All promptsTechnical SEO

Audit ProductGroup schema one product family at a time

Find broken parent-child relationships, variant identity errors and offer mismatches without burying the repair team in syntax noise.

Works in
ChatGPT, Claude, Gemini
You need
Rendered JSON-LD · Variant catalogue export · Live variant URLs
Written for
product variant schema audit prompt

Variant audits go wrong when the reviewer validates a template but never checks the selected product. This prompt follows each sellable member back to the catalogue and live page, where the expensive errors usually appear.

The prompt 110 words
Act as a product-data reviewer. Audit one product family using its rendered
JSON-LD, catalogue rows and live URL evidence.

First map the ProductGroup, productGroupID, variesBy properties and every
Product or hasVariant member. Then compare each member's URL, canonical,
SKU, GTIN or MPN, colour, size, image, price, currency and availability
against the catalogue and visible selected state.

Return: (1) a family summary, (2) a row-by-row evidence table, and (3) a
repair queue ranked BLOCKER, HIGH, MEDIUM or CLEAN. Separate invalid schema
from correct syntax containing wrong business data. Quote the supplied
evidence for every finding. Never infer a variant or identifier that is absent.

JSON-LD: [JSON_LD]
Catalogue: [CATALOGUE]
Live-page evidence: [PAGES]

What to change

Everything in square brackets is yours to replace. Nothing else needs editing.

[JSON_LD]
Rendered structured data from the family and variant URLs.
[CATALOGUE]
The source rows for all sellable variants in this family.
[PAGES]
Visible selection, canonical and offer facts from each live URL.

How to run it

  1. 01
    Choose one complete family

    Include available, unavailable and sale variants where possible.

  2. 02
    Paste rendered evidence

    Use the live output rather than a template or CMS preview.

  3. 03
    Repair identity before syntax

    Fix catalogue and URL mismatches before polishing optional properties.

Questions people ask

Can this prompt validate schema eligibility?

It can identify evidence-backed problems, but final syntax and eligibility checks should still use Google Rich Results Test and Search Console.

A new prompt, most days One working prompt for a real SEO or AI visibility job, what to change in it, and a worked example. No sequences, no offers dressed as newsletters.

Unsubscribe in one click. We never pass your address on.

Run your first audit
in about a minute

Free account, no card. Paste your URL and get a real, scored report of your AI and search visibility.

Measuring rankings in