What does catalog completeness mean?
Catalog completeness means the proportion of a store's products whose records contain every field the page needs to do its job. A complete SKU has a title that says what the product is and the attributes buyers search by, a set of attributes populated for the store's filters, a description that is unique to that SKU rather than copied from a sibling, images with descriptive alt text, valid product structured data, and a Merchant Center record with no disapproval.
The definition is deliberately mechanical. Completeness is not a judgment about how good the copy is; it is a check on whether the data is there. That is what makes it measurable from the catalog alone, and what makes it the first thing a Product Catalog FTE reports.
Which six field groups are scored?
The Product Catalog FTE on ftes.ai scores completeness across six field groups, because each group is read by a different system and a gap in any of them costs the page something specific. The groups are listed below with the system that reads them.
- Title: read by search engines and marketplaces; must name the product and the attributes buyers filter on.
- Attributes: read by the store's filters and by Shopping; missing values make a product invisible to a filtered search.
- Description: read by search engines and buyers; must be unique to the SKU, not a duplicate of a sibling.
- Images and alt text: read by image search and accessibility tools; alt text must describe the product.
- Product schema: read by search engines for rich results; must validate and match the visible page.
- Feed record: read by Merchant Center; any disapproval means the product cannot appear in Shopping.
Why does completeness matter more than copy quality?
Completeness matters more than copy quality as a first measure because a missing field removes a product from a channel entirely, while weak copy only lowers its performance in that channel. A product with no GTIN in the feed is not in Shopping at all. A product with an empty material attribute does not appear when a shopper filters by material. A product with duplicated description text competes with its own siblings in search.
Copy quality still matters, and the Product Catalog FTE works on it, but the order is fixed: presence first, then quality. A store that polishes descriptions on incomplete records is polishing pages that some channels cannot see.
How is the score calculated?
The score is calculated per SKU as a pass or fail on each of the six groups, then rolled up as the share of SKUs passing all six, reported per collection and for the whole catalog. A SKU with five complete groups and one missing is incomplete; the report also shows which group failed, so a collection with a common gap such as missing alt text is visible as one problem rather than a hundred.
A labelled worked example: a collection of forty SKUs where twenty-eight pass all six groups scores seventy percent complete. If ten of the twelve failures share the same missing attribute, one attribute fix moves the collection to ninety-five percent. The Product Catalog FTE shows this arithmetic in the audit so the owner can see which single fix is worth the most.
What makes a field incomplete rather than merely thin?
A field is incomplete when it is absent, when it duplicates a sibling SKU's value where a unique value is expected, or when the feed rejects it. A field is thin when it is present but says little: a title that reads as the product name alone, or a description of one sentence. The Product Catalog FTE treats incomplete as a failure of the completeness score and thin as a candidate for a rewrite, ranked separately.
Keeping the two apart keeps the score honest. A store can reach a high completeness score with thin copy, and the report will say so by showing the rewrite backlog alongside it. Merging the two into one quality number would hide which problem the store actually has.
How does the Product Catalog FTE improve the score?
The Product Catalog FTE improves completeness in a fixed sequence: audit the catalog, rank the gaps by revenue at stake using the store's own sales data, fill each gap from the catalog, a supplier sheet or approved copy, and verify the result in the feed before proposing it. Each fix is a proposal in Approvals with a before-and-after view and the source of every value.
Gaps that cannot be filled from real data are not filled. A missing GTIN, a weight no sheet records, a material the supplier never stated: each is surfaced as a request to the owner with the SKU and field named. The completeness score therefore rises only as fast as real data allows, which is the correct speed.
What should a store do with the score?
A store should record its completeness score before any catalog work starts, set a target per collection in the Targets tab, and review the score monthly against the rewrite and disapproval counts. The score is a leading indicator: it moves before search impressions and Shopping visibility do, and it is the number a manager can hold the Product Catalog FTE to without waiting a quarter.
It is not the only number. Search movement per rewritten page from Search Console and disapprovals cleared are reported beside it. A completeness score that rises while disapprovals stay flat points to a data problem upstream, usually a supplier feed, and the report will show that too.