What's newin Layers

AI agents can now ask for your approval before a merchandising rule change goes live — review the request from your dashboard inbox and approve or reject it before anything publishes.

The sort order preview now breaks a Weighted Group blend down feature by feature — raw value, normalized score, weight share, and contribution — so you can see exactly why a product ranked where it did.

Request the new `collection_titles` attribute on Search, Browse, and Block responses to get every collection a product belongs to, with no follow-up lookup required.

Similar Products recommendations now build from the exact variant a shopper is viewing, not just the overall product, for more relevant matches on color- and size-specific pages.

Reviews.io review ratings and counts now sync as product and variant attributes you can sort, filter, and merchandise by.

Judge.me, Fera, Junip, Smile.io, LoyaltyLion, and Swym Wishlist Plus are now available as Layers catalog integrations.

Watch AI agents apply facet and merchandising changes step by step, each move captioned with the reason behind it.

Variant breakouts can now exclude variants that match a condition you define — like being on sale or out of stock — so they never get their own tile.

Layers now supports Shopify collections that include only some of a product’s variants, so browse, filters, price sorting, and variant breakouts respect exactly what’s in the collection.

Every store now gets an automatic weekly email recapping search and merchandising performance — no dashboard login required.

Zero-result search reporting no longer counts SKU and barcode lookups that miss your catalog, and LayersQL now tracks those misses under their own metrics.

Turn single-type metaobject-reference metafields — like color or material swatches — into storefront facets, with an opt-in flag that resolves each value’s display name and image automatically.