Social MVP for Netflix's Desktop Website
Creating a more meaningful viewing experience through user-recommended content
The task was to create an MVP that would increase social activity and sharing on Netflix’s site between users in order to maintain majority market share*.
As a 3-person team, we conducted research and created a high fidelity interactive prototype over one 2-week sprint. My role was to conduct market research, develop and conduct user interviews and surveys, synthesize research data, refine wireframes, and assemble the prototype
Tools and methods
Design strategy and concept design, user research, interviews, and surveys, affinity mapping, empathy mapping, rapid ideation and sketching, wireframing and prototyping, usability testing, Sketch, Invision
*Not affiliated with Netflix
Users are dissatisfied with what Netflix suggests for them to watch.
Create “buzz” between users through new features to add friends, recommend content, and receive notifications about friend requests and recommendations.
Success would be measured by users who go long periods without using Netflix using it more frequently (KPI: increase viewing by 5 hours per week).
First, we had to get a feel for what the user problem really was. Through user interviews (7) and a survey (309 responses), the main pain points we discovered were:
- Less than satisfactory organization and selection of content
- No customization of lists
- Poor Netflix suggestions leads people to either stop browsing, wait until there’s buzz, settle for something less
- They want to find something to watch but can’t
- Netflix % match is not reflective of user interests
- Users who use Netflix less than 15 hours a week are over 80% more likely to unsubscribe
Which of these problems could be easily solved with a simple MVP?
Users are having trouble finding something good to watch on Netflix
We synthesized our user research into 2 personae:
A user who will browse and watch Netflix everyday, despite sometimes having to settle for inferior content
A user who knows exactly what they’re going to watch, but may go long periods without watching
They're interconnected. The gatherer is browsing and finds content they like, then tells the hunter about it, who then seeks it out.
The hunter is our primary persona, since Netflix stands to gain the most from them: if they trust the recommendations that they get from their friends and family, then they will spend more time watching Netflix.
Our goal was to find a way for users to suggest content to their friends and family from within the Netflix ecosystem.
Designing the MVP
Choosing a primary device:
- TV most used but not conducive to social
- Laptop/desktop second most common our choice for MVP
- Mobile phone fastest growing
Our MVP had to cover flows to:
Add a friend
Receive notifications about friend requests or recommendations
Adding a friend
Our first step in creating this flow was to figure out where to put the social component. We decided that adding a “Friends” button at the top left of the screen next to “Browse” and “DVD” was as good a place as any. Usability testing revealed no issues with this placement. Once a user clicked on “Friends”, they would be prompted to add their first friend by searching for the handle created within each user’s unique profile. Avoiding integration with Facebook or any other social network ensures that this social feature remains intimate among trusted family members and friends.
After some quick-and-dirty guerilla usability testing, we determined that most of our users didn’t know what the majority of the icons in the video player did, and therefore we felt confident in moving the “report playback issues” icon to the upper right of the player and putting a new social icon in its place. Most usability testing subjects didn’t even notice that that’s not where it had always been. Clicking on the icon brings up a box where users can search their friends or choose from a list of people who they’ve recently interacted with to recommend content.
In order for users to view and act on notifications about friend requests and recommended content, we added new notifications to the existing bell icon.
Iterating for Function, Consistency, and Clarity
Allow users to add a second friend
- Added "Add friends" button to "Friends" page and corresponding flow
Reduce friction in Recommend Content flow
- Simplified steps
- Added button to allow users to cancel
- Changed "Recent Recommendations” to “Recent Interactions” to avoid confusion
Replaced friend search results in "stack view" (right) with more traditional "list view" for familiarity and clarity
Reworked “add a personalized note” box when sending a recommendation to decrease friction
Not pictured: ongoing tweaks in wording and syntax for clarity and consistency
How well did we solve the problem?
"It would add on to the experience of Netflix for me."
"I was sure I didn’t want this, but now that I see it, I really really want it."
Reflections and Next Steps
The feedback from our testers indicated that this feature could be very useful for them, and given more time, we would continue usability testing to achieve the most frictionless experience for them.
- Flow to choose handle
- Build out mobile, tablet, and TV versions
- Feature to allow users to recommend to multiple friends
- Group watch feature
- Additional usability testing