Occasionally I take a quick look at where the traffic to my articles comes from. (“Quick” since content ultimately get broad play because people like and share it; if you’re in the content business, any time you spend on doing deep analytics dives that could be better spent creating great content.)
My usual traffic sources are LinkedIn, Google, and to a lesser extent, Flipboard. But then I saw a decent amount of traffic coming from SmartNews.
If you aren’t familiar, SmartNews is a free news app that delivers stories from hundreds of publishers (including Inc.) that makes it easy to discover new topics and media sites. SmartNews has over 10 million active users and over 30 million downloads per month in the United States and Japan. And it won Apple’s “Best of 2013” award and Google’s “Best App of the Year.”
Since I like talking to people behind the products and services I like, I spoke with Rich Jaroslovsky, the VP of Content and Chief Journalist for SmartNews. Rich, a former Bloomberg and Wall Street Journal reporter and the founder of WSJ.com, joined SmartNews during its U.S. launch four years ago.
And why is Rich’s title “Chief Journalist” instead of “Editor in Chief”? Let’s find out.
SmartNews takes a different approach to content curation than I expected.
When I met the co-founders, Ken Suzuki and Kaisei Hamamoto, they showed me a prototype of the U.S. app. I immediately liked it, but I told them the next version should be web-based and highly-personalized.
Kaisei said, “We already built that,” and showed me a product they called ‘Crow’s Nest.’ It was highly personalized, highly customizable, had won tech awards in Japan, won a TechCrunch award… but when he and Ken brought it to SXSW to show it off, it was “a complete failure.” It was seen as just one of a number of similar desktop applications.
On the way back to Japan they concluded that while the technology was fabulous… the product decisions they had made were wrong.
So they set out to do the opposite.
People love to talk about pivoting, but that’s a hard pivot to make.
But it turned out to be a really smart one. They made the app completely and natively mobile. And instead of building something designed to appeal to an audience of one through extreme personalization, they built something of interest to a broad audience.
Also keep in mind that much of the Tokyo population rides the subway every day, and at that time the subway didn’t have Internet connectivity. So they also included features that would allow people to consume content without needing an Internet signal.
The resulting app turned out to be a huge success. It was basically a case of combining great underling technology with smart product decisions — decisions they arrived at because their first set of product decisions fell flat.
In short, it was a classic Silicon Valley pivot — but in the Japan of five or six years ago, the concept of acknowledging mistakes and starting over… that pivot took some guts.
Another classic Silicon Valley move was introducing a product to a local market while ensuring the underlying technology was scalable.
From the very beginning, Kaisei and Ken wanted to not just build a Japanese company and a product but a global company and product. The model and technology we use is highly scalable and transferable from market to market.
At a basic level, we analyze about 10 million pieces of content every day and look for a number of different signals. We look at how widely and rapidly an article is being shared. We use machine learning to analyze and classify the content. We rank stories for importance and interest. The underlying technology is incredibly complex.
But that technology is only one piece of the puzzle. Our model is also extraordinarily publisher friendly. It’s in our DNA to be supportive of publishers. That’s why our incentives are very closely aligned with publishers. I spend a lot of time talking to publishers, explaining the model, figuring out ways we can work with them and they can work with us…
And we’re constantly working with our engineers to refine the algorithms to make the app journalistically smarter.
Which is why you’re not “Editor in Chief”?
When I first joined the company they said I would be “VP for Content and Chief Journalist.” I said, “In the U.S., chief journalist is not a common title. Editor in chief is more common. They said, “If you want to be Editor in Chief, that’s fine, but Chief Journalist doesn’t imply that you’re editing; it implies you’re making news selections on a day-to-day basis, and our goal is to make the algorithm smart enough that we don’t have to be editors in that sense.”
I realized they are right. We’re not editing. We’re finding ways our users can more easily discover news and content they enjoy.
Coming from such a strong journalism background, why was the idea of aggregation so appealing to you?
I’ve been around tech and online news for a long time. I started WSJ.com in the ’90s.
So the first time I saw the app, I got it immediately. The product itself was terrific.
I also believe that as traffic moved increasingly from desktop to mobile, aggregation was the winning model.
In a desktop environment, it’s easy to go from website to website. On a mobile device, unless you have an incredibly strong brand where people will go to the trouble of installing your individual app, from a publisher’s standpoint it’s very hard to attract a mobile audience. People won’t launch 10 different news app to get news from 10 different sources.
That’s why aggregation seemed like such a powerful model, a hunch has proved to be correct. Since we had a thriving and successful business in Japan we had the luxury of being pretty quiet in the U.S. for a couple of years while we refined the product, determined how the market was both the same and different… and so it’s only been in the last year or so that we’ve really started to hit the gas.
Our growth is now up 200 percent year over year in the US. The Japanese market is still bigger, but the U.S. market is growing faster. We have over 300 publisher partners in the US, and I hear from many of them that SmartNews has become one of their top sources for mobile traffic.
As a content creator I appreciate the way the app works. Unlike many news sites, when I tap a thumbnail or headline, that takes me to the publisher’s mobile site. The content isn’t native on SmartNews.
We’re Japanese. We’re very polite. (Laughs.) We like to think we’re good partners.
So you’re right: Users go to the publisher’s mobile site and the publisher gets a direct hit on their site. And if the user taps again they can summon up smart view, an instant-loading format version of the content, sort of like “reader mode” in Safari. And it works even if there’s no Internet signal.
So how do you make money?
When I talk to publishers, my slide titled “Revenue Share” says, “You: 100 percent. SmartNews: 0 percent.”
Publishers like that slide a lot. (Laughs.)
Our model does not depend on taking a cut of that revenue. Publishers that partner with us not only get the web traffic in the first instance, but if users do slide over to smart view, the publishers can also run their own ad in the space and don’t have to share that revenue with us.
We make money on the advertising that appears not on individual publisher stories but within preset subject channels. Those aggregated channels have inline advertising, and that’s where the lion’s share of our revenue comes from.
Which means your main incentive is to create satisfied, long-term users. And yet you don’t do that through heavy personalization, which conventional wisdom would say is the way to go.
Instead of building a product for an audience for one, we built a product of mass appeal.
We are very judicious and conservative about how we use personalization within the app. We emphasize personalized discovery, not just serving our users content they already have expressed interest, in or already know about.
We’re trying to introduce the concept of presenting stories you might be interested in.
Here’s a recent example. A story on our top channel was actually a story from Australia about efforts to preserve Tasmanian devils. There is not a personalization engine in the universe that would ever predict I would be interested in a story about saving Tasmanian devils. (Laughs.)
But I read every word of that story.
I would have, too. Who can resist a Tasmanian devil story?
The algorithm said, “A lot of different people are finding this to be interesting, so we’ll elevate that.”
The key is personalized discovery. Of course that also means sometimes users will see stories they don’t like. I’m generalizing, but a conservative might see a Mother Jones story, or a liberal might see something from Fox News.
That’s probably the biggest complaint we get: That the stories are all “left wing,” or “right wing.” But when we look at studies about user engagement in news apps, our audience is far more engaged than that of any other news app.
My argument is that those are two sides of the same coin. As a 40-year journalist, when all you see is stuff you already know or already think you’re interested in… news gets boring. There’s no serendipity. You don’t get to learn anything new. You don’t get to discover.
That’s the textbook definition of a filter bubble.
Our goal is to puncture filter bubbles.
Does that mean you’ll occasionally see something you aren’t interested in? Sure. But if you look at how engaged our audience is, and how much news they consume, I think it’s a reflection of the power of the personalized discovery.
All of which makes you a machine learning company, not a news company. Similar to when I talked with a Domino’s exec who said, “We’re not a pizza company. We’re a delivery company.”
You’re right. If you ask what SmartNews is, we’re not a news company, we’re a machine learning company. Personalized discovery is the product.
The people here are incredibly smart. I genuinely feel like I’m working with rocket scientists. And their brains are so big, they have all this bandwidth to devote to other cool things.
Me? I’m just an ink-stained wretch. (Laughs.)
So: U.S. growth is rapid. What’s next?
The Japanese business is doing great. The trajectory in the U.S. is on a definite upward path.
Again, the great thing about our approach is that it’s scalable; we can replicate it in markets all around the world. So far we’ve focused on U.S. and Japan, and at some point I anticipate we will create additional national market products for specific major markets.
We have publishers clamoring for us to come into their markets because they see we can drive traffic. Every week I get two or three publisher inquiries from India alone.
We already have a number of partners that are non-U.S. based, like the BBC, South China Morning Post, Reuters, DW (Deutsche Welle) in Germany… so when we’re ready to move into those markets, that will certainly help.
We’ve established our bona fides and have established a model that we can show works for users and for publishers.
Then why not take another page out of the Silicon Valley playbook and rapidly expand?
We prefer the slow runway approach. I love the idea of global domination and I’m not apologetic about it. (Laughs.)
But on the other hand, in a startup environment you must have a sense of discipline. Even though part of you is saying, “Come on, let’s go…” you also have make sure you have this working well before you move to that.
And that’s what we’ve done. For a company like SmartNews, there probably was no tougher market outside Japan to crack than the U.S. market. So we took on the toughest task first — and maintained our focus on that challenge.
Now, in the U.S., we have more than 300 partners. Go to the channel directory on the app and look at the list. There are some pretty great names. And the list of new channels constantly grows as we bring in new partners.
Besides the technology and the model, that’s our other secret weapon: Our fantastic roster of partners.
If we can establish ourselves in the U.S., given how tough the environment is with all the travails of the U.S. media industry and all the competition… if we can establish ourselves here, we can do it anywhere.
Which will let us work, slowly but surely, on doing it everywhere.