How to Run a Technical SEO Audit

A technical SEO audit checks whether search engines can find, crawl, render, index and understand your pages. It does not assess whether your content is any good — it establishes whether your content ever gets a fair hearing.

The output of a good audit is not a list of every imperfection. It is a short, ordered list of things that are actually costing you traffic. Most audit tools produce hundreds of findings; most of those findings do not matter.

Work in dependency order

The checks form a chain. There is no point optimising title tags on a page that is blocked in robots.txt, so work through the stages in order and fix blockers before refinements.

1. 1Crawlability — can search engines reach the page at all?
2. 2Indexability — are they allowed to index it, and are they choosing to?
3. 3Architecture — can they understand how pages relate?
4. 4On-page — is each page's subject clear?
5. 5Rendering — does the content exist without JavaScript execution?
6. 6Performance — does it load acceptably for real users?
7. 7Structured data — is anything eligible for rich results?

1. Crawlability

• ●robots.txt — confirm nothing important is disallowed. A stray Disallow: / shipped from staging is the single most damaging SEO error there is.
• ●XML sitemap — present, referenced in robots.txt, containing only canonical HTTP 200 URLs.
• ●Server errors — 5xx responses in your logs or crawl. Persistent 5xx tells Google to slow down crawling entirely.
• ●Redirect chains — A → B → C wastes crawl budget and leaks signal. Point A directly at C.
• ●Orphan pages — pages with no internal links pointing at them. If nothing links to it, discovery relies entirely on the sitemap.
• ●Crawl depth — anything more than three or four clicks from the homepage gets crawled less often.

2. Indexability

Google's Page Indexing report is the authoritative source here — it tells you what Google actually decided, not what your crawler predicts.

• ●Unintended noindex tags — check templates, not just individual pages. One bad template can deindex thousands of URLs.
• ●Canonical tags — self-referencing on canonical pages, and pointing to the right target on variants. Common failure: every paginated page canonicalising to page one, which deindexes the rest.
• ●"Crawled — currently not indexed" volume. A large number here is a content quality signal, not a technical bug.
• ●Duplicate content from URL parameters, session IDs, trailing-slash variants or http/https duplicates.
• ●Index bloat — thousands of thin tag, filter or search-results pages indexed. These dilute site quality signals.

3. Site architecture

Architecture is how authority moves through your site. A page with no internal links is invisible regardless of how good it is.

• ●Internal linking — do important pages receive links from other relevant pages? Do they use descriptive anchor text?
• ●URL structure — readable, lowercase, hyphenated, stable. Do not restructure URLs without a strong reason and complete redirects.
• ●Breadcrumbs — help users and search engines understand hierarchy, and improve how your result displays.
• ●Pagination — each page indexable with self-referencing canonicals, not all canonicalised to page one.
• ●Hreflang for multilingual sites — must be reciprocal, or Google ignores it.

4. On-page elements

Missing meta descriptions are worth fixing on pages that get impressions, and safe to ignore elsewhere. Google rewrites descriptions much of the time anyway — so prioritise the pages where the rewrite is costing you clicks.

5. Rendering

If your site is a JavaScript application, check what exists in the initial HTML versus what appears after scripts run. Google does render JavaScript, but rendering is queued, resource-limited and occasionally fails.

• ●Use Search Console's URL Inspection → Test Live URL → View Rendered HTML to see what Google actually got.
• ●Main content, internal links and canonical tags should ideally be in the server-rendered HTML.
• ●Links must be real <a href> elements. A div with a click handler is not a link and will not be followed.
• ●Content behind interaction — infinite scroll, "load more" buttons — is generally not seen. Paginate with real URLs.

6. Performance and 7. Structured data

Check Core Web Vitals at the template level rather than page by page — the fix is nearly always in a shared template. Field data from Search Console is what Google scores; lab data from PageSpeed Insights is what you iterate against.

For structured data, confirm your key templates carry valid markup for their type, and review Search Console's Enhancements reports for errors that appeared after a release.

Prioritising what you find

Sort findings by traffic at risk, not by the tool's severity colour. A "critical" warning on a page nobody visits is not critical.

Auditing with Flux N Pro

The Site Audit crawls up to 500 pages and runs more than 50 checks across the categories above, then generates specific fix recommendations for each issue found rather than only naming the problem. A single-page version is available free without an account, and full-site crawls with issue tracking over time are part of the paid plans.