We recently hosted a panel on futurism and tech at our WhipSmart 2.0 event, with panelists from BuzzFeed, Axios, the New York Times, Group Nine, and Quartz. Here’s a roundup of the best bits.
Our WhipSmart panel on futurism and tech was moderated by Aedhmar Hynes of Text100. The goal of the panel was to talk about the challenges of reporting in the digital age, and what the future might look like in terms of technology to aid that. Speakers included:
- Ashish Patel – Chief Insights Officer at Group Nine Media
- John Keefe – Bot Developer and Product Manager at Quartz
- Sara Fisher – Media Reporter at Axios
- Gilad Lotan – Head of Data Science at BuzzFeed
- Renan Borell – Senior Editor of Digital Storytelling at the New York Times
Once the introductions were done, the conversation began:
The role of platforms
Aedhmar Hynes: What I wanted to do in opening up this discussion is just to reflect a little bit on the last year. I would love to get your perspective on how you’re dealing with the changes and in what direction are you headed, in terms of social platforms and the role that they are playing in getting content out to your audiences?
AP: So Group Nine and our four brands are a little bit different than a traditional web media company. We’re primarily focused on building video and distributing that natively through social. Along with BuzzFeed, we’re some of the biggest players in terms of purely socially distributed video so it’s been a challenge for us, I would say.
We’re aware that it’s very risky to a certain extent to have as much consumption as we do on social but that’s where we see our audiences are. So we’re going to continue to figure out what we can while the platforms are constantly in shift.
GL: At BuzzFeed, we’re acutely aware of what we get from every platform, and I think that as a publisher especially as we continue into this uncertain future, you really have to understand what you’re getting and whether your investment is worth it, and a lot of that can be fueled by data. So the better data you have, the better understanding you have of your audiences, and the better you can quantify what you’re getting.
For BuzzFeed, there’s a lot of focus on experimentation and learning. So what can we learn about audiences about formats, about content, what’s a frame that’s successful? And we run a lot of content experiments to understand what are the things that are working with certain audiences and we really dive in and invest in the ones that are working.
RB: I think one of the big takeaways from everything that’s happened is how important it is to have diversified places where you’re putting your content out.
We were going to do a project this year where we were going to sit with all the different desks and talk about all the different levers they could pull to promote their stories and get their stuff seen, and we realized that there are a couple platforms that drive the majority of traffic and readers, and everywhere else your mileage may vary.
We realized that we would have much more luck just writing stories that would be better for digital, if we could shift people’s habits to write, for example, headlines that look like they were meant for digital instead of meant for a newspaper and just put on the internet. That would have a much much more beneficial result than figuring out [something like] we’re going to do Instagram stories, can we turn that into a traffic driver?
We just realized that there’s so many different platforms or so many possible things you can do and I think the most important thing for us is just to write stories better.
Using data to drive content
AH: And just building on that whole notion of the data, everybody’s talking about audiences and the ability to understand through the data exactly who we’re communicating with.
How granular should we go in data and what concerns would you have, or what are the big opportunities looking at that data and using it to drive how you publish content?
SF: I’d say that when it comes to e-commerce you want to go really direct, but when it comes to things like news it’s a little bit tough because there’s a wide discrepancy right now as to how far we should be assignment editing for the American people.
Should we be giving them something that they might not necessarily want to consume but we think is good for them to consume, should we be giving them something that we know that they want to consume but might not be the best thing for them to be an informed citizen?
No one really knows, and I think the key or the trick is to figure out a way to just do both. I think your really sound editorial newsroom will recognize that it’s their duty to have that balance and to reach the consumer on both fronts, and any newsroom that’s just trying to reach the consumer with the news that they want to be getting is kind of leaning into clickbait a little bit and that’s a dangerous place to be.
JK: That’s another role for the data we’re exploring which is, if you’re providing information in not just a personalized environment but a personal platform, like SMS text or Facebook Messenger or a smart speaker, then you’re getting to be into somebody’s world, into the world where they communicate with friends and family. That’s when you can also be smart about not just what you’re providing but when you’re providing it.
So we think really carefully about when we send a push notification out over Facebook Messenger. We don’t want to do it when you’re sleeping, or maybe not when you’re at work, we want to do it when you are more likely to be interested in talking to us. We don’t want to annoy you, because the moment we start to annoy you, you’re going to turn us off, so we’re looking at different ways to use the data not so much in the content tailoring but the tailoring of the delivery.
AP: Data is a very broad term, so it really depends on the context and the goal. So for e-commerce, getting down to that single individual person makes a lot of sense, but for media, you have to create quite a bit of content or personalization ultimately just becomes a targeting challenge at that point.
RB: I sometimes call it the holy grail — the ability that maybe one day I’ll be able to sign into the New York Times and it’ll know exactly what I want.
I think the data is there. I feel like the problem that I run into more — this is another cliche that we use a lot — but it’s not about raising the ceiling about what we can do with the data, it’s about raising the floor and figuring out how we can use some of this data and turn it into actionable insights.
Where I’ve found myself mostly is just trying to figure out how to distill the really smart things you can do with data, and being able to turn that into a comprehensible message that I can take to anybody.
How AI and bots will shape the future
AH: So, building on the data piece and just thinking about the technology and the application of the technology, I’m always fascinated with this notion of “Man Plus Machine” and I’d love to get your perspective on what can we expect in the future in terms of how either AI or BOTS are changing the landscape in terms of publishing content and where is it all headed?
Do we have direct relationships with bots in the future and to what extent can that be predictive of use that’s around the corner that we may not see?
JK: With the caveat that I’ve never been able to predict the future of news I will say that I think that these personal platforms like Alexa, or Facebook Messenger, are just things that are going to be in our life in some form or another.
I also like to say that despite the fact that we do look into natural language processing and the ability for some sort of machine learning to help with that delivery, the real secret to Quartz’s bots are amazing humans. We have teams of people who are writing for our chat platforms and for our bot platforms and it is that writing that makes it stellar and enjoyable and what makes you want to return, so I don’t think the human element is going away anytime soon.
AH: There’s this notion that if you can start to see machines can help us see patterns then actually it may tell humans something that we wouldn’t naturally assume. Gilad, any thoughts on this?
GL: I come from tech, and BuzzFeed is my first job in a proper publisher media company, and to me, it feels like a tech company inside. There’s very much an emphasis on building our own technology. We like to build stuff and I think I think with regards to artificial intelligence the story is around better forecasting optimization ranking.
So we’ve used neural networks in many parts of our publishing layer and are increasingly finding advantages there. Let’s say someone writes a post. The post is written in our CMS, but then where does it go? We have hundreds of Facebook pages, all these Instagram accounts, all these YouTube channels, we have just all these places where the content could go and historically that was a human decision.
At this scale, it’s impossible given the amount of content that’s created and all these channels to make the right matches, and so we have systems internally that take the piece of content that learn from historical sets of data and then performance data from audiences that we’ve captured over time.
Which content will break through the noise
AH: So when you think about breakthrough content or breaking through all the noise, are you seeing trends right now in terms of this is clearly going to break through the news cycle in some ways over others at Axios?
SF: I think we found a very high value in actually breaking news. There’s been such a barrage of news outlets that have popped you can get your news from 40 different websites if you wanted. Good reporting we’re finding breaks through really well.
I think the other thing that breaks really well is having a distinct voice for each platform, understanding which nuances work well. But by far, by far, there’s nothing you could do to manipulate a story or to tailor a story that makes it more effective than just breaking the news yourself.
RB: I 100 percent agree with that. What do I have to tell Maggie Haberman? Nothing! She’s going to break the news, it’s going to take care of itself, there’s nothing I can help with there. There’s nothing better than actually going out there and doing the work and getting the scoop. All the promotion in the world isn’t going to change the fact that if you get a good scoop people are going to share it on Facebook and Twitter.
AH: So for those that are not breaking news necessarily, is there a format emerging as the thing that we know is going to break through well?
AP: I think breaking news at one level is just original content, right? It’s something that you own, it’s something that you created, so we really believe that cutting through the noise is around original stories and telling really good original stories.
GL: Owning certain beats I’d say is something else. So BuzzFeed has been investing in breaking news as well and we’re seeing sort of the benefits of that. But also, for example, the misinformation beat is something we’re known for.
RB: One of the things I’m trying to get people to think about more is the story after the story. So if we get a big piece of news and it is going to change the conversation, we have that first story where we broke the news, but what we should also be thinking about is what are the five or six pieces that are going to come after that and what form should they live in.
That concludes our highlights reel of the future of digital tech, thanks again to all our panelists. If you’d like to see how data can help your content strategy, take a tour of NewsWhip Spike.