All projects
Fitly

Web App · 2026

Fitly

Precision fit recommendation app. Tell it your body, your style, your fit preferences and get matched pieces from brands you actually want to wear.

Role

Product Designer

Platform

Mobile Web (PWA)

Stack

Next.js · TypeScript · Tailwind

Status

Prototype

Problem

Fit. The #1 reason clothes get returned.

Not style. Not color. Just fit. Online shoppers can't try clothes on, and sizing is inconsistent across every brand. The result is returns, waste, and frustration.

Research

The numbers behind the problem.

Online clothing returns due to wrong fit

37%

Coresight Research, 2023

Shoppers frustrated by inconsistent sizing across brands

71%

Narvar Consumer Report, 2023

Annual cost of fit-related returns in the US

$7B

National Retail Federation, 2023

Case Study

From problem to product.

Process

Four steps to the right fit.

01Return data analysis

Mapped the top reasons for online clothing returns. Fit dominates at 37%. Style and color together account for less than 15%.

02Sizing inconsistency audit

Compared size charts from 12 brands. A "Medium" varies by up to 4 inches across brands. No universal standard exists.

03Onboarding model testing

Tested 3 input flows: quiz, form, and conversational. The 4-step structured form completed in under 90 seconds with the highest completion rate.

04Recommendation engine

Built a scoring system that weights body measurements, style preferences, fit type, and occasion to match against a product database.

Screens

Body. Style. Fit. Results.

01 · Body

01 · Body

02 · Style

02 · Style

03 · Fit

03 · Fit

04 · Review

04 · Review

Results

Results

Buy or find in store

From recommendation to purchase.

Each recommendation includes a direct link to buy online and a store locator showing nearby locations with stock status.

Product detail

Product detail

Store & buy links

Shop & store

Solution

Describe yourself. Get matched.

01

4-step onboarding

Body, style, fit, review. Each step is one focused screen. No account required, data stays on device.

02

Scored recommendations

Three product matches ranked by fit confidence. Each shows brand, price, match reason, and a fit description.

03

Buy or find in store

Each result includes a buy link to the brand site and a store locator with stock status and directions.

Design decisions

Color palette

Paper

Sand

Warm

Accent

Ink

Stone

Mid

Typography

Heading

Aa

DM Serif Display

Serif · Display · Editorial

Body

Aa

Inter

Sans-serif · UI · Labels

01

Warm linen palette, zero border-radius. Editorial aesthetic signals quality. Not a fast-fashion tool.

02

No account creation. Profile saves to localStorage. Results in under 90 seconds without signing up.

03

Three results maximum. More options increase decision fatigue. Each match is justified with a reason.

Outcome

Design to working prototype.

Working prototype7 screens designedRecommendation engineFull design systemStore locator concept
View the prototype
Body
Results
Store