Case Study: Wayfair achieves 99% category win rate and 7-point clickthrough lift with Snorkel AI

Wayfair accelerates product tagging with Snorkel Flow 10x faster labeling

Wayfair, the home goods and furniture e-commerce retailer, needed a better way to manage and tag its 40M+ product catalog. Inconsistent supplier data, noisy labels, and manual image labeling that could take eight weeks for just 6,000 images made it difficult to scale search and computer vision efforts. Wayfair partnered with Snorkel AI to improve data labeling and extract richer visual information from product images.

Snorkel AI implemented a data-centric AI workflow using programmatic labeling, weak supervision, and computer vision to build 46 tag models in days instead of months. The result was a 98.97% win rate over previous category-specific baselines, a 7-point lift in clickthroughs, a 5-point increase in add-to-cart rates, and model development cycles that were 10x faster, saving an estimated six years of employee effort.

Wayfair