Why an online polling platform hired a seasoned journalist to run it

Simon Owens
Oct 31 · 20 min read

Advance Publications is one of the largest media companies in the world. It owns dozens of newspapers, the Conde Nast magazine empire, and even Reddit.

A few years ago, it created the Alpha Group, a tech incubator that would launch small startups and try to grow them into thriving, standalone businesses. One of those startups was called The Tylt, and it’s a platform that allows its users to participate in online opinion polls on a wide range of issues. The Tylt was successful enough that Alpha Group spun it off as its own company.

Recently, The Tylt hired Selena Roberts, a seasoned journalist who’s written for The New York Times and Sports Illustrated, to serve as its executive editor. I recently sat down with Roberts to learn about the site’s editorial ambitions and whether online, unscientific polls have any journalistic value.

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/ Google

This transcript has been edited for clarity.

Simon Owens: Hey Selena, thanks for joining us

Selena Roberts: Glad to be here, Simon.

So you’ve been hired as executive editor for a platform called The Tylt. It’s owned by Advance Publications. Can you start by explaining what Advance Publications is?

Advance Publications is a huge organization that started in Staten Island with The Advance and grew over the decades to become one of the biggest forces in media. If you look under the Advance Publications umbrella, you’ll find legacy media companies like Conde Nast, you’ll even find an ownership of Reddit. It’s vast. It also oversees Advance Local, which is where we’re located. When you look at the digital brands, all the newspapers that fall under that brand. You can see where a small newspaper in Staten Island really grew to be a media empire.

Yes, so it owns a bunch of newspapers. It also owns Conde Nast, so The New Yorker, Vogue, Vanity Fair. So this is a very large media conglomerate.

Absolutely. It’s got the premiere titles. It’s done a great job of really pivoting with the times and going into digital, and all the innovation in the areas that they really have made inroads in. We’re part of that. We’re in the incubator side, the area where they’re trying to press new buttons and work with new technology, and that’s where we fall. We fall into that area into new technology, and getting into the fascinating area of data journalism. A lot of people are talking about data these days and what you do with it and how you use it in journalism, and I think The Tylt has found some interesting ways to measure sentiment and find where the passion is around subjects.

You mentioned this tech incubator. I actually had on David Cohn a few months ago and he runs this incubator. Can you talk about what this incubator is and how it works within the company?

We were originally developed by the Alpha Group, which serves as an in-house tech area, this media incubator. It’s all under Advance Local. That is part of where that innovation comes from. Testing new technologies, developing new areas of expertise. What Advance Publications, the mothership, has done is really create these areas where people can put their tech hats on and put forward the ideas about how to do something new with journalism.

Basically Advance Publications was like ‘we’re a big airliner, we can’t pivot that fast. We don’t know what the future of media is going to look like. So let’s start this incubator that can launch all these tiny startups that can grow into the next Reddit.’

I think that’s a good summation. The interesting thing is being part of the Alpha Group and having that startup vibe. We’re not in One World Trade where a lot of the prestige publications reside. We’re in our own little area in Chelsea, and it really does feel like a startup, because it’s new technology applied to journalism. You get that energy, the energy that people are on the cusp of something new. I think creatively, journalistically, it’s a pretty special place to be.

So The Tylt was originally part of the Alpha Group as one of these startups, but then once a startup is successful enough, they spin it out as its own company. That’s what they did with The Tylt, right?

Yeah, we are operating autonomously. The idea is to spin the startup out and see if it can make it on its own. We’re standing on our own as our own company, with our own identity, really looking to form our own community within journalism and within media, but also with brands as well. I think it’s such a special place to be, because three years into this it’s really defining itself in the marketplace. It’s also defining itself with our users and with the people who come to us every day to get on our social polling platform and tell us what they think.

What’s so interesting about that is we capture so much sentiment with the debates we put up, it really consists of voices telling us a story. Every vote we get is a voice that tells us a story. I think it’s a really cool place where we get to reflect that back. We get to tell people, in return for giving us your vote and giving us your sentiment around a topic, we’re going to create stories that speak to that. I think it really empowers users who come to us on a daily basis.

Let’s zoom in on that. What does The Tylt actually do? You mentioned it’s a social polling platform. What does that mean?

When you come to one of our debate questions — and we start everything by asking a question — it can be on culture, it can be on entertainment, on politics. Very often it’s reflecting what the energy of the day is on the internet. Not all the time, but very often it’s about what’s happening today. Are we talking about a new record that Cardi B dropped? Or are we talking about immigration? It can be any of those things. And then we ask a question about it. We want to know what our audience thinks.

Just recently we had an interesting poll question on politics and electability. We ask our audience, does electability matter to you? What we found was that 57% of the people said no it doesn’t. So you can imagine how that kind of platform, when you start every day with a question, and you’re receiving the input, and what we try to do to the input, and what we try to do with every article we post, is we start it with a question. We have the polling apparatus there where you can touch and tell us how you feel, but you can also read contextually both sides of the story. We give a little bit on this side, a little bit on that side. So if you’re looking for more information on a topic, you can read, you can go through the multimedia that we put into the article. And then at the end of the day you can find out how people are voting by touching. We give you automatic real-time polling telling you where you’re sitting on this position.

I think it’s a pretty interesting way to think of users as people who are opting in to tell you what they think. We send these debates out in a variety of ways. You can vote on our site. We can send it out via our beacon widget across the web. You can also find these debates on social media. There are so many ways to vote, and so many ways we can capture that vote. With all that capture we get sentiment, and how people feel about these topics. Certainly our core groups are Gen Z and Millennials. So you can imagine how many great stories we can tell from that data, but also how a brand can get value from that data as well.

Let’s give a concrete example. At the top of the homepage right now you have the question ‘should everyone go to therapy?’ You open up that link, and there’s a poll at the top, and you can either say yes, everyone should go to therapy, or no, it’s not something that’s for everyone. And each answer has its own hashtag. And then you have an article that gives people additional information so it can help them have a more informed decision. There’s an editorial layer to it. So there’s the case rounding up information about why therapy for everyone would be a good thing. But there’s another part showing sources that say that therapy might not be for everyone.

It’s fascinating, right? You’re basically being given as a user the ying and yang of every argument. These are arguments that you can give context to both sides. We don’t ask questions where you can’t present two sides. We don’t get into the weeds of conspiracies. We get into the culture and politics, and you go in and you’re able to read contextually around both sides of the argument. You’re able to use that to inform the vote you’re going to give, or sometimes people already know how they’re going to vote before they read. In realtime you see how you stand against other people.

But over time we leave these polls up and you’ll see that opinion does evolve. More people come in, we get more data points from people that come in. You see over time from our graphics that the vote will change and people come in and add one side or the other. Over time you can see how these votes evolve and where they come from around the country, and where people are voting. I think that in so many ways it gives you an idea of what people believe and think and the sentiment. You can look at the 18 to 34 age group and where they are.

I think it’s a valuable tool journalistically because we’re trying to find out what’s the next story we want to pursue or the next video we want to do for one of our channels. We have that sentiment coming in and it can guide us. You can imagine how valuable that’d be for brands. But when we look at what’s coming in and how we want to reflect that in a story, an example of that could be in politics where we had questions about Gen Z and Millennials who desperately want to vote, but they don’t know who they want to vote for because it seems like there are 500 Democrats running. So we had a program designed to really get into that question, and we did it in a very fun way through a fun show called Election Connection. We did a dating game around all the Democratic candidates, and we put in all the facts, and what each candidate stood for and their platform. We tried to have some levity with it. But we also tried to tell a good story with it with the information we’re receiving from users.

As a journalist for so many years, the special part of that is you get in the room with journalists at maybe a Page 1 meeting, and people kind of come up with story ideas. Sometimes they’re really good story ideas, and sometimes they’re just based on a hunch. The good thing about what we do is we don’t have to act on hunches. We act on information. And we do that editorially. We do that for brands. We can tell stories that are important and relevant because we see what people are telling us.


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:

Ok, back to our scheduled programming…


Is there an app? Or does it mainly operate on the open web?

On the open web. Certainly the voting modules operate within social media too. They will take this right to the voting module if you found this on Facebook or Twitter.

You are coming in with this title of executive editor. You have this background in journalism. It seems like they’re trying to take this tech product and add a journalistic flair. You were a writer at both The New York Times and Sports Illustrated, right?

That’s correct. I was there right before the digital world descended on print and in that transition time when a lot of good things happened and a lot of mistakes that were happening. I think I was there when they were trying to figure out where they were with digital. And I was there during the days when the first reaction was to give away stuff on the internet for free. Obviously, that was a mistake, because once you gave it away for free, it was very hard to pull that back in.

I was there for the pre-digital age, the digital age, and the transition into where we are now. And I think journalism has changed a lot. It’s much more social now. The folks we hire here at The Tylt are very skilled at digital media. Many of them are digital natives. Many of them came upon journalism on the web and don’t know it a different way. It’s interesting to see how they use social media to reach out and engage with audiences, and then bring all that information back and inform what debates they’re going to have that day.

It’s been a great transition. I’ve enjoyed riding the best of print times and certainly gone through the worst of it in print. I think The Tylt, along with some of the legacy titles too, are finding new ways in the digital space to create some great journalism.

How did you end up working with The Tylt? Did they approach you?

I left Sports Illustrated in 2012 to start a streaming service for sports. It probably was a little early. It was back when most folks wanted everything to be ad based, and we were subscription based. So we had an amazing two or three years, and I think we decided that we couldn’t find the right money to go with our idea and that it probably wasn’t place for us to be, so we pivoted into a production company. I produced for five years several documentaries, feature films, and also have written for television pilots, etc… I was pretty happy doing that for the past few years.

But as a journalist, you’re used to that daily feel of what’s going on in the world. In filmmaking, you don’t get that anymore because projects are longterm. You can work on a project for two years. So I did miss that. And a friend of mine was listening to me miss it and said, you know, there’s this interesting thing going on in data journalism, and you should really check it out. I did, and we had some great conversations here at The Tylt, and it seemed like a really good fit. One of the things we discussed was really taking all that data we get and the sentiment we get and all the emotion that we get from the topics we put out, and really turn it back into storytelling.

We’re working on some bigger projects, longer term projects to really represent that. Whether it’s a longer story that’s more magazine like, that really would take everything that we get from data but put it into a multi-media type story, to really give back to those use our product. What did it all mean to them, and contextualize it as well.

All the information we receive and the sentiment we receive really is informative on an R&D level for brands, but also to filmmakers. That’s my background, so we’re talking to production houses and filmmakers about using what we know about Gen Z and Millennials and turning it into storytelling form on a bigger level, whether it’s a documentary or a feature film as well.

Those are the conversations I’ve been having in the three short months that I’ve been here, along with the daily thinking about how we want to even just do an Instagram story. We had one of our reporters cover the climate march, and we wanted to make sure that we were reflecting the information that our audience gives us through debates, but also to be there live and cover something live and to see how that sentiment really does play itself out in a big event like that. We’re doing a lot of different things. We’re doing some stuff I think will be more toward the storytelling format, whether it’s a longform story or something we do with a production house down the road.

You run a story. You have a poll in it. How is that poll being used to enhance your coverage?

Think about climate change and all the different questions that go with climate change. We can ask our audience of Gen Z Millennials about plastic straws. Do you want to ban plastic straws? And 74% of our audience says it wants to ban plastic straws. So as we think of a bigger picture story about climate change, when we think about a documentary about climate change, that’s one piece of information we have with us.

Another one around climate change maybe is it ethical to have children knowing what you know about climate change. What was interesting to me was that 41% of those who responded to that question think it is unethical. That’s a big number when you think historically about people wanting to have children. So that’s another interesting piece of information we can put into a story about sentiment from Gen Z Millennials around climate change. And maybe it’s a longform piece, a piece of multimedia, or maybe it is something that informs a documentary. Or maybe it’s a piece of information that a documentarian would like to have for their documentary.

I think there are multiple creative ways that we can reflect what people are telling us back to them by just asking questions and maybe asking several questions over time. That question about children over time has changed over two or three years. You’ve seen that number rise to 41%. You can see how things change over time. You can see what kind of audience is saying to you. That’s a better way of going after a story than just a hunch or looking at some surveys or not really knowing what it all means.

Just as the Wall Street Journal will hire a polling firm to poll about a candidate or issue, and then write a whole article about that poll, you guys are creating a more scalable version of that that’s pulling in results more quickly that you can then cite in stories, so that you can write articles that are informed by those polls.

You’re absolutely right. It’s having your own in-house, all the time polling center. It doesn’t matter if it’s about climate change or k-pop. It’s sentiment. It’s heat around a subject. And I think that’s super valuable, not for us just as journalists, but brands too.

Imagine you’re a dating app and you’re looking for conversation starters for your first date. We do that all the time. We ask conversation starters all the time, so you can imagine working with a dating app, and we’re delivering your six best conversation starters as voted on by the people who vote with us. There’s something fun we can do with it. There’s something interesting and also informative, and you can imagine if we’re asking about first dates, the kind of information you get from that. We were having fun with one debate we asked about whether you’d go to Disney World on your honeymoon. Those are interesting, fun questions that also give you a lot of insight as to what Gen Z and millennials are thinking about marriage. At the same time that Gen Z and Millennials tell you they don’t want to get on Facebook for love, they’ll tell you I’d rather meet someone in person. But they’re also old fashioned like that. You can get myriad responses that add up to great storytelling for brands and great storytelling for users as well.

News sites have been embedding polls on their articles going back 15 years. How does this differ? Is it because the whole site is built around the polling, so you’re creating this symbiotic relationship where it’s not just the poll that’s thrown in as an afterthought? How does this differ from a lot of news sites have been doing for a long time.

We contextualize our polls. It’s not just a poll question, we give context around it. We also put multimedia in. If we’re talking about Taylor Swift, we put in a Taylor Swift playlist. You make your decision based on all this information we provide along with a poll so it’s not just a question.

We also have a way to deliver it. We use programmatic ad units. We utilize all our vast resources at Advance Local. We have ways to push that out to so many people and we have such a big, scalable audience, that we receive all this opinion in this easy, organic way, and it’s a better model. When you come to our platform, you may vote on one thing, but maybe you see another debate you want to vote on. You have this hub where you can gain agency. That’s what voting is. It’s about having agency.

A lot of these online polls have been unreliable because advocates can flood the zone. One of the benefits of a Wall Street Journal poll is they’re hiring a professional pollster who’d doing a scientific poll. Are these polls scientific at all? If not, how do they add journalistic value.

We do use a signal algorithm. It’s an effective way to measure engagement and exposure, and also provides amplification and gathers more and more data. We have ways that we can filter out bots in our data. It’s a good way to measure sentiment, a heat around a subject, It’s passion around a subject. It’s how do people feel over time on a subject. It’s more of a measure of hearts and minds than specifically a Gallup poll.

So it’s not really scientific, but you’re trying to measure passion rather than saying this is something that could be cited as an official poll.

This is not like Pew Research. It’s more about ways to gather that heat and sentiment, gather that passion. Measure the feelings around the things we talk about on the internet every day. The cultural issues we face every day. And what we try to do over time is measure that sentiment, and reflect it back, and figure out what’s the story people are trying to tell us, from the input they’re giving us on a daily basis. That is incredibly valuable. You’ve been in those newsrooms where you try to figure out what are people saying. You look at a poll, but you know that many of these polls are conducted with landline polls. This is a different way of gathering information. The way we see it is this information is so valuable to us, it’s telling a meaningful story.

Is part of it less about the data output but more about making it more fun to read the content. You’re gamifying it and adding a level of engagement, making it more engaging than the average article on the New York Times?

Yeah, absolutely. It’s a good point. It’s something about finding a way for people to intake information, gamify that. Remember, as we’re contextualizing these debates, we’re drilling down into both sides. In a sense, it’s a more interesting way to read an article. It’s a gamification of that article. And then there’s the agency of voting on something, where someone somewhere is finding your vote interesting to them. And I think it gives the audience more credit for how they feel. I know working for years in journalism, I don’t know if I ever felt a part of an article. I think in many ways, what we do is capture that sentiment and passion around an article, and then maybe they ask their friends about that topic. It’s something denser than an article.

What kind of team do you have on the editorial side? What kind of talent do you have at your disposal?

When I was first interviewing to work here, I met some team members, and I was amazed at their talent. What I mean by that is they’re able to do so many things multimedia, so many things social media, so many things with photoshop, they’re so multi-dimensional as journalists, that’s what blows me away.

They’re cutting their own videos. They’re creating memes. They’re using all that technology. To me, that’s an incredible new talent that’s emerging in this business. The journalists who are creators as well. I have a staff of about 16 to 17 folks on the editorial side. Social media managers. Social media producers. Along with journalists, editors in our categories of politics, sports, culture. It’s not the biggest team in media, but it’s a talented team that can do a lot of things with one story.

It’s mainly focused right now on curating outside content. Are there plans to do more original reporting?

We want to take some of this opinion data and sentiment and turn it into longform. The writers here are incredibly talented and will do terrific jobs with that. If we want to do something with climate change, we’re reporting it out. We’re building relationships in that world by attending the protests, by talking to the leadership of those groups. We do that kind of reporting, and we’re going to do more of that. That’s something you’ll see going into 2020, that we’ll take ownership of what we find out, use that by going out and talking with people. Some of our brand new launches on YouTube are reflecting that now.

We have one new show on YouTube, and it’s our editor at large going to an independent book store in Athens, Georgia and talking to a bookstore owner about what it’s like to sell books in this digital age, and what that community is like. The reason we did that story was that we were seeing a sea change in what Gen Z and Millennials were saying about books. We’d put debates up about books. And so we started to think about what we could do with some sort of storytelling device to reflect that.

Can you talk a little about the business model?

I think what you’re seeing is we’re a great place for branded content. If you think about what we do and the debate structure and how it all starts with a question, you can see where we’d be this great place for branded content. You can look on our site and see those branded content plays. We have driven engagement and elevated fan engagement. We can drive response. We can work with brands to measure customer sentiment. All those things we do on a daily basis.

So there’s a market research angle where a brand might want to test an idea?

If you’re looking for customer sentiment. We had an opportunity where we worked with a natural gas company that was looking for solutions over time and wanted to see what customer preferences are. We posted questions about natural gas infrastructure. That kind of company that’s looking for what people feel about a topic and work with us to track that sentiment over time and watch how people share it across the internet and look at, in this geotargeted area, how people felt about it, and what people wanted out of that question. It’s a valuable took for companies that want to measure sentiment, but it’s also valuable for direct response. We’ve worked with streaming services that were looking for what people think about the Super Bowl. In a sense, it’s about being able to create these custom campaigns that revolve around a question. It’s entertaining. We’re gamifying it. But we’re also giving the brands information about the questions that are being asked and the sentiment around those questions.

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:

The Business of Content

The podcast about how publishers create, distribute, and monetize digital content.

Simon Owens

Written by

Tech and media journalist. Email me: simonowens@gmail.com

The Business of Content

The podcast about how publishers create, distribute, and monetize digital content.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade