Out-of-Stock Experience, Turning a Dead-End into an Opportunity
User Problem
At Farfetch, ~15% of Product Detail Page (PDP) visits were for out-of-stock items. Users often abandoned the page without exploring alternatives, resulting in missed conversions and disengagement.
Findings
High drop-off: A significant portion of users landing on out-of-stock PDPs did not continue browsing.
Customer insight: Peer-to-peer research and role-playing, simulating a luxury store experience, highlighted the opportunity to provide intelligent, personalized next-best-item recommendations—mirroring what a sales associate would do.
Opportunities
Next-Best Recommendations: Introduced multiple recommendation carousels:
Same item, different colors
Same brand + category (e.g., Burberry mules)
Same category + color
A/B/C Testing: Tested which recommendation approach best drove engagement, focusing on relevance and user intent.
Impact
Increased browsing: Users encountering out-of-stock PDPs were more likely to explore suggested products.
Add-to-Bag & GMV growth: Recommendations led to higher add-to-bag actions and increased revenue.
Variant performance: Variant prioritizing similar products over generic brand/category suggestions had the strongest engagement.
Ongoing improvements: Integrated wishlist “Back in Stock” notifications, connecting users to items they cared about and reinforcing engagement (see Relevant Notifications use case).
Mapping the User Problem with Data Metrics to Kick-Start the Project
Assumptions and Solutions
From left to right: Mobile First
Design Close-up, Specs for Dev and Translations in Cyrillic
