Commerce — Selected Studies
A collection of e-commerce flows: PDP, checkout, search, and post-purchase loops.
Before a single pixel.
Discovery, framing, structure and validation — the work that decided what the screens had to be.
- 01Discover
Read the drop-off
Session replays and funnel data across search, PDP and checkout to find where intent dies.
Session replay reviewFunnel analysisSupport ticket mining - 02Define
Surprises kill carts
Isolated hidden fees and latency as the two highest-leverage problems worth designing against.
Hypothesis backlogImpact/effort matrixSuccess metrics - 03Structure
Shortcuts for returning buyers
Restructured entry points so repeat purchase paths sit beside search rather than below carousels.
IA revisionsTask flowsComponent inventory - 04Ideate
Sketch the fast path
Wireframed a total-first checkout and an instant-feedback search before visual design.
WireframesInteraction prototypesCopy drafts - 05Validate
Ship, measure, repeat
Every change ran as an experiment; only lifts survived into the permanent design.
A/B experimentsUsability testingPost-launch analytics
The brief.
A series of focused commerce studies — PDP density, single-page checkout, search-as-you-think, and the loops that bring buyers back. Each one a small system in itself.
Commerce details look small until they cost real revenue. A vague badge, a sluggish search, a checkout that asks twice. These studies pick at the small things until they stop costing.
Talking to real users.
- ✦Funnel analytics
- ✦Session replays
- ✦Moderated usability
- ✦A/B testing at scale
60+ shoppers across 3 storefronts · rolling A/B cohorts
Ongoing (24 months of studies)
"If I don't see the total in one glance, I bounce."
"Don't make me log in again. I've bought here twelve times."
How the study actually ran — four frames from field to evidence.
Read the drop-off
Session replays across search, PDP and checkout to find where intent dies.
Name the leverage
Hidden fees and latency isolated as the two problems worth designing against.
Total-first checkout
Wireframed the fast path before a single visual decision was made.
Only lifts survive
Every change shipped as an experiment; losers were rolled back the same week.
- Insight 01
Speed is a feature
200ms search latency shifted add-to-cart more than any redesign.
- Insight 02
Checkout hates surprises
Every surprise fee at step 3 collapsed conversion. Show the total early.
- Insight 03
Return buyers want shortcuts
'Buy again' next to search bar out-converted every homepage carousel.
How it came together.
- 01
Treated each surface as its own product with its own metric.
- 02
Prototyped friction reductions in code, not just Figma — feel beats spec.
- 03
A/B'd everything against control; trusted the numbers when intuition lost.
- 04
Wrote up the losers too — failure was as instructive as the wins.