Color matching matrix, 10+ shades × highlights × balayage × ombre. How does Magento handle it?

This is the heart of hair-ecom and the #1 reason brands outgrow Shopify.

Magento configurable + simple products with EAV attributes scale cleanly to the cross-product: shade family (10+, jet black, off-black, dark brown, medium brown, light brown, dark blonde, medium blonde, light blonde, platinum, red, auburn, gray, custom), highlight pattern (solid, balayage, ombre, root smudge, money-piece, babylights, dimensional), texture (straight, body wave, deep wave, curly, kinky), origin (Brazilian, Indian, Russian, Eurasian), length (14”, 24” in 2” increments), method (clip-in, tape-in, hand-tied weft, fusion, micro-bead). Cross-product runs 600-1,200 variants per product family. No app-stack ceiling.

The performance gotchas:

• Custom denormalised stock-status table for “in-stock by shade × length” filtering, native indexer slows at 600+ variants.
• Hyvä theme over Luma. Color swatch picker stays under 80ms INP on Hyvä; Luma jQuery re-renders the picker on every interaction.
• Visual color swatches (not text labels) with real hair-photo crops, lazy-loaded. Customer scans the grid visually first, narrows by length second.

Color-matching quiz: customer uploads 6 photos (root, mid, ends, in sunlight, indoor light, fresh-washed), answers 4 questions (natural undertone, current dye state, allergies, desired result). Scoring algorithm (custom rule-based or ML hosted on Vercel / AWS Lambda) returns the top-3 matched SKUs, auto-adds top match to cart with the others as alternatives. Cuts shade-mismatch returns from ~25% to ~10% in my data. The quiz logic lives outside Magento; Magento stores the result against the customer for future orders.