Category Catalog Gap Audit: Find the Data Gaps That Matter

A missing-field count does not tell you whether a customer can buy the right part. A missing color on one category is harmless; a missing thread size, voltage or compatibility note on another can derail the order. Audit by category and by buyer decision, not by one completeness percentage for the whole catalog.

You need a field list for each category you audit. If you don’t have one, start from the matching template in the Attribute Library and adapt it with the category attribute template guide.

1. 1
   Select a sample and record why

   Include high-traffic items, long-tail items, new supplier ranges, frequently returned products, ambiguous searches, and records with linked technical documents. Write down the sampling date and why each SKU is in scope. If you will report a score, keep the denominator and the excluded records alongside it.

2. 2
   Resolve identity before scoring anything

   Check manufacturer and part-number pairs, GTINs where they apply, variant boundaries and pack levels. Do not score attributes on a record you cannot reliably identify; fix or exclude it and say so. The GTIN validator catches check-digit errors quickly.

3. 3
   Compare three views of each field

   For every field in the category template, compare what the public product page shows, what ERP or PIM holds, and what the source documents say.

4. 4
   Classify every gap

   Put each gap in one of five classes: missing in the source, present but trapped in a PDF, extracted but not approved, approved but not published, or contradictory across systems. Each class has a different owner and fix.

5. 5
   Prioritize by decision risk and recoverability

   A required safety rating with contradictory evidence goes to a reviewer first. A facet that many products lack but that sits in their datasheets is a batch extraction job. A field no supplier publishes needs a new collection step with suppliers. Route each finding to the owner who can fix it, with source links and a measurable acceptance check.

6. 6
   Re-run the same sample

   After the next supplier update, audit the same sample again. Report how many gaps closed, how many new conflicts appeared, and which destinations still show stale values.

The worksheet

Keep one row per product, field and destination:

Column	Example
Category	Circuit breakers
Product identity	Manufacturer + MPN + pack level
Field	Interrupting rating
Source	Datasheet revision, page
Destination	Webshop, marketplace, ERP
Observed status	Trapped in PDF
Risk	High: safety-relevant
Owner	Category manager
Next action	Extract and review
Verification date	Date the fix was checked

A good audit isn’t a score for its own sake. It is a queue that makes catalog work smaller and safer, and a baseline you can measure against next quarter.

Why not measure catalog quality with one completeness percentage? +

Because fields do not matter equally. A missing color on one category is harmless, while a missing thread size, voltage or compatibility note on another can lead to the wrong part being ordered. Audit by category and by the buyer decision each field supports.

What types of gap should the audit distinguish? +

Five: missing in the source, present but trapped in a PDF, extracted but not approved, approved but not published, and contradictory across systems. Each has a different owner and a different fix.

How big should the audit sample be? +

Big enough to cover the kinds of record that fail: high-traffic items, long-tail items, new supplier ranges, frequently returned products, ambiguous searches, and records with linked technical documents. Record the sampling date and why each SKU is in scope, and publish the denominator with any score.

How often should the audit be re-run? +

Re-run the same sample after each significant supplier update, and report how many gaps closed, how many new conflicts appeared, and which destinations still show stale values.