Choose your own
adventure
The smartwatch's onboarding was too long, too rigid, and failing dogfooders. Nobody had a plan to fix it. I proposed a ground-up redesign close to feature complete — and got it shipped.
A broken onboarding.
No plan to fix it.
The smartwatch's onboarding flow was too long, too dense, and too rigid. Users had to slog through a fixed sequence of setup screens — device pairing, LTE connectivity, permissions, app configuration — before they could use the watch. Dogfooding feedback was brutal. The UX KPI score sat at 3.6, which meant the experience was a moderate risk for launch. Anything below 4.0 was considered borderline unacceptable.
The team knew it was a problem. Nobody had a plan to fix it.
We were close to a major launch milestone — feature complete. Proposing a ground-up redesign at this stage was viewed as risky. Engineering would need to rebuild navigation logic, design would need to reflow every screen, and the whole thing would need to clear design reviews under significant time pressure.
Research first.
Then the pitch.
I audited competitors' onboarding flows — Fitbit, Apple Watch, Pixel Watch — and discovered ours was significantly longer and denser. Too many mandatory screens, too much friction. The data gave me something to stand behind.
I designed a concept from scratch: a "Choose Your Own Adventure" model where users could complete setup tasks in whatever order they wanted, skip what they didn't need, and get to the watch faster. The mandatory setup was shortened; everything else became optional and self-paced.
I framed it as a "bad first draft" when I presented it to the team — intentionally lowering the stakes so people would engage with the idea rather than defend the status quo. But the concept was solid enough that it immediately changed the conversation.
Then I did the hard part: I built the case. I used the competitive audits and wireframed the full design concept to show the team it was worth the engineering investment. I brought XFN partners together, navigated the reluctance (and some real conflict), and got it greenlit through design reviews under deadline pressure.
Modular onboarding.
Self-contained modules.
A modular onboarding architecture that separated mandatory setup from optional configuration — so users could get to the watch faster without skipping anything critical.
A content framework that gave each setup module its own self-contained narrative — so users could enter any module without needing context from previous screens.
A revised permissions flow that combined, relocated, and removed permissions to reduce friction — validated with Legal, PXFN, and engineering before every change.
The concept traveled
further than the product.
The UX KPI score improved from 3.6 to 3.9 — moving the experience from "moderate risk" to "borderline acceptable for launch." That's not a huge number on paper, but in the context of a product close to feature complete, it was significant.
The bigger outcome: the CYOA concept was adopted by a sister product's onboarding — Ray-Ban Meta AI glasses. Dogfooding feedback from 6 of 9 testers preferred the new prototype, citing CYOA as the reason. They liked that it was shorter, and liked the "Choose your own adventure" style that allowed them to skip anything they didn't want to set up.
Early dogfooders on the smartwatch said CYOA was preferable because "it got them to the watch sooner." That's the whole point.
"You proposed and led a ground-up overhaul of the OOBE flow that simplified the experience and improved its UX KPI score from 3.6 to 3.9."
— Manager review"Your 'Choose Your Own Adventure' concept was also adopted for the Ray-Ban Meta AI glasses OOBE prototype, which dogfooding feedback suggested was a clear improvement over the glasses OOBE due to its brevity and optionality."
— Manager review"Ashlee drove the decision for the NUX team's OOBE navigation revamp. She ensured we got consensus, structured our rationale and communicated it out."
— Product design peerI almost didn't
propose it.
I wasn't sure it was "a battle worth having" — a peer called this out directly in my review. The concept only traveled because I eventually committed to the pitch, but I lost time hesitating. The lesson was that if you've done the research and the data supports the change, trust your judgment earlier.
Don't frame good work as a "bad first draft" to make it feel safer. Own it.