happening. It's an easy cover. It's an easy cover. Yeah, that's what it is. The bad operators are blaming AI for things that aren't happening. 154 00:23:10.508 --> 00:23:18.168 Yeah. Nothing's happening with AI yet that's gonna affect anybody's business. Google may have made other changes that if you rely on search affect your business, but it's not AI yet. And, and- Yet... 155 00:23:18.517 --> 00:23:30.888 and, and as, as much as, you know, the, the relationship with Google has to be figured out going forward, and I'm neither positive nor negative on it, we'll see what happens, the, like, the path is, is, is, like, all to be determined on all of this stuff. 156 00:23:31.648 --> 00:23:41.508 So just to go back to that, I was talking to somebody who works at a publisher that struck a deal. They might not have full visibility into every little term, but I think they are pretty read in. 157 00:23:41.628 --> 00:23:50.208 One of the things that I was asking them about is, is your metadata, the IP from your metadata, part of your deal? They're like, "I have no idea what you're talking about, and I doubt it." 158 00:23:50.228 --> 00:24:00.618 And I'm like, "Well, something I've recently come across is that when you're a publisher and you publish an article, you tag it with, like, 12 different tags. So let's say it's Taylor Swift concert. 159 00:24:00.648 --> 00:24:08.488 You might tag that Travis Kelce, Taylor Swift, Barcelona, where the conference- concert is- Mm-hmm... music, fun, social. 160 00:24:08.568 --> 00:24:18.548 Like, all of those tags are actually viable IP because you're telling a story through the metadata. That can be part of training an LLM." "Have you had that conversation?" 161 00:24:18.558 --> 00:24:21.828 "No, because we didn't even realize that that was something valuable to us." 162 00:24:21.848 --> 00:24:29.028 And I think we're gonna start to learn more about this as we go forward, but the, the, the tension is these contracts are, like, multi-year deals. Mm. 163 00:24:29.128 --> 00:24:40.828 So that's why I think they need to be designed with, like, outs, if you will, or clauses that we can go back to the drawing board as a part of these multi-year deals to be flexible because we're all, including the AI companies, like, learning new things all the time. 164 00:24:40.888 --> 00:24:43.984 Can't, you can't-Legislate the future. And- Yeah... 165 00:24:44.024 --> 00:24:54.174 and, like, there's w- there's a hundred percent chance that big parts of the deal we did with OpenAI probably won't work in a year or two, but you just have to believe we're gonna figure it out because we're all adults and that's what you have to know. 166 00:24:54.184 --> 00:24:59.294 Do, do you anticipate licensing to be a much bigger part of your business? I know you have licensing with, with, with Meredith. I'm not... 167 00:24:59.304 --> 00:25:06.644 Like, not like Better Homes and Gardens at Walmart, which we are, I think, the biggest licensee in Walmart. Okay. You know who will, though? Southern Living at Dillard's, that's good stuff, too. Okay. 168 00:25:06.684 --> 00:25:15.704 I like the Dillard's. The people who will... I interviewed Linda Yaccarino yesterday. Yeah. And she was saying that right now they're using X data to train Groq. Yeah. 169 00:25:16.224 --> 00:25:28.194 But one day, they definitely see a big revenue stream in licensing X data for other LLMs. And so it's interesting, if you have a ton of data, that could be a really interesting play for licensing. 170 00:25:28.204 --> 00:25:34.584 Well, like, assuming this OpenAI deal- Cash is cash. You can call it whatever you want. They're, they're paying you. OpenAI is paying us something. Right. That's licensing to me. Well, what do you wanna call it? 171 00:25:34.624 --> 00:25:37.374 Like- It's licensing. That's what I mean. Great. Then it's licensing. Like, that's licensing. 172 00:25:37.404 --> 00:25:45.784 And so your business was, yes, you have subscriptions or print subscriptions from, from the Meredith properties, primarily an advertising business. 173 00:25:45.824 --> 00:25:55.124 What I'm saying is, like, are you now going to have a licensing business that is growing as the advertising business? Maybe it keeps... 174 00:25:55.244 --> 00:26:01.874 Let's say it keeps growing because you're doing lots of different things, but that licensing becomes a big line item on most balance sheets of publishers. Yeah. 175 00:26:01.884 --> 00:26:07.484 Yeah, again, I'm not an accountant, so I'm not sure exactly where it's gonna sit, but yes, that is the goal. That, that should be the... 176 00:26:07.524 --> 00:26:35.714 For us, if we do it right, and, you know, when we have People and Food & Wine and Southern Living and Travel + Leisure and Verywell and The Spruce, and we have forty brands, if our brands remain really valuable and people keep coming to our various properties and the various places you can get it, from Apple News to our O&O websites to this and that to the other, and they come to our events like Food & Wine Classic we just had in Aspen last weekend, which was, like, a bonkers giant event, this... 177 00:26:35.744 --> 00:26:49.924 You can make a case this is a supplemental revenue stream for us in many, many ways. And if it is, I like our future. If it... Look, you can't be sentimental about how you make money. We, we are a... 178 00:26:49.943 --> 00:26:58.004 We grew, again, double digits the last two quarters, and our print business is probably, like, one-tenth the size of it was ten years ago. 179 00:26:58.064 --> 00:27:07.264 Like, our print business will be smaller next year than this year, but we're growing because you can't be sentimental about the medium. And if we're gonna make money this way or that way or the other way, that is great. 180 00:27:07.464 --> 00:27:14.104 We are gonna go to where the money is, a-assuming that, you know, we, we can continue our value proposition with consumers. Like, I don't... 181 00:27:14.144 --> 00:27:23.324 Sort of like people are like, "Oh my God, this licensing revenue," why is it bad? I don't get it. Like, it's good. It's, it's the world [chuckles] evolving to potentially the next thing. 182 00:27:23.864 --> 00:27:32.494 And we haven't seen any degradation in our core business, and all of a sudden we have the beginnings of this other revenue stream, so I'm like, "Oh, you can actually be pretty optimistic here." Can I ask you a question? 183 00:27:32.604 --> 00:27:46.264 I'm just genuinely curious. Because so much of your content is evergreen, do you worry that in- after your first deal, they will have obtained enough of your data that they won't need to renew? So it- it's interesting. 184 00:27:46.324 --> 00:27:56.584 I don't think that's gonna be an issue because the thing people don't understand about evergreen content that works is, I'm gonna, I'm gonna get these numbers slightly wrong, but say half of our traffic is from URLs that are more than five years old, right? 185 00:27:56.624 --> 00:28:06.884 Across whether it's Verywell in Health or Allrecipes or you name it as our parents. But that content has been updated hundreds and hundreds and hundreds and hundreds and hundreds of times. 186 00:28:06.944 --> 00:28:14.294 If you have health content, everything in our health corpus is touched at least once every three months. Everything. We've a hundred... I don't know how many articles we have. 187 00:28:14.304 --> 00:28:21.944 We probably have a hundred thousand articles on Verywell. Everything is written by a medical writer, reviewed by a doctor. Like, the amount of work that... 188 00:28:21.984 --> 00:28:26.544 Everyone assumes evergreen content is you write a recipe and you make money forever. That is absolutely not how it works. 189 00:28:26.964 --> 00:28:38.864 And we spend, we probably spend fifty, sixty, seventy percent, depending on the time of year, of our content spend working on old content, new images, new videos. You need a celiac version of this recipe. 190 00:28:38.944 --> 00:28:50.024 Like, all of this stuff. And that is our value and investment. So our belief and our bet, and I think one of the things we got OpenAI to understand, is that there is no set it and forget it in content. 191 00:28:50.064 --> 00:28:56.464 There are a lot of evergreen guys that are gone now because they stopped doing this. Like, we have never slowed down our investment. 192 00:28:56.544 --> 00:29:05.584 And look, I think th-the reason why we're so out front about we need to be compensated for the work we do is because we need to be compensated for the work we do. We do... 193 00:29:06.004 --> 00:29:16.584 Like, these articles on health cost thousands and thousands of dollars to maintain. If it's an article about, you know, some deep symptom of diabetes that you have a bunch of doctors writing about, it's really hard work. 194 00:29:16.644 --> 00:29:20.744 And so a-anyway, it's a very long answer to I don't think it's gonna be an issue. 195 00:29:21.444 --> 00:29:31.744 When do, when do you expect or do you not expect or you just don't know to see the impact of AI becoming part of the search experience on traffic? 196 00:29:31.794 --> 00:29:39.454 'Cause it seems pretty clear that, you know, zero-click searches have gone up historically with Google, and this will just accelerate that trend. 197 00:29:39.884 --> 00:29:50.064 And then if you look at, like, a Perplexity, I don't think Perplexity's gonna last, but Google's gonna copy Perplexity because it's a, it's a pretty good experience. I, I don't... 198 00:29:50.524 --> 00:30:00.224 R-right now, for those people who, like, follow Google closely, it's very answer boxy, right? Remember the answer box? You know, what is Lincoln's birthday- Yeah... and it tells you in a box, right? Yeah. Okay. 199 00:30:00.444 --> 00:30:09.024 But then the answer box started to come to more complicated things, and they would summarize it and... but you'd still get the link in the answer box. There are links in most of their AI answers. 200 00:30:09.044 --> 00:30:17.364 Not all, but most right now. And it seems to be if people are... You're getting enough clicks in those links that for us it seems to be, like, net neutral in some way, I guess. 201 00:30:17.404 --> 00:30:23.284 It's very hard to track because the pages are changing every day. And I, I don't know. 202 00:30:23.384 --> 00:30:33.064 I think Google is gonna have to make a bunch of decisions around that search page, because if you were my dad, that search page is very confusing right now. There's, like, nineteen different things going on it. 203 00:30:33.104 --> 00:30:38.574 There's, like, product galleries going this way and AI at the top, and then, uh, you just want your listing of links, you can't even find... 204 00:30:38.574 --> 00:30:45.774 There's three of them, and then there's, like, a Reddit th- like, ah, what's going on? So I, I think it's all gonna have to find its own level. On the Reddit thing, I... 205 00:30:46.684 --> 00:30:57.882 Reddit would tell you that-They see their value on the Google search page as being hu- real human experience and conversation- I'm sure they do... which is the, uh-- of course, right? 206 00:30:57.892 --> 00:31:08.802 Which is the opposite of these AI summaries, right? Like, uh, does this s-skincare make my face itch, right? Like, you want real people talking about that as opposed to, like, AI summarizing it. 207 00:31:08.802 --> 00:31:11.992 You know, but what Google has done- [laughs] I know. This is the only thing. This is my one qualm. 208 00:31:12.072 --> 00:31:21.372 Google has spent the last twenty years saying, "The search page is this precious place, and we're only gonna put the most credible domains and the most credible writers with the most credible content on the search page." 209 00:31:22.052 --> 00:31:35.812 Then they have individual Redditors that know nothing about nothing giving skincare advice ranked ahead of wem- WebMD and Verywell, who have doctors and nurses and skincare professionals talking about the exact reason the same thing is happening. 210 00:31:36.152 --> 00:31:44.762 Actually, it's not credibility. Google cares about relevancy to your query, and the reason I say that is because relevancy is not necessarily credibility. 211 00:31:44.812 --> 00:31:59.432 In the news example, they made ch- big changes in, like, twenty nineteen to how they ranked news, but what would happen is the New York Times or whoever would write a twenty-thousand-word p- story, and at the bottom paragraph, it would say some revelation. 212 00:31:59.512 --> 00:32:09.132 I don't know, "This new flower is purple." And HuffPo would come in and do a one-hundred-word piece with a big headline, "New Flower is Purple." 213 00:32:09.142 --> 00:32:18.852 And when someone would do a search, they'd say, "What color is the new flower?" Google, from a relevancy perspective, would serve them the HuffPost first because it was easier for the consumer than The New York Times. 214 00:32:19.212 --> 00:32:29.552 Then they shifted the algorithm to put a priority on original reporting. But y- I'll tell you something, it doesn't provide as relevant in some cases of a search experience. So that's actually what they're weighing. 215 00:32:29.572 --> 00:32:35.572 It's not necessarily- Well, yeah... credibility, it's relevancy. Well, they create the incentives, and people respond to the incentives. I mean, if you don't like- An incentive-... 216 00:32:35.652 --> 00:32:43.052 the, the, the backstories of the recipe pages, don't blame the publisher, blame Google because- Well, well, here's the thing, though. Google's incentive is to make it 217 00:32:43.952 --> 00:32:52.292 the, the easiest as possible to serve the consumer, and the c- exactly, the consumer does not necessarily want that t- whatever it is, twenty-thousand-word essay. 218 00:32:52.692 --> 00:32:58.232 That's what they're gonna elevate, and so I don't think this is them saying, you know, "This Reddit post is more credible than yours." 219 00:32:58.332 --> 00:33:04.102 I think that that's them saying, "We're more likely to get someone to click and engage, and we can sell ads against it by putting it up there." 220 00:33:04.172 --> 00:33:11.812 Coincidentally, they did a deal with Reddit, like, a lo- uh, right around that time they started to show up really prominently in search, didn't they? Yes. 221 00:33:11.852 --> 00:33:19.572 Well, it's int- it was a sixty-million-dollar deal or something like that. Yeah, I'm just saying they are truly business partners, but- Yes. Maybe some reporter should- Yeah... look into that. 222 00:33:19.672 --> 00:33:28.412 Maybe someone should look into that. Uh, it's interesting. So you... L-let me ask you, do, do you use Perplexity at all? I've used it, yeah. Yeah. What do you think about it? It's definitely interesting. 223 00:33:28.832 --> 00:33:33.072 Are you following the Forbes situation? Yes, I am. That's why I'm wondering, like- Wait, I don't, I don't know the Forbes situation. 224 00:33:33.082 --> 00:33:42.672 So Perplexity basically ripped off a Forbes article to do one of their own little AI-generated articles. Oh, yeah, I saw that. Forbes, y- I reported this, they basically threatened... 225 00:33:43.172 --> 00:33:50.692 did all but threaten real legal action. They gave them like ten days to respond to whatever their queries are- One of those... saying, like, "You, you can't do this anymore." 226 00:33:51.072 --> 00:34:02.252 But the thing is about all these AI companies is, like, h-how scared should you really be, right? Like, do you really believe this legal threat from these publishers? No, I don't think so. They, they, he, he gave... 227 00:34:02.442 --> 00:34:09.932 The, the CEO responded. Yeah. He's like, "Oh, this is a work in progress." But, like, they're not gonna stop what they're doing because they got a letter from a publisher. Agree. Yeah. 228 00:34:09.992 --> 00:34:24.822 The risk/reward profile for, I don't know Perplexity enough to call them out specifically, but for that ilk is the reward is way greater than the risk, so just do it. Yeah. And- Forgiveness, not permission. 229 00:34:25.032 --> 00:34:33.252 Yeah, and a lot of it is just, like, you know, very valley platform mentality, content is worthless. 230 00:34:33.312 --> 00:34:43.692 We are going to provide an algorithm and a system that makes everyone's lives better without regard to where this stuff came from or without regard to where this content came from, and that's gonna be a problem. 231 00:34:43.892 --> 00:34:51.162 So why do the deal with OpenAI then? I mean, you said that they're, that they're... I mean, won't, won't they just do the same thing that has happened repeatedly? 232 00:34:51.312 --> 00:35:00.332 I mean, we used to use this GIF of Charlie Brown pulling, getting the football pulled out from under him- They, ugh... a million times for every story about Facebook and publishing. 233 00:35:00.452 --> 00:35:20.912 You, you can be paralyzed by fear and not making decisions, or you can have relationships with people that are saying the right things, acting and doing the right things, that are doing fair economic transactions, that you believe their intentions, and you move forward, and that's what we did. 234 00:35:20.972 --> 00:35:31.762 I actually... OpenAI's been great for us so far. Like, I'm... We're super happy with it. Anything can go wrong at any time, but you cannot run a publisher, and again, uh, we're not news, we're not... 235 00:35:31.792 --> 00:35:37.192 You cannot be scared. You have to be aggressive. You have to defend what is yours. 236 00:35:37.232 --> 00:35:49.092 A-and look, you need brands that get you proprietary audiences that will end up in proprietary data, in quotes, and then you can have a business. And if you can have those three things, you are going to be great. If... 237 00:35:49.172 --> 00:36:00.852 This is all in service of that. If, if I were to, like, you know, bet with you, it... Like, do you expect to be making most of your money from bringing people to webpages and putting ads in front of them in five years? 238 00:36:01.792 --> 00:36:14.812 I, I mean, like- Or you have no idea? Like, we sat eighteen months ago with the OpenAI guys before anything was public, and we saw this th- like, saw this thing, and the world is so different eighteen months later. 239 00:36:15.242 --> 00:36:24.472 I have no idea what's gonna happen eighteen months from now. I have no clue. I, I actually... And this is what you have to get your organization to know. Like, I don't care. 240 00:36:25.032 --> 00:36:38.452 As long as we're smart and flexible and we can figure it out, if the world were just sending people from Google to our webpages at an increasing rate in a fairly good economy, that would be amazing, but that's not what's happening. 241 00:36:38.492 --> 00:36:50.272 We still manage to do well in a world where everything's changing. That's your job as a publisher. It's freaking hard. Every day is a battle. You're at war. War is a too strong a word. You're battling somebody. 242 00:36:50.522 --> 00:36:59.410 [laughs] It's not war. You're battling somebody every day for something, and somebody-Is always trying to take your business or take margin out of your business because they think that they can. 243 00:36:59.720 --> 00:37:02.240 You always have to skate to the puck. Absolutely. That's the way I look at it. Yeah. 244 00:37:02.320 --> 00:37:14.820 And the puck is constantly moving, and you have to have a low ego and be willing to say, "Okay, what I've been doing professionally for the past five years, what I've trained myself, what's easy, what's now low margin work because I know what it's like, I can't do-- I can't do it anymore. 245 00:37:15.010 --> 00:37:20.300 Doesn't work. And now I have to do this whole other thing." And you have to just get comfortable with doing that. 246 00:37:20.340 --> 00:37:28.720 And some of these whole other things you might not like, but that is the reality of being in this business. If you don't like that, then you should not be in this business. And I'll give you a good example of that. 247 00:37:28.770 --> 00:37:38.180 I mean, I, I, like, I want to print out what you just said, and I wanna hand it to every single person I can. [laughs] Because that's exactly what people have to understand, and nobody understands it. 248 00:37:38.260 --> 00:37:54.560 Everybody just complains. Yeah, I mean, a good example of that. Right now, I see the digital advertising business as a barbell. I think you either are high impact brand sponsoring a beautiful villa in Cannes. Congrats. 249 00:37:54.590 --> 00:38:00.099 Right? That's you. Wonder what that looks like or you're p- you're just pure efficiency, right? 250 00:38:00.120 --> 00:38:09.430 You're a walled garden, and you are helping people c-click to complete a sale, and everything in the middle is kind of getting bottomed out. Well, you're kinda doing both. You're the whole barbell then, right? 251 00:38:09.430 --> 00:38:18.100 I mean- Well, that's the thing. You can get-- do the both sides of the barbell. I, I, I think we are both, but it's-- we don't have a single brand that's in the middle of that barbell. 252 00:38:18.400 --> 00:38:23.160 We're-- every brand's at one end or the other. E-exactly. Yeah. You have to go... There's no middle anymore. 253 00:38:23.560 --> 00:38:33.260 That middle, that n- you know, twenty dollar run of site CPM, like, that's getting really hard to compete on because you're not efficient enough, and at the same time you're not high impact enough. 254 00:38:33.300 --> 00:38:43.020 And so in terms of skating to the puck, if you don't like moderating events, that's a real big problem for you right now as a content creator because so much energy and momentum is there. 255 00:38:43.100 --> 00:38:50.620 If you don't like doing that, you gotta find something that is sponsorable or we can monetize. Maybe you're amazing at data analysis. 256 00:38:50.660 --> 00:38:54.470 Okay, can we put some of your data analysis behind a paywall, and we charge niche subscriptions? 257 00:38:54.480 --> 00:39:04.300 But, like, whatever it is, you as a journalist, you as a content creator have to figure out, like, how can we-- how can you skate to a puck that's monetizable, and if you are unwilling to do that, you have a big problem. 258 00:39:04.469 --> 00:39:09.680 So I remember when I first started coming to Cannes, it was, like, all in agent- all agencies, right? It was about the creative work. 259 00:39:09.720 --> 00:39:17.660 It was-- it would actually go until Saturday, which is insane, which is when the, like- Oh, my God. The fireworks on the beach thing? Yeah, the TV commercial at- I'm old too... 260 00:39:17.740 --> 00:39:25.320 awards, what they would call film, would- [laughs] Which is-- I would always joke, I was like, "If your ad agency is calling it a film, you paid too much." 261 00:39:25.860 --> 00:39:34.620 But now it's like the grocery store chain is, like, out front [laughs] Retailers- It's amazing... financial services companies. Yeah. Anyone that has attention- [laughs]... 262 00:39:34.660 --> 00:39:44.290 will marry that with first-party data, and that's y- that's the winning formula. Are you competing with, with retail media? And is that, like- We're-... friend or foe? Or how do you look at that? 263 00:39:44.320 --> 00:39:55.860 So far it's, it's been friend, friend-ish. What I would say is it's a-- it's for us, it's another demand source to tap into, but it's also another place that gets between you and your clients' money. 264 00:39:55.900 --> 00:40:06.280 But by and large, retail media networks for us have been excellent. They're very data-driven. They're very smart. There's only, like, a couple of them that are, are scaled enough to matter right now. 265 00:40:06.880 --> 00:40:18.530 But because we're, like, evergreen, that gives us... most of our stuff is very high, very intent driven- Yeah... very performant, and we're the biggest affiliate referrer to p-probably every retailer in America. 266 00:40:18.600 --> 00:40:27.140 I might be off by one or two. So we're very good at commerce, so we tend to do really well with retail media networks because we kinda understand what makes consumers buy stuff. 267 00:40:27.180 --> 00:40:37.680 So we've been happily coexisting with them. But so the, the-- to me, the impact of a retail network, and again, the other ones I mentioned, is that they are contributing to the leveling out of the middle of the barbell. 268 00:40:38.150 --> 00:40:47.280 Like, they're adding to the efficiency pressure- Yeah... so that it doesn't make sense anymore to be in the middle. Yeah. So final thing is, like, I mean, you have a very intermediated business, right? 269 00:40:47.440 --> 00:40:51.400 Like, I mean, at the end of the day, you're intermediated on the distribution side. 270 00:40:51.980 --> 00:40:59.480 You know, you won't give me the figure, but let's just say it's seventy-five percent 'cause you've got Google that's fifty percent and then a bunch of other algorithms that are intermediating. 271 00:40:59.920 --> 00:41:07.040 People are not-- most people are not typing verywell.com into their browser, like, in the morning, right? People.com. People.com, yes. Okay. 272 00:41:07.580 --> 00:41:21.160 But and then on the other side, you're often intermediated from the clients, right? You've got The Trade Desk, any DSP. You've got agencies. Like, you know, you came from outside of the publishing world. 273 00:41:21.220 --> 00:41:35.200 That, that seems, like, not ideal. It, it's not ideal. I think we, we, um, fundamentally understand better the advertising side of it because you can correlate that with brands, like, performance, data. 274 00:41:35.240 --> 00:41:47.020 Like, that stuff we, we really can play well with. And there is always disintermediation in some form between you and your clients. It could be the agency. It could be the RMN or retail media network. 275 00:41:47.040 --> 00:41:56.240 It could be whatever. There's a million people trying to get between you and your money, and, you know, you put a dollar into the programmatic universe, and you get fifty cents of buying power at the end of your buy. 276 00:41:56.330 --> 00:42:01.460 So there's a lot of stuff going on in there, but I think we understand it. We've dealt pretty well with it. 277 00:42:01.560 --> 00:42:11.320 I think the disintermediation on the supply side, I don't think it's very different than it's been for the last decade. I just think that you have to be-- it's, it's exactly what... 278 00:42:11.380 --> 00:42:18.890 Interviews are boring when everyone agrees, but I agree with Sarah. It's, it's what everyone has to understand. You cannot be sentimental, and just because you did really... 279 00:42:19.460 --> 00:42:26.920 Facebook sends zero visitors to publishers now. None. Zero. For some of our sites, they were twenty percent of our traffic. 280 00:42:27.080 --> 00:42:35.440 We're still growing traffic in aggregate year over year because you have to know what is next and do what is next. And, like, we live in a world... 281 00:42:35.460 --> 00:42:42.739 And I, I think on the, on the ad side, the, the incentives are a little bit more aligned, right? Performance, this, brands, this. 282 00:42:42.840 --> 00:42:51.840 On the other side, on the, on the sup-supply side, you gotta understand what everyone's incentives are, and Facebook's incentives have absolutely nothing to do with the publisher's health and wellbeing. Zero. 283 00:42:52.440 --> 00:43:02.290 So knowing that going in, you just have to set up your business so you can respond as things change. And it's part of why this is, like, work every day and a struggle every day 'cause something... 284 00:43:02.320 --> 00:43:09.390 Everything feels great, and then, like, Tuesday, something ridiculous happens. You honestly have to just enjoy the chaos. Like, that's the only way to make this good, to make this work. 285 00:43:09.680 --> 00:43:14.680 I'll tell you a tension that I'm really dealing with because there's no f- more Facebook traffic to news. 286 00:43:14.700 --> 00:43:25.807 When I was writingFive, six, seven years ago, the type of content that was going bananas for me was when I would do like a media analysis of Trump or something like that. Hmm. Nowadays, like that's so hard... 287 00:43:25.948 --> 00:43:37.408 It's still important work civically, so I will do it. But in terms of monetizing that like heavily trafficked story, there's less business reason to do that. Where is there a business case for me right now? 288 00:43:37.828 --> 00:43:52.307 It's like serving my niche B2B marketing, uh, and publishing audiences with insights, like charts and analysis- Mm-hmm... on the ad market. That is two very different jobs, okay? 289 00:43:52.368 --> 00:44:03.087 One is me getting on the phone with people in Mar-a-Lago, and the other is me putting my brain in s- Excel spreadsheets, and I have to be completely indifferent to either, and I'm okay with that. 290 00:44:03.207 --> 00:44:09.887 But I think a lot of reporters would not be. Yeah, that's true. Awesome. This was awes- this was really good. I'll say, and you and I have disagreed a lot about things in the past- I know... 291 00:44:09.887 --> 00:44:16.907 not just like Philadelphia Eagles topics. Yeah. I'll say this has been, and I, and I really mean this truly, sincerely, the best conversation I've had all week. Oh, good. 292 00:44:16.947 --> 00:44:26.307 Because you guys are informed, level-headed, thoughtful- Sober... not sensational- sober-ish. Like s- Sober for now. S- We're at a villa with a ton of wine. 293 00:44:26.367 --> 00:44:38.827 But, but like I wish everybody could think about our business in this way. We'd be in a much better place. Like- I agree with that... the math is mathing. Like it's not that hard. It's not that complicated. All right. 294 00:44:38.927 --> 00:44:48.047 I'm either going between the math is mathing for the, the title, or what is it? Enjoy the chaos from Sarah. Well, A, B test. I, I like enjoy the chaos. [laughs] I like it too. Awesome. Thanks, Brian. Appreciate it. 295 00:44:48.087 --> 00:45:01.027 Thank you so much. Thank you. Thank you all for coming. [outro music]