Discussion starter: an entity SEO question, posted with an answer from Niraj Raut to open the thread. If you have dealt with this on a site, reply with what you saw, especially where it differs.
- Entity SEO
- Structured Data
How do you QA structured data across 200 templates after every release, and which tools catch real problems rather than false positives?
Disclosure: Niraj Raut, who posted this answer, runs the SEO consultancy linked at the end of it.
1 reply
-
Niraj Raut
At 200 templates, manual structured data checks after each release do not scale, and generic validators produce too many false positives to be trusted. What catches real problems is an automated check that runs against a representative URL for each template, extracts the structured data from the rendered page, and tests three things: that required properties for the search features you rely on are present and valid, that values match what the page shows, and that nothing has changed unexpectedly compared with the previous release. Search Console’s enhancement and merchant listing reports then act as a lagging confirmation. The false positives mostly come from tools that flag every missing recommended property, or that treat schema types without Google features as errors.
What to check, and why
Check Catches False positive risk Markup present on each template that should have it Markup removed by a release Low JSON-LD parses Broken syntax from template changes Low Required properties for used features, such as price and currency on product Offers Loss of rich result or merchant listing eligibility Low Values match the visible page Stale or wrong prices, availability, ratings Low to moderate; needs careful extraction Diff against previous release Unexpected changes in types, properties or values Moderate; expected changes must be whitelisted Every recommended property present Rarely anything important High Building the check
- Pick representative URLs for each template, including edge cases such as out-of-stock products, products with variants, sale prices and articles with several authors.
- Render the pages as they are served in production or on a production-equivalent staging environment, including JavaScript-injected markup.
- Extract structured data and the visible values it should match, using a crawler with custom extraction or a script.
- Validate against Google’s documented requirements for the features you use, not against every schema.org property.
- Diff against the last release and alert on removed types, removed required properties or changed values.
- Run it in the release pipeline for critical templates, and on a schedule in production for the rest.
Reducing false positives
- Limit checks to the structured data types that produce features you care about.
- Treat missing recommended properties as information, not failures.
- Whitelist expected changes that accompany planned releases.
- Review and prune checks that fire without leading to action.
Hypothetical example:
A retailer with 200 templates relies on a generic audit tool that reports thousands of structured data “issues” after every release, mostly missing recommended properties. The team ignores the reports. After a release removes the Offer from product markup on one template, merchant listings for that category disappear for two weeks before anyone notices. The team builds a targeted check: representative URLs per template, validation of required product and organisation properties, comparison of markup prices with visible prices, and a diff against the previous release. The next time a release breaks the markup, the pipeline blocks it before it reaches production.
Watch for this:
Validating only the HTML source misses markup injected by JavaScript, and validating only staging misses production differences such as CDN or tag manager changes. Check the rendered output in production as well as in the pipeline.
Group templates to keep the suite manageable
Two hundred templates rarely mean two hundred distinct structured data implementations. Most share a smaller number of markup generators, such as product, article, organisation and breadcrumb components. Map templates to the components that generate their markup, and test each component thoroughly on a few representative templates, plus a lighter smoke test across all templates. That keeps the suite fast and maintainable, and when a component changes, you know exactly which templates to retest.
Use Search Console as confirmation
Search Console’s merchant listing, product snippet and other enhancement reports show what Google found, with a delay. Watch them for sudden changes in valid or invalid item counts by template, and investigate any jump. They are a safety net for problems your checks missed, not a replacement for checks, because by the time they show a problem, it has usually been live for days.
Automate checks per template on representative URLs, test required properties, value consistency and changes from the last release, run them in the pipeline and in production, ignore recommended-property noise, and use Search Console reports as a lagging safety net.
Need help with this on your own site?
Niraj Raut works with businesses in Nepal, Australia, the UK, Europe and the US on technical SEO, ecommerce SEO, local SEO and AI search.
Add a reply Cancel reply
Replies are open to approved members. Log in or apply to join.