Consumer app
Pic-A-Boo
- App design
- UX research
- MVP validation
- Role
- Industry
- Year
- Product strategy, App design, UX research
- Consumer app
- 2023
Project overview01
Choose your best photo with help from friends.
Choosing between eight near-identical photos is a genuinely common problem, and the existing answer is spamming a group chat. Pic-A-Boo makes it a poll: create it, share a link, get an answer.
We led the design and the research behind it. The interesting work was scope discipline: deciding what this product refuses to be.
The challenge02
Every feature we could imagine pulled towards becoming a photo platform. Accounts, albums, profiles, history. Each one would have killed the thing that makes it useful, which is that a friend can vote in forty seconds without installing anything.
At a glance03
The problem, and what changed
Why it stalls
Choosing a photo happens in group chats that were never built for a decision: images scattered across replies, opinions to reconcile by hand, no result at the end.
Decision flow
Voting takes no account and no download, polls expire after 72 hours, and a question is attached to every poll so the answer carries context.
Voting surface
The voting screen was designed before the creation screen: swipe, comment, mark the best, then a result that shows who chose what and why.
Scope of work04
What the work covered
No account for voters, ever
The person voting is doing you a favour. Anything that asks them to sign up spends goodwill you do not have.
Constraints that keep it a decision, not an archive
Eight photos per poll and a 72-hour lifespan. Both limits keep the product a temporary decision tool, which is the behaviour worth validating first.
Design the voting screen before the creation screen
Most people meet this product as a voter, not an author. That screen got the attention.
Results that read at a glance
Best and second best, with the voters who chose them. Enough to decide, not enough to argue about.
What we built
A mobile upload flow, poll creation with a custom question, a shareable link, and a voting experience where the voter only types a name. Comments live on each photo, the poll expires after 72 hours, and the MVP caps a user at ten polls a month.
Voting without an account
The invited friend opens the link, sees who invited them, and starts voting. No signup, no download. Every second of friction before the first vote is participation lost, so the flow removes all of it.
Results with the reasoning attached
A poll does not end in a ranking. Each photo shows who picked it and what they said, and the creator sees the winner together with the argument for it. Knowing why a photo won is worth more than knowing which one did.
Outcome
- A defined MVP with deliberately narrow scope: a social decision tool, not a photo platform.
- The 72-hour window and poll limit make real behaviour measurable before anything scales.
- A complete flow designed end to end, from empty state to shared result.
