Netflix / Improving Recommendations (Fan Redesign)

Dive into my fan-driven passion project where I tackled Netflix's watching recommendations. Driven by an imagined goal of making Netflix the go-to streaming platform again, I ran a small study and prototyped a new feature to help Netflix users discover new titles with more affinity and satisfaction.

Incidentally, this feature was later introduced in Netflix's official app. Let’s take a look at the work!

Stemming from a conversation during an app critique interview, I wanted to imagine a world in which Netflix takes fuller advantage of its user data in order to provide better recommendations. To begin with, I ran some validative research both quantitative and qualitative to confirm that this was, in fact, a real and valuable enough people problem to tackle.

It turns out, as suspected, that few actually use recommendations from the platform. For the general population, watching was influenced by recommendations from friends, while for “filmies” (like foodies) the attributes of a film or series were equally important. 

That got me thinking about Pandora Radio, a music streaming service that created a complex, internal analysis operation called the “Music Genome Project” where they hired actual musicians to parse all the titles in their library into “genes,” or musical attributes.

Some quick secondary research to validate that listenership is driven by recommendations on music streaming platforms, particularly Pandora, and I was ready to get to work on a new front-page Explore feature to facilitate filmies’ discovery of relevant, impactful recommendations. 

Interestingly, while this was a self-directed pitch project, Netflix actually introduced this feature directly in their official app later that year. I must have been on to something!

If you want to know more, please contact me and let’s chat!