home

life lately

schedule a call

what did i forget?

routing adventures

get to know me

vibing to

currently working on

see my projects :)

let's connect!

send a message

resume

drafts

Reformation app on two phones over an editorial photo
Reformation  •  Mobile app concept

A catalogue you can actually check out of.

Role
Product Designer
Participants
12 shoppers
Methods
Usability tests, Maze, card sort
Scope
iOS app, end to end
The problem

The dreaded, but inevitable, browser crash.

Picture this: TikTok serves you the top you've seen on Hailey Bieber and Emrata. They have your size. You're mid-checkout, debating expedited shipping for your girls' date tomorrow, annnd the browser crashes. You refresh, but you can't remember what the top was called. The visual thread's snapped, soo you move on with your day.

184s
until the mobile browser typically crashed mid-session
3,171
items to browse, and counting
2m 14s
to find "a dress under $200," vs. 45s on Aritzia and Sézane
Research

Beautiful, but hard to shop.

Before commencing design, I gathered a cohort of 6 people to navigate Ref's mobile browser. They loved the imagery, but most weren't sure what was tappable.

Screen recording
4 of 6
paused on the hero for 10+ seconds, then went to the hamburger menu. Nobody used the photo as a way in.
“I wasn’t sure if I could tap it. It just felt like a photo.”
Think aloud
3 of 6
weren’t sure which editorial collection held the item type they wanted.
“Romance Yourself... does that mean dresses? I’m not sure.”
Maze
5 of 6
Safari sessions hid Ref’s mission tagline behind the browser bar, a limit of mobile web itself.
Being naked is the #1 most sustainable option. We’re #2.

The shopper's mood, stage by stage.

Tap a stage
What they do
Reviews product images, looks for similar items, checks size availability.
What they think
“This would go really well with the blazer I have.”

12 shoppers sorted 22 features. They think in app patterns.

That gave me a navigation I didn't have to invent: Shop, Bag, Loves, and Account, with product details kept on the product itself.

Feature
Browse Products
9
3
Filters
7
3
2
Size Guide
2
10
Fabric & Care Materials
2
9
1
Store Locator
2
8
2
Wishlist
8
4
Save for Later
10
2
Edit Quantity
12
Shipping Cost Estimate
12
Estimated Delivery Date
8
4
The design

Keep the editorial feel. Make every photo tappable.

Design 1/4

Onboarding, simplified.

Testing put the ideal onboarding at 3 to 5 screens. I kept it to 2, with Face ID so returning shoppers skip the password entirely.

2
screens from open to browsing
Design 2/4

Flipping through a catalogue.

Shoppers love Ref’s editorial feel, so the home page reads like a lookbook. The difference: every photo is now a doorway into what’s in it.

4/6
testers had tried tapping hero photos that went nowhere
Design 3/4

Checkout, made simple.

Price, tax, and shipping sit up front. Editing size, quantity, or moving an item to Loves all happen inside the bag, so nobody has to leave it.

0
times you need to leave the bag to edit it
Design 4/4

Loves, aka the wishlist.

Wishlisted items had a 67% chance of being bought, so Loves earned a spot in the nav bar. Named lists like “Capsule Closet” turn saving into planning.

67%
of wishlisted items were purchased

The wishlist went from a dumping ground to a decision tool.

Only 20% of shoppers revisited saved items before. Named, organized lists changed that.
+78%
wishlist utilization
+41%
revisit rate on saved items
67%
of wishlisted items were purchased
Key takeaways

Every tap is a decision point, and a drop-off risk.

Two questions shaped every decision: why does this product deserve your attention, and what are you actually here to do?

When you're dropping $200+ on a single piece, you want to feel good about it (no buyer's remorse, please). My cohort wanted fabric sourcing, care instructions, and sizing upfront, not buried behind a support line they'd have to dig for. And lo and behold, Ref's product pages now have all three.

Since then

Ref's been closing the same gaps.

Since this study, Ref has rolled out fit details, material and care info, and even a "Ref under $200" edit (hi, Maze task). All gaps my cohort surfaced. The one piece still unsolved is where this whole thing started: shopping Ref on mobile without the browser giving out on you.

What I'd continue

The final "Complete Purchase" tap.

Shoppers could edit their cart without leaving it, but I never fully tested what causes abandonment at that last step. Hesitation, distraction, price shock, or all of the above? That's where I'd foray into next.