PEAK6 / Poker Power

First-week retention, tripled.

I rebuilt the app's navigation and designed a guided first-game experience, so new members learn poker by playing a real hand instead of reading about it.

RoleStaff Product Designer
Timeframe2023 – 2025
TeamSole designer across 3 pods
PlatformMobile app

Sequence of Poker Power first-game screens, each with a short coaching callout over a live table

The complexity

A live table is intimidating.

Poker Power teaches women confidence, risk-taking and negotiation through poker. Most new members have never played.

The app mixed playing and learning in a way that confused them, and too few started, or finished, a first game.

New membersMost have never played, and a live table is intimidating.
InstructorsReal-time control of a live game while they teach, without hiding the table.
BusinessA new membership launching at the same time: videos, lessons and daily tasks.
TeamOne designer, me, across three product pods.

How I untangled it

I built one reusable design system and brand identity that carried the same look, feel and UX across the existing app, the new membership features, the marketing website and even our live events. A shared set of design principles made three pods' work feel like one product, and gave every new feature a foundation to build on.

Key decisions

What I changed.

01
Separate Play from Learn

A global tab bar gave each mode its own home, so new members always knew where they were and what to do next.
02
Teach inside a real hand

With my PM, I designed a choreographed, pop-up-video-style first game that walks players through a simulated match one rule at a time.
03
Bring them back on their terms

Players choose when to be nudged about community games, and the core table got more contrast and less friction.

The tradeoff

How to teach a first game.

Option A

Explain the rules

The obvious tutorial. But my hypothesis was that people don't want to read how to play poker. They can do that anywhere online.

Option B

Drop them into a game

Learning by doing works, but bots alone don't teach, and user feedback kept telling us a table of real people is intimidating.

Option C · Chosen

A guided experience

A real game against bots, choreographed like pop-up video, with hints and tips right where the action happens on the table.

Why

I didn't want story-style screens the whole way through, and some ideas need more room than a tooltip. So I blended two formats: larger explanations overlaid on part of the table, and bright tooltips that point at exactly what they're teaching. The result is a 17-step game that takes about three minutes.

The tradeoff: it took less engineering time, but moved work onto my PM, who mapped every hand with a poker instructor so each teaching moment landed in the right order without getting wordy. We got a repeatable experience we fully controlled, and new players got the feel of a live game. It drove +86% game starts and +82% first-game completion.

Shipped