UPDATED 21:26 EST / NOVEMBER 03 2015

NEWS

How a data scientist sold Twitch on a ‘recently watched’ tab

Amazon-owned video game livestreaming platform Twitch.tv has enough viewers to rival some television channels, which is quite a feat for a site that (for the most part) specializes in live content focusing on a very specific niche. It would be easy to assume that Twitch viewers are interested in watching all sorts of games, but according to Twitch data scientist Danny Hernandez, the opposite is actually true, a realization that changed the way Twitch approaches content discovery.

“Recently watched came up a couple of times in the past as a ‘nice to have’ project, a.k.a. another one for the ‘maybe never’ pile,” Hernandez explained in a post on Twitch’s official blog. “After all, recency is everywhere … But it wasn’t clear it’d be valuable for Twitch until I asked our data the right question. How much of our viewership is already on recently watched channels?”

Hernandez said that the more he thought about it, the more a recently watched tab seemed like a perfect fit for Twitch, but first he had to prove that the feature would be worth it.

Proving it with viewer data

“With my increased belief in recency came clarity as to the next step. I needed to find out how much viewership was already on recently watched channels,” Hernandez said. “It only took a day to build up Recency in Redshift, AWS’s data warehouse. Half of what people watch are channels they recently watched. Boom!”

Hernandez explained that the actual percentage is even higher when you only look at users who have been using the site long enough for recently watched data to be relevant. He looked deeper into the data to prove his hypothesis and found that while the site’s Follow feature is useful for power users (whose Following page shows at least five live channels at once), those users account for only 30 percent of the site’s total user base, roughly the same number of users who do not log in to the site at all.

“Recently watched isn’t just strong, it’s strong when compared to the incumbents,” Hernandez said. “Recently watched also requires less space than popular, so it’s easy to integrate. The most recent live channel is almost 4x as useful as the most popular channel, so it’s perfect for limited, highly visible real estate.”

Hernandez said that within a week after sharing his findings internally, Twitch began planning an experiment using a recently watched feature, and it quickly found that its treatment group increased their viewership by 1.1 percent. While that number may not sound very impressive, Hernandez explained that it is actually one of the site’s most successful experiments ever.

“Twitch has moved the needle more with huge projects, like launching an iOS app,” he said. “But we only get to do that once.”

Screenshot via Twitch.tv

Since you’re here …

… We’d like to tell you about our mission and how you can help us fulfill it. SiliconANGLE Media Inc.’s business model is based on the intrinsic value of the content, not advertising. Unlike many online publications, we don’t have a paywall or run banner advertising, because we want to keep our journalism open, without influence or the need to chase traffic.The journalism, reporting and commentary on SiliconANGLE — along with live, unscripted video from our Silicon Valley studio and globe-trotting video teams at theCUBE — take a lot of hard work, time and money. Keeping the quality high requires the support of sponsors who are aligned with our vision of ad-free journalism content.

If you like the reporting, video interviews and other ad-free content here, please take a moment to check out a sample of the video content supported by our sponsors, tweet your support, and keep coming back to SiliconANGLE.