ecommerce_ops · ecommerce · workflow
Wayfair uses Google Gemini LLM to generate free-form personalized customer interests powering homepage product carousels
Traditional customer understanding models relied on fixed taxonomies and required extensive training data, missing implicit or latent interests not directly expressed in customer behavior.
How it works
Common implementation structure
How this type of workflow is generally built, generalized across documented cases — not tied to any one vendor's stack. Click any stage to read what happens there. Specific products that implement these stages appear in “Tools commonly seen” below.
Stage 1 · Customer behavioral signals ingested
Customer search queries, viewed product details, Add to Cart, wishlist and purchase signals from historical data are fed into Gemini using curated prompts.
Tools used
GeminiDynamic Page ConstructorUI Composersemantic search model
Outcome
Wayfair's new interest-based product carousels on the homepage are already driving measurable gains in engagement and revenue, while the in-house data compression model yields significantly lower operational costs.
Results
Volumeup to 70%
Cost replacedsignificantly lower costs
Grounding & classification
Source type: technical build writeup
27 fields verified against source quotes.
content generationpersonalizationpredictive analyticsrecommendation systemknowledge baseproduct catalogbuilder submittedmetric backednamed customerproduction runtime claimedtools describedworkflow describedecommerceconversion increasecost reductionrevenue increasetechnical build writeupecommerce opsmarketing opsdata sync enrichmentextract classify route