Why the owners of Condé Nast launched a tech incubator
Advance Publications is one of the world’s largest media companies. It owns Conde Nast, home to magazines like The New Yorker and Vogue, as well as dozens of newspapers across the U.S. It even has a majority stake in Reddit.
A few years ago, it launched Alpha Group, an incubator meant to build brand new tech products and grow them into fully functional companies. Rather than acting like a traditional media company, Alpha Group takes a much more expansive view as to what constitutes a 21-century media company, and its products span a wide range of functions, from a social polling app to a Facebook chat bot.
I sat down with David Cohn, a senior director at Alpha Group, to talk about how the company comes up with new product ideas and what it’s like to try to launch multiple tech companies from scratch.
To listen to the interview, subscribe to The Business of Content on your favorite podcast player, or you can play the YouTube video below. If you scroll down you’ll also find a transcript of the interview.
iTunes/ Stitcher/ Google Play/ Overcast/ Spotify
Simon Owens: Hey David, thanks for joining us.
David Cohn: Thank you for having me.
You work for Alpha Group. It’s a kind of incubator inside of a much larger company called Advance Publications. Lots of people might not know that name, but it owns lots of titles that people have heard of. Can you give us a sense of what falls under the company.
Advance is a big company. It owns a lot of news organizations, local news organizations like The New Jersey Star Register, The New Orleans Times-Picayune, The Cleveland Plain Dealer. It also owns the American City Business Journals. I believe there are 40 of them. So like The New York Business Journal, The San Francisco Business Journal. It’s the sister company of Conde Nast and the Conde Nast magazines.
And then there’s a couple other properties as well. At one point it owned Reddit outright, but it spun that out. It’s still a partial owner of Reddit. My team doesn’t work with any of those brands. We’re specifically an incubator team within the larger company.
What’s the history of Alpha Group? When did it launch? What was the thinking behind it?
It’s relatively new. Alpha Group is a little over three years old now. It really does come out of the idea that a team can be dedicated to looking into new products, new projects, and new audiences. The existing brands, they have a very specific audience, each of them, and they’re doing their daily grind. I don’t work with any of those existing brands because our job is to think about new projects and new products.
I think the history of it comes down to there were a couple things people were talking about wanting to try, and they wanted to just build a team that would have the space to go ahead and do that, and kind of do that on a regular basis. Our entire reason for being is to repeatedly build MVPs, launch them, test them. If we like what we see, then we build them out a little bit more and turn them into fully functional companies as my team steps back. My team continues to make new MVPs while those new properties grow and go through the drunken walk of the startup.
You’re not trying to launch traditional publications like the kind Advance traditionally owns. The New Yorkers or the Vogues or various newspapers. You guys are trying to launch the next Reddit or Facebook or something like that.
Yeah, that’s correct. I would say we’re always looking to leverage technology to differentiate what we do. And you’re right, I think if the organization wanted to launch another traditional publication or a publication that served a local audience in the way that their other publications serve local audiences, I think they have the talent and the knowledge on how to do that. We’re specifically trying to do things that are a little bit outside the purview of what they’re already doing very well.
Over the past few decades you’ve heard this criticism from certain corners: why didn’t newspapers invent Google or Facebook or even Craigslist? This idea that they’ve been too rigid in terms of what they consider their purview as media companies. It seems like Alpha Group falls into that kind of thinking, that a media company can’t get too rigid in what it considers a mass consumed editorial product, that it needs to think about platforms and building out innovative new products.
Definitely. I think the Alpha Group has a pretty broad conception of media. I wouldn’t say we only work in big J journalism.
What is a toy today but could be something very important and meaningful tomorrow? My team has been looking at Twitch. A lot of people would look at that and say, oh, it’s a bunch of people playing video games. That is true, but there’s something really interesting going on there as a platform. So we want to explore with those kind of spaces.
I was hoping to talk a little bit about your past work. I remember that you started a very early crowdfunding platform for journalism. This was before Kickstarter and Indiegogo became household names. Can you talk about that?
I started off as a technology reporter. And then I founded something called Spot Us, which was basically like a Kickstarter for journalists, but it was before Kickstarter. So I used to have to take 20 minutes to explain crowdfunding, which today sounds funny, but at the time it was a funny concept.
We would raise money for independent journalists who cover stories. We would get 50, 60 people to throw in $30 each, and that would be enough money to get a reporter to cover a story and publish it.
And the idea is that they would say ‘I want to write an article about this.’ And they would post to your platform, and then the idea was that if you wanted to see an article about subject X, you would pledge money to that, and then once they hit the threshold that would make this worth their while from a monetary standpoint, they would be able to investigate or do the research and create that piece of content.
That’s exactly right. The journalist would put up a pitch. The way I looked at it, because I was a freelancer for a while, is that if a freelancer pitches the editor, because the editor is the person who has the freelance budget, they can say yes or no because of money. So really, in this case, the freelancer was pitching the public. Here’s the story I want to do, here’s my background, all the kind of stuff you’d put in a pitch to an editor, but in this case it was to the public, and if enough members of the public wanted to chip in a little bit of money, then we would have enough. And then they would do the story, and at the end we would try to sell that story to publications. If we were able to sell the story to the publications, we would actually give the money back to the user in the form of credits. Not actual cash. In credits they could reinvest into another story.
And if we weren’t able to sell it we would just publish it on Spot Us.
Were there any successfully funded campaigns that stood out?
Probably the one we were most recognized for is we raised about $10,000 to send a woman on a boat to go to the middle of the Pacific Ocean, to go to the great Pacific garbage patch. She blogged the whole time. People who donated were getting regular updates from her as she was out there. And then it was on the front page of the science section in The New York Times. So that got us a lot of recognition and praise.
There were others of course. We had things published in The Oakland Tribune and different local magazines. We did mostly local stories. We actually paid to have an independent reporter at the Los Angeles trial for the BART police officer who shot Oscar Grant. The trial was in Los Angeles because they didn’t think he’d get a fair trial in Oakland. We actually were able to have a local reporter in Los Angeles report on it for a lot of local independent publications in Oakland who otherwise wouldn’t have had anyone on the ground there.
Do you think there were any flaws in your assumption about how journalism could be funded? I remember thinking at the time that this just seems like too much work on me, the user. I kind of just want to subscribe to a publication and have them come up with the ideas and bring the stories to me that I didn’t even know I wanted. Whereas here I would have to actively hunt out stories that I would want and then help fund them, and then maybe six months later I would have this thing delivered to me. Did you find any flaws in your assumptions for how the public would embrace such a tool?
Sure. I think the number one thing I would have changed, and we started trying to do some of this, is to fund a reporter as a beat reporter, and not specific stories.
So like a Patreon.
Exactly. Much more like a Patreon. Really you’re becoming a monthly member of a specific reporter who covers whatever the topic is. And you’re going to get, to your point — you’re not sure what you’re going to get from them week after week, but you do know that you enjoy them, you like them, and you want to get content from them about a general topic.
In fact that ties very much into one of the projects that we’re doing right now called Project Text, but we can talk about that later. But that would probably be the main thing that, when I look at Spot Us, that we did wrong. We were doing it very much like a Kickstarter model of a specific reporting project, an investigative piece, when really we should have done it as an ongoing beat reporter that you’re supporting.
I know you also worked for Circa, which was this really ambitious app that wanted to reorient how we consume news. It’s a little bit vague in my memory, but you guys were going to update stories as they were going and create threads — thinking about a different way an article can be constructed as news developed.
Yeah, I still think Circa was really ambitious, and I still get excited when I think about it. In fact I just hung out with some old Circa employees last night. Just sort of a reunion.
What Circa was — what we fundamentally challenged was the idea of the article. When you think about the article, it’s the base unit of information. If you’re a reporter, at the end of the day you’re supposed to produce an article. But what we would do is tell stories through atomic units of news. They were facts, quotes, statistics, events, or images. And we would thread those together over time to tell a story.
So you could follow a story kind of the way you follow a person on Twitter. An example — and I apologize that it’s a morbid one, but at Circa we did a lot of breaking news — would be there’s an explosion at the end of the Boston marathon, unknown how many are dead or injured. We’d send that push notification out, and people would decide whether to follow that story. And then a little bit later, 20 minutes or an hour, we send a push notification of a photo, and maybe 30 minutes after that we push a quote from the police chief.
If you remember, that was a weeklong manhunt that took place. Throughout it all all the other news organizations were writing article after article, much of which would repeat itself, which was frustrating for someone who already knew certain things and wanted to know what they missed. But at Circa we could keep track of what a person has or hasn’t read. So if you were on vacation or were brand new to the story, we’d start you at the beginning, but if you were following along the whole time, we’d skip all of the background because you already got it and relay just the latest information.
It was almost the opposite of 24 hours news. When there’s breaking news, CNN, because they feel like they need to fill so much air, they have to repeat themselves over and over again, and they do a good job of filling that air, but it just feels vapid and annoying. It’s the opposite of what a lot of people want, which is hey, when there’s something new, let me know. And because we would break out bits of information and make them discrete units themselves, we were able to do that at Circa.
[LIKE THIS ARTICLE SO FAR? THEN YOU’LL REALLY WANT TO SIGN UP FOR MY NEWSLETTER. IT’S DELIVERED ONCE A WEEK AND PACKED WITH MY TECH AND MEDIA ANALYSIS, STUFF YOU WON’T FIND ANYWHERE ELSE ON THE WEB. SUBSCRIBE OVER HERE]
If I remember correctly, it ran out of funding. And I wonder if it was too early for its time. It’s only within the last year that we’ve seen this backlash against social media, and publishers are reporting really successful traffic from news apps like Apple News, Flipboard, and Smart News. It seems like the adoption of news apps is really ticking up now.
Yeah, I also think our timing was poor in that this was just in the run up to the pivot to video. We used a very ‘just the facts ma’am’ approach. We tried to be objective and clear and neutral, and a lot of people said that just won’t work, you need more tone, more voice. This was still the era of Gawker. And of course then since 2016, a lot of people have really wanted some sort of objective voice or a sense of who’s a fair and honest player and how can we check facts. Which again, when you’re breaking things out in discrete units, it’s a lot easier to say ‘this is a fact’ than when you have an entire article. At that point the article gets attacked for any misused word.
I think Circa was a little early from a technical perspective, but it was also going against the grain and trends in a lot of respects.
How did you come to work for Alpha Group?
After Circa, I actually went to AJ+ and was an executive producer there for the first year of its launch. AJ+, to what I was just saying, was almost the opposite of Circa. It very much leaned into Facebook and leveraging the social network on Facebook to spread its content and its videos. When I left AJ+, I think it was the fifth or sixth largest news video producer on Facebook and the world.
Advance reached out to me and described the job, which when you describe it, it just sounds like a dream job. When I tried to explain it to my parents, it was like well, I’ll sort of be the mad scientist at this media organization. At first, I thought that it wouldn’t work out, because I’m in the Bay Area, they’re based in New York, and I really could not leave and move to New York. But they actually said I could stay in the Bay Area. For me, that was kind of a clincher, because again I really enjoy working on — I’m a big believer that all code is political. The platforms you build will end up impacting the editorial that is produced. And so I think at Alpha Group, one of the things we’re thinking about the question: what kind of platforms can we build that will produce content that we feel proud of or that we think is filling in holes that currently exist, both in terms of audience and markets?
That’s kind of how I ended up there. I think it originally was through someone at the New Jersey Star Ledger who connected me with the Alpha Group.
What are you doing on a day to day basis? What does your daily schedule look like?
It kind of depends. We have several different products. We have the Tylt. I’m not as much in the day to day in the Tylt, because they have a staff now and they’re kind of running it. We have Elsewhere, which allows you to make video memes. And then we have something called Pigeon that’s kind of in the Reddit space in that it involves sharing information, voting on the information in a way that allows the best content to float to the top. And something called Project Text.
I’m usually trying to work on the projects that are the newest, that have the least support, that we haven’t hired people for. And right now that’s mostly Pigeon and Project Text. Whereas for Elsewhere and the Tylt, we have brought on people who are taking more leadership roles on those projects, and that’s fantastic. At that point, I like to say that our role is similar to that of maybe the role of an early investor with the startup. We obviously want them to succeed, we have a stake in it, we have an expertise in what they’re doing. We did help build the MVP and we’ve obviously thought about the space. But we’re not there in the day to day. We’re not over people’s shoulders. They’re setting the quarterly strategies and reporting to us, but we’re really trying to let them get more and more ownership. Obviously if they came to us and they said hey, we’re going to take this thing into a casino, we would step in. But we bring people on who understand the vision and are ready to take the mantle and push that forward.
Are you like a roaming CEO who’s just there long enough to get the thing incubated so you can put someone else in charge?
Yeah. I don’t know if I would say roaming CEO, but certainly we are working on multiple projects that are so early that they’re not ready to talk about. I see us as mad scientists, trying to start something in a lab and seeing if we can get a spark of life. And if we are able to get a spark, it’s about nurturing a fire. If you’re nurturing a fire, there’s a certain point where it’s sustainable, and we need someone who can keep putting fuel on it, and tend to it and build it up.
But when it’s very early it may die out. And so we’re usually looking at different projects and saying, ok, can we actually make this spark into a real flame, or is it just going to die out? That happens as well. That’s totally part of the space that we play in. But ideally we find something that has all the ingredients to become a flame we can then pass on. It’s healthy enough for us to pass on to other people.
What’s the process for developing a new product?
It’s different every time, because every product is unique. One thing I would say is we try to think about what is the cheapest, easiest, fastest way to test early assumptions. We’re usually coming to these things with an assumption of: would this be interesting to an audience? Would people actually want to do this?
I’ll use Project Text as an example. Would a reporter want to send text messages to an audience? Would an audience want to receive text messages to their phone from a reporter? It’s actually pretty straightforward. Would this be obnoxious, or would it be something they get a delight out of? The quickest, fastest, easiest, cheapest way to build something to accomplish that goal and see what happens.
Every project has a different hypothesis or different thing we want to test, and so it’s a different process for each one. But we try and go after projects that we can see an early and easy way to test our assumptions.
My team talked a lot about Blockchain a year ago. But when we were thinking about it, it always came down to: is there a really easy and fast and cheap way to test this? And we struggled to come up with an easy one. I think there are a lot of experiments we can do there, but we’re a four person team, and to do that would really require all of our time for a long time, and we decided that that would not be in our best interest.
Is there a regular monthly pitch meeting where you’re pitching new ideas? Or is it more ad hoc where someone just starts talking about it, and it just kind of gains steam organically? How does it bubble up?
It’s actually a little bit of both. We do meet regularly. We chat about once a week, and about every other week, some of it is more like logistics about what’s going on with our existing projects, and then we’ll try to find some time to brainstorm. That’s not to say we’re brainstorming to decide on something that week. We’re not turning things around every two weeks. We play with, hey, what are we observing, what trends are there, does anyone have specific product ideas that are building around any trends? Those percolate over time through various conversations. And we start to kind of coalesce around some. Again, we are all kind of opinionated, we have different ideas sometimes, and that’s totally fine because in the end, what are the different assumptions that we’re working with when we’re talking about this idea, and can we test that?
Once we get to that kind of point, unless there’s enough interest and excitement for an idea over several conversations, if we feel like we have enough space based on whether or not we’ve handed off flames that we’re tending to, that’s when we’ll decide, ok, of the ideas we’ve been talking about, we’ve just sort of handed something off, we’ve brought something on that can continue the good work we’re doing with this project, what’s the next thing we’re going to pick up and push forward?
Would you say that Tylt, in terms of user adoption, is your most successful product so far?
I would hate to say successful. They’re all my children and I don’t want to pick a favorite.
Just from a user adoption perspective.
I would say from a reach perspective, yes. Their reach on social, I want to say, is 50 million. A reach of 50 million of Millennials and Gen Z people. It’s a social polling platform. People kind of come in and weigh in on topics on the site, but they can also chime in on social using hashtags, and that will register as a vote.
It is the biggest in terms of staff, the people we’ve hired. I don’t know the total staff number now. I want to say it’s between 20 and 30. I would say that as far as a company that has grown and has its own leadership and has a really strong reach, the Tylt is probably the biggest.
So people download an app, and they want to launch a poll for fun — it’s not like a scientific poll, right?
It’s not a scientific poll. Users don’t create the polls actually. We have a team that creates the polls. We have our own content and editorial teams. They’re creating the polls based off what’s happening in the world. Those are voted on on site, they can be embedded on other sites, and then we’ll create social content that people will engage with, and as they engage with it, those engagements can be registered as votes.
You said a lot of your work is on Pigeon right now, which is a chatbot service. I feel like there was a lot of hype around chatbots a few years ago saying they were the future of news, and then they really fizzled out really quickly. I remember trying a bunch and it was really just a way for people to use my Facebook Messenger to send me stories. They weren’t doing anything innovative. But you guys are doing something somewhat interesting in how you recommend links through the chat box.
Yeah, you’re right that there was a pop of chatbots. I personally was kind of dissatisfied with most of them. It felt like a simple RSS feed and it wasn’t solving a problem. Thanks New York Times for the chatbot that’s sending me your articles, but I know how to find your articles. That’s not a problem for me. I just go to your website. The bot doesn’t make that any better.
What we built with Pigeon is around news and information, but it’s not the content of any single organization. It’s actually similar to Reddit or the original Digg.
I actually compare it to StumbleUpon.
Yes. I was an early user of Digg and StumbleUpon. I was on Netscape Navigator for a little bit, which was another competitor in that space. There was this exciting time between 2004 and 2006 where there we these social news sites. There were a lot of them. It was just the idea that the wisdom of the crowd would be able to select and determine what is the most important piece of information.
On Pigeon, if you submit something, the bot says ‘thanks for the submission, and if you’re a new user, we guarantee this will go to five other people.’ Whereas if you submit something on Reddit you’re not sure what will happen to it. But on Pigeon, the bot says ‘thank you very much, I’m going to show this to at least five people.’ If all five people vote it up, it’ll go to 20 people. If all five vote it down, it stops spreading. And, if you are repeatedly voted down, the next time you submit, your reach diminishes. So now you’re going to go to four people, or three people. Or, if you’re continually voted up, your reach increases. Now you’re going to go to six people, seven people, eight people.
And another element is that when people vote on the content, they don’t know who submitted it. It’s anonymous. It’s not without consequences. There are consequences to whether people like or dislike the content you submitted. But they don’t know it was you who submitted it. I could submit something in Pigeon and Ashton Kutcher could submit something, and people wouldn’t be voting on celebrity status, which on Twitter would be the opposite.
And that’s why I compare it to StumbleUpon. Because it can’t be gamed, you can’t post something and then get your million Twitter followers to go upvote it.
Exactly. You cannot create a mob of people to come and find your content and upvote it. Again, to that point that I said earlier about all code being political, the decisions you make about a platform impact the content. We really wanted to create a platform that put content front and center and took personality and ego and tried to put that on the backburner.
Maybe ego less because there is some pride in knowing your piece of content was accepted and spread, and there is a leaderboard that you can go up and down. In order to do well, you have to have good content. The content is front and center, versus I will compare it to Facebook — when you see a piece of content you’re seeing the content but you’re also evaluating it and judging it based on who showed it to you. Is this my crazy uncle? Is this my colleague? And that will change how you react with it as well.
Where does this all live?
Right now just Facebook Messenger. Again, while we were building this, we were like, where’s an easy place to quickly build a chatbot, and Facebook Messenger really lends itself to building a chatbot easily and getting it in front of an audience.
In theory it could live anywhere a bot can live. It could exist on Slack or WhatsApp. And again it’s a chatbot for news and information, but really it’s not about — the bot is kind of mediating connections between people. It’s really a social network in a way. That’s what I think was missing from some of the earlier news bots that you mentioned earlier. They were really just about giving you the content, it had nothing to do with connecting you to other people to that content.
Or allowing you to have a hand in what content was chosen.
Exactly. I could imagine a version of Pigeon on smart speakers. And at that point it might be audio submissions and you could literally imagine a Pigeon community for Berkeley. And I’d say hey everybody in Berkeley, did you notice that there’s graffiti on the school? And that would go to five random people, but it would be people in the Berkeley Pigeon community. And potentially spread or not spread based on the relevance to the wider Berkeley community. That would be a way where any of these sub communities can have a conversation with themselves where content is front and center, and they’re allowed to focus on the value of the content that’s being submitted.
How does one initially engage with the app?
It’s actually very easy. You can just go to Facebook.com/pigeoncorps and send a message. When you go to that Facebook page, in the right hand side there’s a button that says ‘send message.’ You just hit that button and say ‘get started,’ and you’re engaging. At that point you can select ‘read’ and see what other people have submitted. Or you can post and submit things yourself.
Right now the community focus is on internet culture and memes and humor and things like that. But again in the future we can imagine lots of different topics.
How does it spread to new users? Is it 100 percent word of mouth? Or is there a viral mechanism within it that allows it to spread?
A little bit of both. Right now there are ways where users can share the Pigeon bot with others, and that will help with distribution initially. Some people share the bot with their friends. And then we also do some Facebook advertising and marketing to get new users. And we’re able to see the stickiness we have.
What kind of user adoption have you seen so far?
Pretty strong, actually. I want to say we’re at 70,000 users, and app usage is pretty good right now. Our dailies are pretty strong. We recently did a lot of new features and tweaks, and we’re seeing the increasing rate of people getting content and wanting to read more content, and also posting.
One of the things we were able to start doing is seeing segmenting styles of curators. So some people, for example, really enjoy curating a lot and want to be exposed to fresh content. Like something that has just been submitted by a new users. And others have a lower threshold and really want content that has already been vetted. And we’re able to start to distinguish between those and give people a little bit more of what they want in terms of the content. Adoption has been pretty good. We’re pretty happy with it. We’re trying to bring someone on to take a little bit of the helm and expand into new areas. We’re not totally sure what those are going to be, but just like there is unknown number of subreddits, you can imagine a countless numbers of versions of Pigeon that focus on specific topics.
Did you like this article? Do you want me to create awesome content like this for you? Go here to learn how you can hire me.
Simon Owens is a tech and media journalist living in Washington, DC. Follow him on Twitter, Facebook, or LinkedIn. Email him at simonowens@gmail.com. For a full bio, go here.
Related articles:
