Catalog & Content Quality
Catalog Quality Auditor
Seven deterministic checks that score catalog data quality: duplicates, gaps, mismatches and suspicious claims.
The Problem
Bad catalog data quietly hurts search, SEO and conversion — and it is invisible until someone audits it.
How It Works
- Run seven checks per product: duplicates, missing material or description, poor titles, category mismatch, suspicious claims, inconsistent colour.
- Penalise 1.2 per error and 0.6 per warning.
- Report an overall score with the offending product IDs.
Technically Interesting
Fully reproducible scoring — the same catalog always gets the same score and the same list of fixes.
Why It Matters
Clean data is the precondition for any intelligent commerce feature, AI or otherwise.
Scope & honesty. Despite the name, there is no LLM: the checks are deterministic. Demo catalog is deliberately issue-dense (score 33/100).