In recent months, publishers may have been noticing an influx of traffic from a not-usual suspect: SmartNews, the news app that was founded in Japan in 2012 and has been operating in the U.S. since 2014. The company is Parse.ly’s most reliably growing external traffic referrer this year, growing at an average of 8.8 percent per month across the 3,000 sites that Parse.ly tracks.
Publishers are happy about this traffic. But whether SmartNews, which now counts 15 million monthly active users globally, can be more than a traffic driver — whether it can actually convert readers into subscribers — is still a question.
Unlike semi-competitors like Flipboard, SmartNews doesn’t require or even allow users to login. Everybody who opens the app sees pretty much the same thing: Lists of news stories under the categories “Top,” “Entertain,” “Lifestyle,” “U.S.,” “Politics,” “Sports,” “Biz,” “Tech,” “Science,” “Buzz,” “Social” (where you can connect your Twitter account so SmartNews pulls in stories shared there), and “Discover,” where you can choose publishers to follow. Tap on an article and you’ll view it the way it looks on the publisher’s mobile site — ads and all — though you can also select a stripped-down, fast-loading “SmartView” version if your internet connection is bad. (That feature was originally built for Japanese users riding the Tokyo subway.)
“We are deliberately optimized for news discovery, as opposed to excessive personalization,” said Rich Jaroslovsky, SmartNews’ VP of content and chief journalist. (In digital journalism circles, he’s probably best known as the founder of the Online News Association back in 1999; ONA gives out a Rich Jaroslovsky Founder Award each year.) He oversees partnerships with about 370 English-language publishers, as well as the content operations at SmartNews’ U.S. and international English-language editions (the Japanese-language edition is managed separately). He and other SmartNews executives like to say that SmartNews “busts filter bubbles” — the core experience of the app is designed to present a diversity of content, including things you didn’t necessarily know you were interested in.
“We hope users return because they are discovering something new,” said SVP of product Jeannie Yang, “At the core is what we call personalized discovery, which is different from personalization or personalized recommendation. How do we understand what the user’s comfort zone is? How do we stretch them beyond that interest to surface hidden gems and also really break out of their filter bubble?”
The algorithm does turn up some suggestions based on what users have previously liked and clicked on. But the goal is largely to provide news quickly, not to turn up long think pieces you might have missed otherwise. That differentiates it from sites like Flipboard and Pocket.
“Flipboard’s origins were iPad-centric and to me, at least, Flipboard is still much more of a lean-back experience, something that you look at when you have time,” Jaroslovsky said when I asked him about the differences between the apps. “I think of SmartNews as much more of a lean-forward experience, something you look at when you don’t necessarily have a lot of time.”
The publishers I spoke with see SmartNews as a fairly simple set-it-and-forget-it traffic booster. Partnering mostly requires just giving SmartNews your RSS feed; publishers then have access to an analytics dashboard, and they can put their own advertising in SmartView and keep 100 percent of that revenue. Bobby Blanchard, the social media manager for The Texas Tribune, said that the publication’s traffic from SmartNews varies a lot, but in the rare instances when one of the site’s articles gets included in on SmartNews’s “Top” screen, there’s a big traffic spike; in one recent instance, it was “like a Drudge Report sort of spike — same class as that.”
MIT Technology Review signed on with SmartNews in February. “We’ve since seen 16.92 percent growth, month over month, coming from SmartNews,” said Rosemary Kelly, the publication’s head of audience development. SmartNews makes up 4.3 percent of the Review’s overall referral traffic.
“We have a very broad international audience at Tech Review, we want to continue to grow that, and so this seems like an easy thing to test,” Kelly said. “The hope is that SmartNews readers will subscribe to our newsletter” and that the most engaged of those will then become subscribers.
“These are great tests,” she added, “but in the end, if they don’t become subscribers or monetize in some capacity — whether it’s events or subscriptions — then it’s just traffic to the website.”
Vice, meanwhile, has seen success as SmartNews has opened up its pitching process. Publisher partners can suggest stories for promotion when, for instance, they have exclusives or stories that are doing particularly well.
Since SmartNews doesn’t require users to sign in, “our demographic information is not as rich as it might be otherwise,” Jaroslovsky said. Research — both internal and external — tends to suggest that the SmartNews U.S. audience is broad, between the ages of 30 and 54, and slightly more male than female. There are also interesting differences between the U.S. and Japanese audiences. “The push notifications and breaking news alerts we send out in the U.S. are big drivers of traffic to us and to our partner publishers,” Jaroslovsky said. “Japan is less of a breaking news culture, and we send out more news alerts in the U.S. than we do in Japan.” The U.S. and Japanese app interfaces are also different: Japanese people tend to swipe left to right, while Americans prefer to scroll.
In Japan, for the last couple of years, SmartNews has had programs that send not just traffic but revenue to its 2,000-plus Japanese publishing partners; over the next year or so, those will expand to the U.S. as well. “In the U.S., we have been more of a source of traffic than of revenue, but that will start to change,” Jaroslovsky said. “If the publisher ecosystem is healthy, then SmartNews is healthy. That’s going to be an important thrust going forward.”
This article was originally published by Nieman Lab. It was republished on IJNet with permission.
Main image CC-licensed by Unsplash via Saulo Mohana.