StyleGraph
Recommendations that actually go together
A style-aware recommendation engine that learns outfit combinations from the lookbook and the shopper's fit profile — suggestions that make sense, not just "also viewed".
- Outfit -level output, not item-level
- Live Re-rank on every action
- Fit -aware by default
StyleGraph builds a real graph of how items pair: what the lookbook styles together, what shoppers buy together, and what the fit profile says will work. The output is recommendations that feel curated, not a random column.
Every time a shopper adds to cart, the graph re-ranks the rest of the visit; every time an outfit is bought as a set, the underlying relationships update.
Everything StyleGraph handles for you
-
Lookbook-informed
Captures the stylist's intent from the lookbook — the graph learns what should pair.
-
Fit-aware
Filters recommendations by what will fit the shopper, using their FitProfile.
-
Live re-rank
Re-ranks on every cart action — the last add shapes the next suggestion.
-
Outfit output
Recommends outfits, not just single items — a lookbook-style set the shopper can add with one tap.
How StyleGraph plugs into the fashion stack
StyleGraph is a recommendation engine on top of catalog, lookbook, and fit data. It does not change inventory — it steers attention.
- BoutiqueSite renders the "complete the look" module on every product page.
- LookbookCDN supplies the curated outfit examples the graph learns from.
- FitProfile filters recommendations by the shopper's fit rules.
- OutfitStudio consumes graph output when stylists build fresh outfits for a drop.
Wire StyleGraph into your product today
Book a consultation with our founders and we'll walk you through the whole microservice stack — not just this one — live on your domain.