Home

About

Projects

Download

Home

About

Projects

Download

Shipped 2020 · Mobile app

Reducing Decision Friction to Increase Playback

When people are tired, choosing what to watch is hard. A homepage feature that made the decision easier and got more people to actually press play.

Role

Concept · Flow · UI · Prototype

Timeline

Approximately 2 months

Team

Devs · UX Designer · Product Owner · Data Analyst

Methods

Analytics · Interviews · A/B testing

Outcome

Reduced decision friction

Overview

The gap between browsing and watching

The problem

Many people looked around for a while, then left without playing anything.

The goal

Gapfilm wanted more people to actually watch, not just browse.

Why it mattered

When no one presses play, the content goes unwatched and the app loses its point.

My role

What I owned and what I contributed to

I owned

The core feature idea: helping people choose what to watch based on mood, instead of comparing titles.

The mood-based structure behind it: Rewind, Feel, and Escape.

The user flow for the feature.

The Figma prototype, which I built myself.

The final screen design for this feature.

I contributed to

User interviews, where I joined as an observer alongside another UX designer.

Reviewing the A/B test results with the team.

Shaping how existing data insights could support the experience.

What the data showed

People browsed, but didn't watch

Our data analyst found a consistent pattern in the analytics: people spent real time inside the app, but often left without pressing play. Four behaviors stood out.

Long Scrolling

They scrolled through large parts of the catalog without choosing anything.

Going back and forth

They opened title after title, comparing them.comparing themcomparing them.

Few plays

They were interested, but didn't start watching.but didn't start watching.

Watching trailers

They watched trailers, but still closed the app without starting a movie.

Talking to users

Why people don't press play

To understand the why behind the behavior, we ran user interviews. Four reasons came up again and again:

Mentally tired

Overwhelmed

Unable to decide

Afraid of wasting time

People weren't struggling to find something. They were struggling to decide.

There was plenty to watch. The hard part was choosing. So the goal shifted: not help people search, but help people decide with less effort.

The challenge

From browsing to watching

A tired user doesn't need more options. They need an easier way to start watching with the full catalog still available. That gave me four constraints to design against:

User control

It is critical that users feel in control of their choices at all times.

→ The experience must remain predictable and easy to exit at any moment.

Mental state

Users are mentally tired and unwilling to evaluate options.

→ Content should minimize comparison and reduce decision pressure.

Low-effort interaction

Users are not motivated to explore complex interfaces.

→ Interactions must avoid filters, menus, or multi-step flows.

System balance

The platform should not reduce access to the rest of the content catalog.

→ Users should still be able to access the full catalog whenever they choose.

Exploration

A few ways to make deciding easier

Show fewer titles

Reduce the number of films shown on the screen, so users have fewer options to compare. It makes the choice easier, but part of the catalog becomes less visible.

Many people looked around for a while, then left without playing anything.

Stronger suggestions

Reduce the number of films shown on the screen, so users have fewer options to compare. It makes the choice easier, but part of the catalog becomes less visible.

Many people looked around for a while, then left without playing anything.

A simpler choice

Instead of asking users to compare many titles right away, I kept the full catalog available and introduced simpler ways to begin. It lowers decision effort while keeping users in control.

Many people looked around for a while, then left without playing anything.

Concept

Three simple choices

I designed a flow that helps users start from their current need, rather than from comparing a large catalog.
The flow begins with one simple question:

“What do you need tonight?”

Rewind

For familiar, low-effort watching. Predictable stories help you settle in without thinking too much.

Many people looked around for a while, then left without playing anything.

Feel

For when you want to feel something. Emotionally engaging stories can help you process, reflect, or release.

Many people looked around for a while, then left without playing anything.

Escape

For switching off from daily stress. Immersive stories pull you into another world.

Many people looked around for a while, then left without playing anything.

The flow

A short, simple path

Pick one

A suggestion

Play

Approach

From comparing to choosing

scroll · compare · repeat

Browse and compare

What do you need tonight?

Rewind

Feel

Escape

Choose by need

How the suggestion was chosen

The app already had a recommendation system based on watch history, usual viewing length, and time of day.
My work was to design the choice flow on top of it, so users could start from simpler entry points instead of comparing the full catalog.

Testing

Comparing the old and new homepage

We ran an A/B test: the existing homepage as the control, the new mood-entry flow as the variant. The task was identical on both: open the app, find something to watch, and start it.

The metrics mapped straight back to the four behaviors we'd seen in the analytics, so we were measuring the real problem instead of a proxy:

Browsing Time

how long users spent browsing titles before choosing something to watch.

Time to first play

how long it took from opening the app to pressing play.

Many people looked around for a while, then left without playing anything.

how long it took from opening the app to pressing play.

Playback rate

how often browsing led to an actual movie play, not just a trailer.

Many people looked around for a while, then left without playing anything.

how often browsing led to an actual movie play, not just a trailer.

Before

Escape

Feel

Rewind

After

The result

Outcome

Less hesitation

Faster decisions

Less browsing

More playback starts

Overall view

What it didn't solve

Some people still preferred to browse on their own.

The choices were optional, so they lost nothing, the full catalog stayed one tap away.

The words meant different things to different people.

"Rewind" might be a comedy for one person and a family drama for another. Clearer labels would help.

Without fresh content, the same suggestions could start to feel repetitive.

Rotating them would be a sensible next step.

What I learned

This project changed how I scope a UX problem. I first thought the homepage needed to be easier to use; the data showed the actual barrier was decision-making, not navigation. Learning to separate those two what's a UI problem versus what's a cognitive one is something I still rely on. The project also showed me a blind spot: I treated the labels as copy to finalize at the end, and they turned out to be the part most open to misreading. Now I test wording alongside the flow, not after it.

This project changed how I scope a UX problem. I first thought the homepage needed to be easier to use; the data showed the actual barrier was decision-making, not navigation. Learning to separate those two what's a UI problem versus what's a cognitive one is something I still rely on. The project also showed me a blind spot: I treated the labels as copy to finalize at the end, and they turned out to be the part most open to misreading. Now I test wording alongside the flow, not after it.