Lead Product Designer · 2024 · Real Estate
Lifting Find Your Home completion by 31pp for a Top-11 U.S. homebuilder
Find Your Home completion rose from 60% to 91%, and Configure & Buy from 80% to 94%, measured on the same 37-person study before and after.
Oakwood Homes — a Top-11 U.S. homebuilder — was losing first-time buyers mid-journey on a fragmented digital buying flow: plans, packages, financing, and forms that asked for six-figure confidence without earning it. As Lead Product Designer I owned the Find Your Home and configure-and-buy experience end to end. Constraint: ship a redesign that sales and engineering could defend, measured against the same 37-person study before and after.
01 · The problem
Four in ten buyers never finished finding a home
Stakeholders asked for better filters and stronger buttons. The baseline said something more specific. In a controlled study with 37 buyers aged 26–49, only 60% finished “find an available home” and 80% finished “configure and buy,” and the time people took varied widely on both. The brief described the screens. The numbers described a confidence problem.
That gap between the brief and the evidence became the real design problem: polish the screens the organization already believed in, or reframe the work around where confidence actually broke.
02 · What I found
People stalled between steps, not inside them
Before designing anything, I re-ran the study, recorded every session and mapped where people paused. They didn’t stall inside package comparison or form fields. They stalled between steps: finishing a kitchen package, hunting for what came next, going back to check whether the price had updated, then second-guessing the choice they’d just made.
The filters and buttons in the brief were real problems, just not the ones blocking completion. Drop-off clustered where the interface stopped explaining the next commitment, not where forms were hard to fill.
03 · Key decisions
Three calls that shaped the redesign
Decision 1
Redesign the handoffs, not the filters
I pushed to redesign what happens between decisions (running price, step summaries and clear next steps) and used the pause evidence to win the argument.
- Instead of
- Shipping the requested filter and button polish: a faster political win with a likely small lift.
- Why
- The session recordings showed confidence breaking between screens. Polishing the screens themselves wouldn't reach it.
- Trade-off
- More engineering cost, and a harder story to tell in status meetings. Filters and buttons still improved, but as support for the strategy rather than the strategy itself.
Decision 2
Keep price and package on screen at every step
Every step names the decision, what it costs and what it doesn’t commit you to. Pricing and the chosen package stay visible, so buyers stop going back to check.
- Why
- Going back to check whether the price had updated was one of the most common pauses in the sessions.
Decision 3
Cut the roadmap to what protected trust
I scoped a larger roadmap than we shipped, then cut anything that didn’t protect trust between screens. What stayed: one labeled filter pattern so learning transfers across Find Your Home, a clear hierarchy for map sorting, and AI chat inside the flow so questions don’t pull people out of context.
- Trade-off
- Fewer headline features to point to. I chose unglamorous continuity work over flashier features I could have claimed.
04 · Outcome
Completion rose 31 points, and became more predictable
After launch I re-ran the same tasks with the same 37 participants. Find Your Home completion went from 60% to 91%, and Configure & Buy from 80% to 94%.
Process
Same study, before and after shipping
Re-ran the same tasks with the same 37 participants after the redesign.
Find Your Home completion
Configure & Buy completion
Time spreads tightened too: Find Your Home 28s → 18s; Configure & Buy 86s → 42s.
The number I’d defend first isn’t the 31-point lift. It’s the time spread: 28s down to 18s for Find Your Home, and 86s down to 42s for Configure & Buy. Hesitation didn’t just average out; it consolidated, so finishing became predictable under the same study conditions.
05 · Reflection
Conversion lives in the joinery
For a first-time buyer, finishing a homebuying flow is an act of trust, not a usability trophy. This version won because we measured the real failure mode and disagreed with a tidy brief when the evidence said otherwise.
I still apply that to high-stakes consumer flows. People rarely abandon because a UI is ugly. They abandon when, between two screens, the product stops giving them a reason to take the next step.