Case Study — 2026
FASHIONHERO
Turning a competitor's 0% commission attack into a data-backed retention experiment.
FashionHero is a fast-growing fashion marketplace (+28% YoY revenue, 2.4M active buyers, 4,200 sellers) with one structural risk: 100% of revenue comes from transaction commission. When a competitor ("Forte") launched a 0% commission campaign aimed directly at FashionHero's most valuable sellers, I led the product strategy response — from quantitative segmentation and seller interviews to a validated retention experiment and a working prototype.
Roles
- Product Strategy
- Data Analysis
- Prototype Development
Timeline
Capstone Project AI Product Heroes 2 — June 2026
Live
Visit Website →Tech Stack
- Claude Code
- Neon (Postgres)
- Next.js
- Vercel
Simulated seller P&L data
FashionHero's revenue is 100% commission-based with zero diversification. A competitor launched a 0% commission offer aimed at "backbone" sellers (10+ reviews, 3+ months active) — a segment that generates 2.6x higher margin per PLN of GMV than top sellers, but doesn't realize its own value. Losing this segment means losing a buyer base that won't come back, within a 6-9 month window before the competitor can build its own.
I designed a Retention Campaign — 50 free Promoted Listings credits for at-risk sellers, paired with a real-time ROI dashboard showing sales impact within days. The approach was validated through an Opportunity Solution Tree, scored against 3 alternative opportunities using a PRES framework (Pain/Reach/Evidence/Strategic fit), and de-risked with a 6-week A/B test design (200 test vs. 200 control sellers).
Projected +10pp retention lift among at-risk sellers, laying the foundation for FashionHero's first non-commission revenue stream — projected at ~105K PLN/month (~1.26M PLN/year) if scaled into a paid Promoted Listings tier.
This project's differentiator isn't the interface — it's the reasoning that led to it. Three stages, each with its own method, turned a competitive threat into a validated experiment.
01 — Research
Quantitative segmentation of a 500-seller dataset into trophy, backbone, and new-seller cohorts, backed by 3 qualitative seller interviews and competitive research into Shopee SEA's 0%-commission precedent, Etsy Plus, and Allegro Ads.
02 — Prioritization
An Opportunity Solution Tree scored 4 candidate opportunities against a PRES framework (Pain, Reach, Evidence, Strategic fit), ranging 12–18 out of 20. The highest-scoring opportunity — a seller considering migration to a 0%-commission competitor — scored 18/20 and was chosen.
03 — Validation
A 6-week A/B test design (200 test vs. 200 control sellers): primary metric of +10pp retention lift, secondary metrics of ≥30% credit adoption and ≥3 dashboard sessions/week, with a stop-rule if adoption fell below 10% by week 4.
Real-Time ROI Dashboard
Sellers see their campaign effectiveness — "+227% this month" — plus a live sales chart, before spending a single credit.
One-Click Promote Flow
A 3-click campaign launch with 1/3/7-day credit tiers, tracking live impressions, clicks, and sales as they happen.
Smart Product Catalog
Tags surface "trending" and "promoted" products against a baseline sales snapshot, so sellers know exactly what to promote next.
Market & Trend Analytics
Live search-trend data and price benchmarking against the category median, turning buyer search behavior into a seller's next move.
