Exclusive: Jonah Peretti Explains Why He Sold BuzzFeed  🔗 **Source:** [Exclusive: Jonah Peretti Explains Why He Sold BuzzFeed](https://share.snipd.com/episode/c6f706e5-8cae-4185-ab76-b624310b23e4) ## Highlights - **Social Mobile Created A New Medium For BuzzFeed** - BuzzFeed's rise was driven by social + mobile becoming a new medium that enabled sharing as the core of media. - Jonah credits early success to understanding platform dynamics that made content a social connector, not just a broadcast piece. Transcript: Nilay Patel Hello and welcome to Decoder. I'm Neal I Patel, Editor-in of The Verge, and Decoder is my show about big ideas and other problems. Today I'm talking to Jonah Peretti, who is technically still the CEO of BuzzFeed, although that will be coming to an end very soon. Just days before we spoke, Jonah agreed to sell 52% of BuzzFeed for a total of $120 million to Byron Allen, who owns The Weather Channel, a number of broadcast stations, and several other Websites. The deal is a bit of a life raft for BuzzFeed. The company was once valued at $1.6 billion. But just last quarter, the company told investors it was at risk of running out of cash. Now there's a new lease on life and new leadership. As part of the deal, Jonah himself is stepping down as CEO and taking on a new role as president of BuzzFeed AI. Byron Allen himself will become the new CEO of BuzzFeed. It's obviously a huge structural and organizational change and a really big decision. This is prime decoder bait. And of course, I've always been very interested in what digital media companies are doing to adapt and survive in an information landscape dominated by algorithmic social platforms. After all, I've been saying for a long time that the original sin of digital media was Jonah and BuzzFeed betting they could so consistently go viral that platforms like Facebook would Pay them for content, just like cable companies pay carriage fees for channels like ESPN. This was the big bet for a lot of companies, all chasing BuzzFeed's influence and valuation, and it has all come crashing down. ([Time 0:00:33](https://share.snipd.com/snip/0a0c6c17-ef4e-467c-8dc7-8968b7dfda3a)) - **Deals Matter More Than Pure Tech Today** - The current media era favors deal‑makers over pure tech experts because distribution now relies on partnerships and ad/deal relationships. - Jonah says organic platform-driven growth is 'anemic' compared with the 2013–14 open distribution era. Transcript: Jonah Peretti I mean, I think we're in an era right now where deals really matter a lot. And so deal focused executives are having a lot more success than trying to just build organically. The connections with advertisers and the connections with partners and marketers and being able to see from their perspective and sell them things that they get excited about and Inspired about is really important. I think, you know, in the kind of peak growth period of BuzzFeed, kind of 2013, 2014, you know, in that era, really understanding the technology was the key to being able to build a good Media company because the technology was changing so quickly with the rise of social and social platforms, that, you know, you really needed to be deep on the tech side. And there was tons of organic growth available, like the platforms were very open in a way that allowed you to drive tons of organic growth. I think today, you know, if you look around the landscape, it's a lot more about deals and partnerships and business development and things like that. And that can be really transformative for a company. Whereas the kind of organic growth on the big platforms is really just, you know, quite anemic. ([Time 0:07:13](https://share.snipd.com/snip/aaaa819b-ed6e-4d21-961c-7c768f938642)) - **Platforms Briefly Subsidized Professional Content** - Platforms paid BuzzFeed millions early via rev-share programs for video and news, but those payments were short-lived as creator supply exploded. - Jonah argues platforms should have continued underwriting professional content to avoid today's toxic, attention-maximizing incentives. Transcript: Jonah Peretti So, I mean, first, just as a factual matter, we did get paid millions of dollars by Mark Zuckerberg. So the prediction that they would pay for content was accurate. It just was short-lived. So we got paid for that exploding watermelon. I mean, we were paid for... Nilay Patel But those are all those pilot programs, like stand up your Facebook Live team. We'll give you the seed funding to buy cameras. There was never actually payment for content. There was no ongoing revenue relationship. Jonah Peretti Oh, there absolutely was. We got paid rev share for video on Facebook. So, you know, tasty, the short, tasty videos that initially blew up, there was no payment for that. But then when Facebook realized they needed to make longer content so they could put mid-roll ads, you know, there were programs that generated millions of dollars for our company Where we were making longer form video for Facebook and we were getting rev share payments on that. Millions of dollars from YouTube. It's not that it didn't happen. It did. It did. It did happen. The news tab was another example of it. We got paid millions of dollars for putting news on Facebook. So I think what you're saying is accurate in the sense that it went from free distribution and sending traffic to our site to hosting content on their platform and getting paid for it And a rev share for it to then an explosion of creators, which made the value content and the platform's willingness to pay much lower. ([Time 0:11:40](https://share.snipd.com/snip/6566d86b-7dba-461e-91ae-d5dbfadafdee)) - **Exploding Watermelon Was Actually Paid Work** - Jonah recounts BuzzFeed being paid for the exploding watermelon program and Facebook news tab rev-share programs. - He argues those programs worked but were abandoned, which he sees as a strategic mistake by platforms. Transcript: Jonah Peretti I mean, we were paid for... Nilay Patel But those are all those pilot programs, like stand up your Facebook Live team. We'll give you the seed funding to buy cameras. There was never actually payment for content. There was no ongoing revenue relationship. Jonah Peretti Oh, there absolutely was. We got paid rev share for video on Facebook. So, you know, tasty, the short, tasty videos that initially blew up, there was no payment for that. But then when Facebook realized they needed to make longer content so they could put mid-roll ads, you know, there were programs that generated millions of dollars for our company Where we were making longer form video for Facebook and we were getting rev share payments on that. Millions of dollars from YouTube. It's not that it didn't happen. It did. It did. It did happen. The news tab was another example of it. We got paid millions of dollars for putting news on Facebook. So I think what you're saying is accurate in the sense that it went from free distribution and sending traffic to our site to hosting content on their platform and getting paid for it And a rev share for it to then an explosion of creators, which made the value content and the platform's willingness to pay much lower. And so it was happening and it was working and it was exciting. And I actually believe that it was a mistake for Facebook and Mark Zuckerberg to not continue with the news tab, to not continue to pay professional content creators. ([Time 0:11:56](https://share.snipd.com/snip/ffcac5d2-e720-43e2-9652-66438eac3cc9)) - **Prioritize Profitable Core Before New Experiments** - Restructure to build a profitable core before launching new initiatives so you can weather audience or ad-market drops. - Jonah plans to cut burn, set a profitable foundation, then layer new AI-driven products on top. Transcript: Jonah Peretti Mean, we're still a public company. It used to be we were a public company that I controlled through super voting stock. And now we're a public company that Byron controls through the 52 percent ownership. Nilay Patel So you've given up your super voting stock in this? Yes. That was actually my next question is where are the 40 million shares coming from? They're coming from you? Jonah Peretti No, we're issuing new shares. Okay. But you're dissolving your super voting shares. I just am converting my shares to just be low vote shares or normal common stock. Nilay Patel Great. And then of the 20 million, you said this is going to give you some operating cash. How much runway do you get out of 20 million? Jonah Peretti I mean, I think infinite runway is, you know, the, the, the, the, the plan. I mean, if you look at our cash burn, it's been, you know, not pretty, pretty insignificant. And I think we also announced on our earnings that we plan to do a restructure. And so I think, you know, setting the company up to operate profitably and to do that for the long term and then have a strong foundation that we can build on top of. ([Time 0:24:48](https://share.snipd.com/snip/80ec35d2-666c-49d6-acaa-236ba67615c5)) - **BuzzFeed AI Is Internal Tooling Plus New Apps** - BuzzFeed AI is twofold: augment internal teams (nervous system for what works) and incubate new AI-native apps via Branch Office. - Examples: BF Island (meme/messaging), Conjure (camera challenge) as experiments in AI media formats. Transcript: Jonah Peretti Of it is how do we use AI across the business as a whole? And I can share some things, but not everything. But part of what I think AI is really good for is it's almost can be like a nervous system that is able to understand and detect, you know, what's working and what's not working and how content Is being shared and engaged with. And some of the early stuff that BuzzFeed did about social content, I think you could do on this whole nother level when AI is able to actually understand the content and then push challenges To creative employees to make new things, but they're making new things with a lot more information about how people are engaged with the content. So just to put it in more simple terms, using AI to help our core business and help our people be more creative. So that's one part of it. Another part of it is building entirely new apps. And so Branch Office is an incubator that creates new apps. We just today launched BF Island, which is a new kind of messaging app that allows you to play and message and talk with your friends. They, you know, using using Gen AI. So it's kind of like an inside joke engine. We found that, you know, making AI content and posting it on Instagram or TikTok just kind of feels like slop. But making funny things with your friends that are about the, you know, specific things that you're joking about and talking about is just a lot more fun. And so we, we built BF Island for that use case. And then we also launched an app called Conjure, which is a, a camera app where you get a challenge every day to take a picture and kind of discover a mystery every day. You can, for example, take, take pictures to see if there might be Uapps hiding in the sky near where you live or get messages or fortunes or other things on Conjure. So I think when you think about BuzzFeed AI, it's a combination of things that integrate across the entire company and help our teams do their jobs better. And then it's also new apps and new experiments that allow us to test AI as a new medium that could potentially lead to new businesses or new apps that break out. ([Time 0:27:19](https://share.snipd.com/snip/f9f22b22-0326-416c-91dc-e409b0780bdf)) - **Writers Buying Code Created Hundreds Of Games** - Jonah describes how non‑engineer writers now 'buy code' to ship games, letting BuzzFeed produce hundreds of games quickly. - He credits toolkits that let writers create interactive experiences, driving time‑spent growth and audience love. Transcript: Jonah Peretti It's pretty, I mean, the, the, the BuzzFeed games have, have really exploded recently. And, and that's another great example of the fact that writers can buy code games. So instead of, you know, the New York times launches one game every, every, uh, every year. And, you know, we've done hundreds of games this year with, with creative people who are, who are writers and not programmers, having a toolkit that allows them to, to build in that space. And we've had a bunch of breakouts, you know, it's, it's become a big percentage of, you know, big percentage growth in time spent. And the comments on the, you know, from the audience are just like that. They love them and it allows us to iterate and evolve them more quickly and change them more quickly. ([Time 0:34:22](https://share.snipd.com/snip/5c0d25da-108e-4c24-8d24-5b54ed5176f7)) - **Direct Traffic Enables Launch Of New Products** - BuzzFeed's current distribution is largely direct, letting the company promote new apps across its properties rather than rely on platform referrals. - Jonah expects BF Island and Conjure to be marketed inside BuzzFeed to reach millions without platform gatekeepers. Transcript: Jonah Peretti Yeah, I mean, we have a majority direct audience now. So, you know, some of that is because of declines in other audiences, you know, but but tens of millions of people, you know, every month, you can look at our comm scores. You don't have to trust, trust me on it. It's like, we're the biggest publisher in our competitive set. You know, we're bigger than people. Um, we, we have a lot of people coming and when we see new formats, like for example, these, these Buzzfeed games that we have been creating, um, they come back more frequently, they spend More time, they engage more, they send the games using iMessage to friends. So you have this new kind of social that's more personal, that's more private. And so we have distribution that, like if someone was creating a new company and trying to do any of these things, it'd be very hard to achieve. We haven't really started promoting our branch office apps, these new AI apps, but soon Conjure will be all across BuzzFeed and BF Island will be all across BuzzFeed to market it and Let people play it. ([Time 0:36:28](https://share.snipd.com/snip/640e8e01-b66e-4694-ae5b-b60a8c8f50b7)) - **Launch Freemium Now And Ride Falling Model Costs** - Use freemium and iterate rapidly because model costs fall quickly; early subscribers may break even later become profitable. - Jonah notes AI model costs drop (Moore's Law analogy), so locking users now can pay off as compute gets cheaper. Transcript: Jonah Peretti I mean, I think this is pretty established that the freemium model for these kinds of apps is, you know, is, is, is pretty natural. People have shown a willingness to pay for AI services. And so I think- Wait, have one? Nilay Patel I mean, like, you know, Chattapete has, what, 900 million users and like very, very few of them pay anything. Jonah Peretti I mean, a lot of, you know, how much revenue is Chattapete generating from subscribers? I mean, you know, it depends- Not enough to pay their bills, I would say. But but there are a lot of smaller freemium apps and a lot of companies that, you know, have 5 million, 10 million in revenue and are profitable. And, you know, you might not hear about those because they're not big venture backed, you know, companies. But I do think that these kind of midsize and smaller types of apps can be really good businesses. And in some ways, it's a little bit like a new kind of media business. It's the economics might be more similar to, you know, a YouTube show or a, or a series or a podcast where, you know, that the, the cost of the, of the talent and the show and, you know, is, Is a, is a hurdle that the advertising or the direct or the subscribe subscription to clear. But I think there's a lot of examples of that that I've seen talking to other founders and getting a sense of that market. So I feel like there's a bunch of different ways to monetize. It's also reminiscent a little bit of the earlier era when you had Moore's Law making everything cheaper and you could have a product that lost a little bit of money, but if you wait a year, The processing is cheaper, the costs are cheaper, and AI is progressing incredibly quickly. So anytime we use a new model, you know, six months later, that the cost of it is like a 10th of what it was. And so if we could lock in a bunch of users to an experience that's not about creating the most beautiful images, it's about joking with your friends, getting ideas for what to make. And it's a, you know, BF Island is kind of an idea engine where there's all these cool contexts and things you might not have thought of that you can then use to delight your friends or play. The cost of that is making those images is going to and has been going down a lot. ([Time 0:42:47](https://share.snipd.com/snip/c764f345-bb60-41b3-95c2-4c0e878d3438)) - **Creator Independence Broke The Media Apprenticeship** - Creator disaggregation shifted value to top talent who can monetize independently, breaking the industry apprenticeship model. - Jonah says top creators leaving for Substack/YouTube removed the redistribution that funded new talent development inside media orgs. Transcript: Nilay Patel Yeah. Jonah Peretti I mean, when you think about BuzzFeed video, it was really awesome for the team that was there because they had all these collaborators. A lot of them came in as interns or fellows. The Try guys came in that way, you know, a bunch of, of the talent came in, you know, really almost like interns that became fellow got fellowships, and then became junior producers, And then, you know, kind of worked their way up. The challenge, I think, became when when when the platforms really started to, disaggregate the media companies, like essentially having individual creators create their own channels Without media companies. It meant that the very top creators could make more money if they left. And it kind of started to kind of break the cycle of development, because essentially, almost every media company, if you look over, you know, the last 100 years, there's some stars, And they get paid the most, but they get paid twice as much or something that, you know, or three times as much, but they're producing 10 times as much value. And, and then you have a lot of new, you know, new talent coming in, and they're learning, and maybe they're working their way up, but they're getting kind of subsidized by by the sort Of, you know, bigger talent. And so it creates this like natural tension in these businesses, where the whole process of having a community and an organization professional development and new talent coming In and, you know, like all of that kind of breaks if the top talent is like, we'll just do a sub stack or we'll just do a podcast that we own and take, you know, 90% of the profits from it or, You know, we'll just make our own YouTube channel, then, you know, the process kind of breaks where you don't you no longer have that like redistribution that sort of helps the next generation Learn the young reporter or the new video producer or whatever. And so I think early BuzzFeed was trying to build essentially like a collective in an organization. And at a certain point, some of the talent was like, I'm making too much money. I mean, I'm not making enough money. I'm going to go create my own thing. And, you know, that worked for some of them. A lot of them found that it really sucked to try to be a YouTube creator and having to call in friends and favors and not having resources and not having collaborators and the loneliness Of it and the burnout. And YouTube and TikTok love to give the trending designation to new talent, but then a year or two later, it's like they don't get that anymore. And so it really is kind of broken for a lot of the talent that was frustrated and left, like didn't work out that well. ([Time 0:48:19](https://share.snipd.com/snip/36a755f2-d574-4cea-a5d9-20c5669ddda2)) - **Use AI To Augment Journalism, Not Replace It** - Use AI to elevate journalism, not replace it; apply tools for research, cross‑referencing, and presentation while keeping human reporting for trust. - Jonah rejects AI-written news for HuffPost but supports AI-assisted reporting and interactive story formats. Transcript: Jonah Peretti I mean, I think that there's things that AI can do in journalism. I think it's, you know, the example of being able to, you know, buy code and create new games and new formats is one example. I don't think a HuffPost journalist would, you know, you remember back when like snowfall was this like big, you know, story. We did tons of snowfalls on the verge. Nilay Patel Yes. And they're big, complicated layouts. We love them. Yeah. Jonah Peretti Yeah. And, you know, that kind of thing is trivial to do now. Right. You know, you can you can design interfaces that, you know, are, you know, it's trivial from a technical standpoint. Like you don't necessarily need an engineer to do that. You need someone who has, you know, a good sense of design and good vision or whatever, but and you need the journalism to be good and the story to be good. But so I think a lot of it depends. It's like, how can you use AI to elevate journalism, to promote it better, to connect people to it, to research and find, you know, like crawling the web is something that, you know, how Many times have journalists spent three days going around to websites or whatever? And now with CloudCode, you can basically say, I need this data source, I need this data source, I need this data source, I wanted them all in folders. Okay, I want to cross reference them, you know, the kind of fact gathering and things like that that you can do is is something that could help elevate journalism and make journalism Better. ([Time 0:53:20](https://share.snipd.com/snip/5d1037a2-6f20-4785-84ba-926704367e6d)) - **Turning Down Disney Kept BuzzFeed's Independent Run** - Turning down Disney's 2013–14 buyout traded near-term financial upside for independence that allowed cultural influence like BuzzFeed Video and investigative hiring. - Jonah acknowledges a later sale might've maximized value, but values the independent growth period they had. Transcript: Jonah Peretti Million. Do you think he should have sold back then? So, I mean, from a financial standpoint, it's hard to say that that wouldn't have been a good deal to take. It was, I think, $450 million with an earn out that could get to that $650 million number. And, you know, we were just hiring Mark Schuess to build our investigative team. We were just starting with BuzzFeed Video and that whole cultural phenomenon of BuzzFeed Video was kind of just about to take off. There were a bunch of things that I'm really glad we got to do as an independent company. And I wouldn't really trade those years where we were able to, you know, build and explode and hack culture and, and influence the internet in a massive way. And I think it would have been harder to do that, you know, inside of inside of Disney, I think it would have been a great financial deal. And I do wish that, you know, my partners and investors and others had had been able to have that financial, you know, windfall, which, you know, you know, was unfortunate that they that, You know, that we weren't able to sort of live up to that. The other thing I would say is that I don't think selling the company at that moment would have been the maximization of value. I think selling it two years later would have been. And there was, you know, a lot of interest that people don't know about, you know, in those sort of subsequent years. And so the Disney deal was the first time someone came and I think Iger was ahead of the curve and saw that there was a lot of growth and excitement there. ([Time 0:57:40](https://share.snipd.com/snip/907d0741-2d86-407b-ac32-8236cb7179c7)) - **SPAC Timing And Complex Deal Hampered Cash Position** - The SPAC and Complex acquisition missed a hot window and left BuzzFeed with debt instead of cash, constraining later reinvention. - Jonah says buying Complex added debt; selling those assets and Byron's investment restores balance sheet to pursue AI strategy. Transcript: Jonah Peretti Well, I think the biggest issue was we missed the window. So again, if you had perfect hindsight, if you could do it over again, I think we would have gone public via SPAC and we would have done it without buying complex and we would have just done It right away because the market was super hot. We could have gotten a lot of capital into the business and we didn't need Complex. Like our business, you know, we were bigger with Complex, but our business was probably better. Our numbers were probably better without Complex than with. And so we could have just done the deal more quickly and had more cash on, on the balance sheet. And, and then we would have been able to have a lot more leeway and, and latitude to reinvent the company when the market got really tough. And so the, the, the problem is when, when we started to explore this complex acquisition, it slowed down our ability to complete the SPAC transaction because we needed to agree with, You know, on a sale and agree on all these terms and negotiate. I mean, with Hearst and Verizon, who co-owned it. And so and then we needed to then, once we went public, like figure out how to integrate and work with cultures that were not really well suited to each other. The complex culture was just very different than than the BuzzFeed culture. And so in retrospect, I think. Hitting this back market while it was hot, bringing in a bunch of capital, doing it with just BuzzFeed and then using that to reinvent the company as a public company, we would have been In a really exciting position and we would have had capital to do acquisitions and things like that. Instead, we missed that window. The SPAC market got ice cold. We were already kind of far along on the complex deal. And so we ended up with debt instead of cash. And so the debt to buy complex basically has been a burden that we've slowly been able to alleviate, you know, first by selling complex, then by selling first we feast and and now with This partnership with with with Byron Allen and that investment. ([Time 1:00:21](https://share.snipd.com/snip/23ceed7c-423c-4c8a-9df6-9fac855213d2)) - **Design AI Products Around Human Agency And Social Play** - Build AI experiences that prioritize human agency and social connection to overcome generational skepticism. - Jonah suggests apps that look like games or private sharing (WordChain, BF Island) will be better received than obvious 'AI' products. Transcript: Jonah Peretti I mean, I think that a lot of what we do does not read or appear as AI. I think people don't like certain things about AI. If you just ask them in general, what do you think of AI? There's a lot of pretty negative responses. I think people love playing WordChain, and I don't think they mind that AI coding tools were used in creating WordChain. And so I think the human agency and human intention is in the in the product, a person who designed the game was is obsessed with making a really fun game, the people playing it really like It. The creator of the game is in the comments talking to them. And they're like, when are you gonna make the next one? And I love this. And, and so I just think that saying, do you love AI? It's kind of like saying back in the day, do you love mobile? You know, or like, do you love mic? But a lot of people would say, yes. Nilay Patel You know, my career is based on, do you love mobile? Like the answer was millions of people being like, yeah, I want a whole website about that. Jonah Peretti Yeah. I mean, there were industry people who liked mobile. But I think most people were like, they liked the things they can do with it and they like the way they use it. Or, you know, it was, you know, it's like, do you like microcomputers? It's like, well, no, I like that I can do spreadsheets now and I don't have to have, you know, like, so I think if you think of AI as a computing platform, it's the question is what are the Apps built on top of that computing platform that Gen Z will like? And there already are many of them. And, you know, it's, you know, whether it's using them to do their homework or help them, you know, solve problems that they have with, you know, personal they have and talking to ChatGPT Or Claude or whatever. Like, I think there is a lot of love for AI and a lot of hate for AI. And I think we just live in a world where like everything is super polarized and everything is kind of love and hate. And so I think fighting for a new way of using AI and a new way of thinking of AI that puts people first, is creative, that it magnifies people's agency, that is not about feeding everyone The same slot, but is about like helping people have a more personalized and connected experience and connect with their friends. Like, I think that's going to be something that you will have to fight for because there'll be all kinds of dystopian uses of AI. But I think we can build something that is special, that makes people see the power of this new computing platform, but not through the platform itself, but through the applications And through the fun things that they can do with it and the fun ways they can connect with their friends. ([Time 1:05:47](https://share.snipd.com/snip/aa708ddb-3f38-43ad-a839-b2a7e9763429))