PolandRetail and e-commerce2026 – ongoingAllegro
AlleCompanion complementary-product recommendation retrieval
Allegro reported 21.25% higher recommendation-linked web GMV in AlleCompanion in-cart tests
Problem
Products bought together did not always work together. Allegro needed recommendations that reflected compatibility.
Approach
Allegro deployed AlleCompanion, combining a neural retrieval model with category constraints. It evaluated the model alongside existing production systems in two-week online experiments.
Results
Takeaway
Results varied by placement. The reported in-cart gains measured GMV linked to carousel interactions, rather than total marketplace sales.
Categories
- Industry
- Retail and e-commerce
- Business function
- Marketing and sales
- Innovation type
- Process innovation
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
Allegro's researchers described production deployment and online comparisons with existing recommendation systems. They reported significant in-cart GMV gains on web and app, with mixed results elsewhere.
Full details are in the original sources.
