Skip to main content

The science behind Netflix’s first major redesign in four years

The science behind Netflix’s first major redesign in four years


How Netflix is using a mountain of data to get you to watch more stuff

Share this story

Back in March, a developer named Renan Cakirerk wrote a small piece of code that made a big impact on Netflix. Cheekily named "god mode," it addressed one of the most annoying aspects of trying to use Netflix in your browser: scrolling through the company's ever-growing list of movies. Once enabled, it would simply give you one, big list. Instead of sitting there, holding your mouse in anticipation, you could simply find the title you wanted and get on with watching.

The web lit up with stories hailing it as a crowning achievement of little-guy ingenuity. TechCrunch called it a fix for a "wonderful problem." CNET dutifully noted that it "saves you from the slow monotonous horizontal scrolling." Mashable, meanwhile asked, "Are you listening, Netflix?"

Inside Netflix, though, god mode was old news. In fact, the company had already tried it out on thousands of unsuspecting users years earlier. And it was a total disaster.

"Who the heck has time to go through every title?"

"It's the difference between what people say they want, and what they actually want," says Todd Yellin, Netflix's VP of product innovation. "Consumers say they want to see every title in a catalog, but who the heck has the time to go through every title?"

Netflix began testing it on small portions of its user base — typically new users, so as not to confuse existing ones. It was as if they'd been given a 60-page restaurant menu. Instead of making people watch more, they ended up watching less. "We're guessing that people were just overloaded," Yellin explained to a group of reporters at Netflix's Los Gatos, California headquarters this week.

But Cakirerk wasn't wrong in trying to improve that aspect of Netflix, and Netflix wasn't oblivious to the fact that its old system was just plain bad. In fact, a fix for the issue is one of a few subtle but substantial changes people will notice when they start using a new version of in browsers over the next few weeks. The company began rolling that update out to some of its users this week, and will have it out to everyone else by early next month.


At first blush, the new design doesn't seem markedly different. It's got a black background now, to match what people are used to on their phones, tablets, and TV sets. Those little arrows that scroll through the service at the speed of a glacier are still there, but now they jump between entire rows of choices. And the service does a much better job of letting you see information about a show as you click around, instead of accidentally playing something you only wanted to know more about. Under the hood though, the changes are the culmination of years of research aimed at gleaning every nuance about how humans hunt for things to watch. Netflix has been tossing out breadcrumbs in various configurations, and seeing how we gobble them up. This is the newest handful for us to taste test, and it comes with the hope that we'll feast.

To aid in that ongoing psychology project, Yellin and his team are gathering and analyzing a mountain of data, then twisting the dials on what we're all seeing in real time. By the time you see the cover for the next season of House of Cards, it likely will have already gone through several rounds of virtual focus groups to see which design drew the most intrigue. And you yourself could be an unknowing participant in countless other tests designed to get you to watch more, without ever making you feel like you're being led astray.

"It's not good enough just to see what they click on, because you can show the most sensationalistic thing: show a lot of sexiness on the cover, even though there [are] no sex scenes. Then they'll hit play, but they won't watch it," Yellin says. "We're not just looking for clicks here, because that's not a good metric. We're looking for finding the right people to watch the show, because we want to promote our shows to the right people who will actually play it through."

Netflix has snuck fake shows in research emails

Changing covers is one of the least crazy things Netflix is doing to gather data. Last year, the company sent a dozen researchers out to conduct more than 1,500 interviews in people's homes to understand how they use Netflix. They also sent out 15 million email surveys, sometimes with fake shows and star ratings to determine which is more effective in getting someone to watch. The exercise has the potential to tell Netflix any number of things, like if people really care about half star differences in ratings, down to if they're still checking their email.

Yellin says that while Netflix is paying attention to what people ask for in these surveys and feedback requests, it ultimately spends far more of its energy watching what they're doing on the service. "Most of our personalization right now is based on what they actually watch, and not what they say they like," he says. "Because you can give five stars to An Inconvenient Truth because it's changing the world, but you might watch Paul Blart: Mall Cop 2, three times in a few years ... so what you actually want and what actually say that you want are very different."

The stakes in all these experiments are high. There has never been more competition in online video; between HBO Now, Amazon Prime Instant Video, and Hulu, retaining customers has never been more challenging. Netflix's goal is to keep its 62 million existing users there month after month, while also converting everyone who completes a free trial into a paid subscribers. By changing just one aspect of the service across groups of as many as 100,000 new members, Netflix tries to spot differences that get them to pay. Which should make it clear — Netflix's new design may be the culmination of past research. But it's also the beginning of another test.