Priya's Portfolio

Netflix Mood Selector

Netflix Mood- Based Discovery

Redesigning content discovery to match viewers' emotional state in the moment, reducing decision fatigue and increasing engagement.

ROLE

Product Designer

UX Researcher

TIMELINE

3 Weeks

TOOLS

Figma, FigJam, Google Forms

PROJECT AT A GLANCE

Problem

Netflix's algorithm recommends based on history, not current mood. Users browse for 10+ minutes and often leave without watching anything.

Solution

A one-tap mood selector that returns a single curated recommendation, lowest possible friction, zero added cognitive load.

92%

Said the experience felt more personal than browsing by genre.

88%

Said they'd use this regularly if Netflix launched it.

6/8

Completed the full flow without any prompting.

Outcome

92% User satisfaction. 88% of the users would use it regularly. Almost all the users completed the full flow unprompted.

PROJECT AT A GLANCE

Problem

Netflix's algorithm recommends based on history, not current mood. Users browse for 10+ minutes and often leave without watching anything.

Solution

A one-tap mood selector that returns a single curated recommendation, lowest possible friction, zero added cognitive load.

92%

Said the experience felt more personal than browsing by genre.

88%

Said they'd use this regularly if Netflix launched it.

6/8

Completed the full flow without any prompting.

Outcome

92% User satisfaction. 88% of the users would use it regularly. Almost all the users completed the full flow unprompted.

Overview

Research

Design

Impact

Testing

Reflection

01 - PROBLEM

The paradox of too much choice

Netflix's content library is a strength that creates a real friction point: users spend

significant time browsing without finding something that matches how they feel in the moment.

The existing algorithm is history-based — it surfaces what you've liked before, not what you're in the mood for right now. Those are different questions.

🎯 DESIGN QUESTION:

"How might we help users find content that matches their current mood — without adding more cognitive load to an already overwhelming experience?"

02 - COMPETITIVE CONTEXT

How others approach this

Before designing anything, I looked at how other platforms handle mood or context-based discovery.

Spotify

Strong precedent

Mood-based playlists ("Focus", "Chill")

Users don't search - they browse by feeling.

Netflix Today

Gap identified

History- and popularity-based rows

"Because you watched X" is the primary control signal. Mood is not a first-class input.

Apple TV+

Same Gap

Curated editorial collections

Staff picked themes, but no user-driven mood input.

The pattern across competitors: Mood works well as a discovery input, but none of the major video platforms treat it as a primary entry point.

03 - RESEARCH

Survey Findings

I ran a Google Form survey with 40 participants (ages 18–45, all active Netflix users) to pressure-test the problem hypothesis before designing anything.

82%

Spend 10+ minutes deciding what to watch

per session.

74%

Say mood directly shapes what they

want to watch.

67%

Have wanted mood-based suggestions while browsing

~40%

Left Netflix without watching due

to indecision.

Selected verbatim responses

"I scroll for ages, nothing feels right, and then I just give up and go on my phone instead."

— Survey respondent, 27

"I scroll for ages, nothing feels right, and then I just give up and go on my phone instead."

— Survey respondent, 27

"I spend more time deciding than actually watching sometimes."

— Survey respondent, 22

"I spend more time deciding than actually watching sometimes."

— Survey respondent, 22

Research limitation: The sample was convenience-based and self-selected, so findings are directional rather than statistically representative.

04 - IDEATION

What I explored — and what I cut

I used FigJam to map out several directions before committing to any of them.

Mood wheel / sliders

Visually interesting but created a new problem: users had to think too hard before getting any value. Cognitive load test failed.

Discarded

Multi-step Questionnaire

Felt like a survey, not a feature. 3+ steps before a result was too high a barrier for something that should feel effortless.

Discarded

Tappable mood tags

One tap to start, instant feedback. Works with Netflix's existing visual language. Lowest friction path to a result.

Chosen

Single recommendation output

Showing a list reintroduces choice paralysis. One curated result with a "try another" escape valve solved this.

Chosen

DESIGN OBJECTIVES

Under 2 min to a recommendation

Zero added cognitive load

Feels native to Netflix UI

Easy to adjust or undo

The solution

INTRODUCING MOOD PICKS

A mood-based recommendation feature that helps users discover content aligned with their current emotional state. By selecting mood tags, users receive personalized suggestions that match how they feel—turning choice overload into an enjoyable, intuitive experience.

05 - SOLUTION

Mood Picks — a three-step flow

Access

Via "Mood Picks" in the

navigation or a "Need Ideas?" button on the home screen.

Select Mood

Tap pre-defined mood tags or type your own. Toggle between Movies and TV Shows.

Get a pick

One curated recommendation. Play now, get more info, or try another.

Designing the experience:

Hover Element

Matched Netflix's signature red-hover microinteraction so the feature feels native, not bolted on.

Mood Tag Control

Users can edit Mood tags - add or remove with a single tap.

Loading Animation

Loading state uses the Netflix 'N' with a subtle zoom-in/zoom-out glow animation — on-brand.

06 - DESIGNS

Home Page

To use the content recommendation feature, users can click "Mood Picks" in the top menu. Or click the "Need Ideas" button.

Mood Selection Page

The users can choose the tags they like or type in what they want to watch. Users can also choose between Movies/TV Shows.

Recommendation Page

Displays a curated title (e.g., Oldboy) with quick options: Play Now, More Info, or Try Another.

07 - TESTING

Usability testing — what worked, what didn't

I ran moderated usability tests with 8 participants using a think-aloud protocol.

92%

Said the experience felt more personal than browsing by genre.

88%

Said they'd use this regularly if Netflix launched it.

6/8

Completed the full flow without any prompting.

92%

Said the experience felt more personal than browsing by genre.

88%

Said they'd use this regularly if Netflix launched it.

6/8

Completed the full flow without any prompting.

08 - REFLECTION

Mood Picks — a three-step flow

92%

Said the experience felt more personal than browsing by genre.

88%

Said they'd use this regularly if Netflix launched it.

6/8

Completed the full flow without any prompting.

The strongest design instinct I developed on this project: resist the urge to add. Every complex idea I explored failed because it asked users to do work before they got value. The simplest version, tap a feeling, get a result — was also the one that tested best.

If I were to continue this: I'd run structured interviews to explore edge cases (users who can't name their mood, users who want to deliberately break their pattern), test whether the single-recommendation format holds up over repeated use as novelty fades, and explore how mood data could feed back into Netflix's broader recommendation engine without feeling intrusive.

The biggest open question is what happens at session 10, not session 1. That's the research I didn't get to do in 3 weeks.

Self-initiated concept project. No affiliation with Netflix.