Product feed

What it is

A product feed is a file or API stream that describes every item in your catalogue using named attributes agreed with the receiving platform. Common formats include XML, CSV, TSV and Google Sheets, with required fields such as id, title, description, link, image_link, price, availability and brand, plus identifiers like GTIN or MPN. Most retailers maintain a primary feed generated from their ecommerce platform, with supplemental feeds for overrides and enrichment.

Why it matters

Feed quality decides whether products are eligible to appear at all, and how well they match a shopper's query. Titles, attributes and images in the feed are usually the strongest signals a channel has about an item, so weak data suppresses impressions and wastes budget. As assistants and AI shopping experiences summarise products, clean structured attributes give them something reliable to quote on price, availability and specification.

How it works

Practitioners map catalogue fields to each platform's specification, then validate the output in tools such as Google Merchant Center, fixing disapprovals and warnings before scaling spend. Feed management is ongoing: scheduled fetches or API pushes keep price and stock in sync, feed rules and supplemental feeds tidy titles, add custom labels and fill gaps, and enrichment adds attributes like size, colour, material and GTIN. Many teams pair the feed with matching Product structured data on the site so on page and off page data agree.

When it applies

It applies to any ecommerce business listing products on shopping channels, paid or free, and to marketplaces and affiliate networks that ingest catalogues. It also matters when a retailer wants accurate product facts available to AI assistants rather than scraped guesses.

Examples

• A fashion retailer rewrites feed titles to the pattern brand, product type, colour, size so Shopping queries match more closely.
• A homeware brand adds a supplemental feed with custom_label_0 set to margin band, then uses it to split bidding in Google Ads.
• A stockist fixes 300 disapproved items in Merchant Center caused by missing GTINs and mismatched prices between the feed and the landing page.

How it is measured

• Feed approval rate and count of disapprovals or warnings by error type
• Attribute completeness, for example percentage of items with GTIN, colour, size and material
• Impressions, clicks and conversions per item, plus share of catalogue with zero impressions
• Price and stock accuracy, measured as mismatches between feed values and live product pages