StreamNow
Your next watch in under 90 seconds.
People spend around 45 hours a year just deciding what to watch. Streaming catalogs are huge, browsing UIs are optimized for inventory rather than decisions, and the result is decision paralysis — you scroll, you sample trailers, you give up, and your downtime shrinks. The design question: how do you get someone from “opening the app” to “watching something they'll enjoy” in under two minutes?
Decision fatigue is the real product problem
People scroll, sample trailers, second-guess, and give up — the catalog isn't the bottleneck, the deciding is.
Choosing is social
A large share of 'what should we watch' happens in groups — the decision is a negotiation, not a query.
Context beats catalog
Mood and available time predict what gets watched better than genre taxonomy does.
≈45 hours a year lost to scrolling, sampling, giving up.
One 90-second match flow — three confident picks.
Quick Match — under 90 seconds end to end.
Quick Match over endless browse
The core flow asks for minimal input and returns exactly three recommendations in under 90 seconds. Three, not thirty — the constraint is the feature. Every additional option re-introduces the paralysis the product exists to remove.
Taste Match for users who want depth
A second, longer path covers directors, actors and previously watched titles, so recommendations compound over time. Casual users never see it; invested users self-select into it.
Mood and duration as first-class filters
What you watch depends on the evening you're having, not on the catalog taxonomy. Mood and available-time filters sit directly in the matching flow instead of buried in settings.
Watch Party for shared decisions
Half of “what should we watch” is a group negotiation. Watch Party lets multiple people feed the same match, so the recommendation settles the debate.
- ✕ Rows of thumbnails with no opinion
- ✕ Trailers as commitment tests
- ✕ Catalog-first navigation
- ✕ Group nights become remote wars
- ✓ Exactly three recommendations
- ✓ A 90-second match flow
- ✓ Mood + duration come first
- ✓ Watch Party settles the debate
The final design takes a user from open to a confident pick in under 90 seconds. The flow was prototyped at low and high fidelity and refined through user testing across the matching, filtering and watch-party flows.
- Iterating from user feedback beat my initial assumptions more than once — the testing rounds reshaped the matching flow.
- Thinking like a business: which features earn their place against monetization and build cost, not just user delight.
