Emerging TechnologyAIMachine learning and choice paralysis: How Netflix personalizes title images

Machine learning and choice paralysis: How Netflix personalizes title images

A sophisticated recommendation engine has been a crucial part of Netflix's success. Using machine-learning algorithms, Netflix personalizes everything on your homepage, right down to which title images you see.

The visual cortex takes up 30% of the human brain, so people naturally process images 60,000 times faster than text. As a result, we’re all guilty of judging books by their covers, which is a crucial element of Netflix’s engagement strategy.

Netflix has a seemingly unlimited supply of content, which can be a double-edged sword. That volume of content produces choice paralysis, where people have so many options that they can’t decide what to watch. They ultimately don’t watch anything, which underlines the importance of Netflix’s sophisticated recommendation algorithm. Netflix customizes everything on the homepage, right down to the title images.

“We don’t have a product. We have hundreds of millions of products because we deliver personalized experiences,” says Tony Jebara, Director of Machine Learning. “We’re looking to not just make recommendations, but have members believe in them.”

AI driving engagement

Machine learning drives Netflix’s algorithms, which play a huge role in the company’s success. By recommending the most relevant content, they increase engagement significantly, saving the company an annual $1 billion.

The algorithms dictate the personalized homepages and “top picks,” taking each subscriber’s viewing habits and preferences into consideration. The company utilizes a process called interleaving, which identifies the most promising ranking algorithms from a large set of initial ideas and then A/B tests using the pared-down algorithms. Compared with traditional A/B testing, interleaving works faster with smaller sample sizes.

“You end up kicking yourself because during the whole period of innovation, you’re not giving the best possible machine-learning algorithm to the users,” says Jebara.

What does this mean for title images?

Netflix used to use the generic title images, provided by studio partners. They were often scaled-down versions of DVD cover art that didn’t necessarily fit. The company has since performed tests to determine the images most likely to catch people’s limited attention. Engagement metrics include click-through rate, aggregate play duration, and what percentage of views had short durations.

Images with expressive facial emotions perform well, as do those featuring particular people. For example, Unbreakable Kimmy Schmidt got the best engagement with an image including both the titular character and fan favorite Tituss Burgess.

What performs even better are the images that are specifically selected for you.

“Let’s say Good Will Hunting is recommended. If we know someone really loves comedies because they’ve watched Zoolander and Arrested Development, the title image might have a picture of Robin Williams,” explains Jebara.

Do you watch a lot of movies with John Travolta or Uma Thurman? That factors into which Pulp Fiction title image you see.

A quick experiment

At The Next Web’s TNW event, Jebara’s presentation included visuals. But they were the “official examples” from Netflix. I was curious, how did this factor into my own personal Netflix homepage?

As a Shameless fan, I searched for William H. Macy movies, like Fargo and Magnolia. None were available to stream, but even if they were… William H.Macy is a big enough star that he might be in the title image anyway. So I decided to focus on the filmographies of less-established cast members.

Mall is 2014 movie I’ve never heard of about disaffected suburbanites who come together in a mall following a shooting spree. Look at the difference between Mall‘s movie poster and my title image on Netflix. Guess what TV show the redhead is on.

Netflix’s gateway

The Netflix experience revolves storytelling, in a way.

“Storytelling is the heart of the human race,” says Jebara. “It’s how we share information, pass on our culture and teach the language. It started with cavemen drawing and now we’ve got PowerPoint.”

Netflix views title images as “the gateway to the stories” and the machine-learning algorithm serves the right ones for each user.

Whitepapers

US Mobile Streaming Behavior

Whitepaper | Mobile US Mobile Streaming Behavior

4m
Winning the Data Game: Digital Analytics Tactics for Media Groups

Whitepaper | Actionable Analysis Winning the Data Game: Digital Analytics Tactics for Media Groups

4m
Giving a Voice to Your Brand

Whitepaper | AI Giving a Voice to Your Brand

4m
Mobile Messaging Masters

Whitepaper | Mobile Mobile Messaging Masters

4m

Related Articles

Five ways to up your social media ante with conversational AI

AI Five ways to up your social media ante with conversational AI

4d Kartik Walia
Practical personalization: Adobe’s tips for creating meaningful moments with AI and automation

AI Practical personalization: Adobe’s tips for creating meaningful moments with AI and automation

2w Emily Alford
Deloitte ad agency Heat launches AI practice to predict near-term online trends

AI Deloitte ad agency Heat launches AI practice to predict near-term online trends

4w Barry Levine
Is martech a threat to marketing jobs?

AI Is martech a threat to marketing jobs?

2m Luke Richards
Personalization, segmentation, testing: How email marketers can use AI

AI Personalization, segmentation, testing: How email marketers can use AI

2m Mike O'Brien
Benefiting from AI and deep learning for video summarization

AI Benefiting from AI and deep learning for video summarization

2m Divya Jain
CMO spend on analytics expected to double in next three years, despite overall economic uncertainty

AI CMO spend on analytics expected to double in next three years, despite overall economic uncertainty

3m Luke Richards
Salesforce announces new Einstein features for devs to build custom AI and ML

AI Salesforce announces new Einstein features for devs to build custom AI and ML

3m Kimberly Collins