Answer
What does a Technical SEO audit actually check?
Whether pages can be crawled, rendered, indexed and understood.
What does a Technical SEO audit actually check?
Whether pages can be crawled, rendered, indexed and understood. That covers robots and real crawler access in logs, rendering parity between raw and rendered HTML, the indexed set against the intended set, canonical and redirect consistency, structured data validity, internal link distribution and orphans, and field performance per template.
The longer answer
The first group is access: what robots.txt allows, what the CDN and firewall actually do to each crawler, and what the server logs show was really fetched. Configuration and behaviour frequently disagree, and only the logs settle it.
The second group is comprehension: whether content, links and markup exist in the raw HTML or only after JavaScript runs, whether the indexed set matches the intended set, and whether canonical, noindex and robots directives agree with each other rather than contradicting.
The third is structure and experience: internal link distribution and orphans, click depth to commercial pages, structured data validity against the right types, and Core Web Vitals in field data per template rather than one lab score for the homepage.
What actually helps
- Verify crawler access in server logs, not in robots.txt alone
- Compare raw against rendered HTML for content, links and markup
- Reconcile the indexed set against the intended set
- Read performance from field data per template, not one lab test
Where this changes by service
The answer is not uniform across the work:
- AI Automation and Workflows, where it decides ending the monthly rebuild of the same report
- AI Platform Optimization, since building the comparison page the assistant currently has to invent is only as good as this is
- Agentic Commerce, in which it is the first thing checked before testing the buying journey as an agent would walk it, not as a person would
- App Store Optimization, which cannot be done properly without it
Where this changes by sector
Nor is it uniform across sectors:
- For a Finance and fintech business the stakes are higher, because regulated claims, compliance review in the publishing path, and trust signals that have to be real.
- In Home and trade services the consequence is sharper, since emergency intent, service-area geography, and a phone that has to be answered.
- Manufacturing and industrial: the sector’s constraint here is that low volume, high value, and a specification sheet that decides the deal.
The caveat
An audit that produces a list of four hundred issues with no prioritisation is a report rather than a plan. The value is in what to fix first and why.
If this answer does not fit your situation, the reason is usually one of the two lists above rather than the answer being wrong. This connects directly to 301 redirect or Canonical tag, and less obviously to Broad match or Exact match.
What to check on your own site
Four questions that establish whether this applies to you:
- Whether it is the same for a returning visitor as for a first-time one
- Whether it is actually true on the site rather than true in the documentation
- Whether anybody has measured it, or whether it is being taken on trust
- Whether the person who set it up is still the person who understands it
Related reading
Still not answered?
Send the version of this question that applies to your own site. If the answer is that you do not need us, that is the answer you get.
Find out what is actually holding this back
Send us your site and your goals. You will get an honest read on the gaps worth fixing first, whether or not you work with us.