The Technical SEO Audit: Fixing the Foundation

A website can pass every technical SEO check going and still fail to get indexed. It is worth explaining why, because the gap between "the audit is clean" and "the site is working" is where a lot of budget gets wasted.

Automated audits are useful. They are also easy to over-trust, and the things they miss tend to be the things that actually matter.

What a technical audit actually checks

A technical SEO audit examines whether search engines can discover, crawl, render and index your content. The standard scope covers four areas.

• Crawlability. Robots.txt rules, XML sitemap accuracy, internal linking depth, and redirect chains that dilute link equity.
• Indexability. Canonical tags, meta robots directives, and what Search Console reports as indexed against excluded.
• Renderability. Whether JavaScript content is visible to crawlers, Core Web Vitals thresholds, and mobile usability. This matters even more for AI search, as our guide to getting cited in AI answers explains.
• Structured data. Schema markup validity and eligibility for rich results.

That list is genuinely useful and we are not about to argue otherwise. The problem is what happens when a site passes all of it and still underperforms.

Why a clean report is not a clean bill of health

Automated tools check what is cheap to check: character counts, tag presence, status codes on URLs you supplied. Those are real checks and they catch real problems. But three structural blind spots come as standard, and no report will mention them, because a report can only describe what it looked at.

Coverage is a choice, and gaps are invisible. Every audit runs against a list. If that list is shorter than your actual site, whatever sits outside it has never been examined and nothing will tell you so. The first question to ask of any audit is not what it found, but how many URLs it looked at and where that list came from.

Rendered and served are different documents. Most tools execute JavaScript before checking. Crawlers on a low-authority domain may not get that far. If your tool renders and the crawler does not, your tool is grading a page that Google never sees. A canonical tag injected client-side will pass a rendering audit and be absent from the served response.

Presence is not correctness. A canonical tag exists. Fine. Does it point at the right URL? Is it the same in the served HTML as in the rendered DOM? Does Google actually agree with it? Google overriding your canonical is common, and no tag-presence check will ever surface it.

The problems that hide behind a passing score

These are the failures that hide behind a passing score.

Navigation that does not survive rendering. A footer or menu built by JavaScript can be missing entirely from the crawler's snapshot. The audit confirms the page has links. It does not confirm those links exist in the version a crawler sees.

Whole sections reachable only via the sitemap. If a blog or resource area has no rendered internal link pointing at it, those pages are technically discoverable and practically orphaned. From a sitemap-driven audit, everything looks perfectly reachable.

Unknown URLs returning HTTP 200. Request a page that has never existed. If it returns a success status with homepage metadata rather than a 404, you have a soft 404 factory generating unlimited crawlable junk. Audits only check URLs they were given, so they never ask for one that should not exist.

Pages Google has never heard of. This is the one that costs real money. There are three index states, not two: indexed, crawled and rejected, and completely unknown. That third state appears in no coverage export, because errors require a crawl to have happened first. A commercially critical page can be missing from every report simply because Google has never fetched it.

What we would check instead

If you are auditing a site, or paying someone to, these are the checks that earn their place. Most take minutes.

• Fetch a page without JavaScript and read what comes back. Not the rendered version. The raw response.
• Request a URL that does not exist and confirm it returns a genuine 404.
• Compare served HTML against Search Console's rendered snapshot. Use URL Inspection, choose Test Live URL, then view the tested page. Anything in one and missing from the other is your problem list.
• Confirm every important page has an internal link that survives rendering. A sitemap entry is not a link.
• Inspect your key commercial URLs individually. Not in bulk. One at a time, because that is the only way the "unknown to Google" state becomes visible.

How to prioritise what you find

Not every issue deserves the same urgency. We rank on three dimensions.

• Impact. How many pages, and how much traffic, does this affect?
• Effort. What does the fix actually cost in development time?
• Urgency. Is this costing you rankings now, or is it a missed opportunity for later?

High impact and low effort goes first. That usually means fixing render-blocked navigation and orphaned sections before touching a single meta description, even though meta descriptions are what most reports flag loudest.

A list of character-count warnings is worth very little next to a set of pages no crawler can reach. Work a typical report strictly top to bottom and you can spend an evening tidying descriptions and change nothing that matters.

Our honest position on audit scores

Technical SEO audits are necessary. They are not sufficient, and a clean score is not evidence that a site is healthy.

Be wary of anyone who hands you a percentage and calls it a diagnosis. A good audit tells you what it checked, what it could not check, and what it found when it stopped ticking boxes and started reading raw responses. If the deliverable is a colour-coded score with no reproducible method behind it, you have bought a report rather than an answer.

Ask whoever audits your site one question: how would I verify this myself? If there is no answer, the audit is decoration.

A technical website audit is one of the things we do, scoped around what a particular site actually needs rather than run off a fixed checklist. If your traffic has flattened and nobody can explain why, that is usually a foundation problem rather than a content one. Our SEO services start by finding out which. If you are earlier than that and want the fundamentals first, start with our beginner's guide to SEO.