Craig Campbell SEO Training & Consultancy

Tutorials

Technical SEO audit: find and prioritise the fixes that matter

By Craig Campbell · 16 Nov 2026

4 min read

A technical SEO audit checks whether search engines can reach, understand and index the pages that matter, and whether visitors can use them reliably. The goal is a prioritised list of fixes with evidence, not a long export of every warning a crawler can find.

This guide gives you a practical audit order. It complements the older technical SEO guide, which remains unchanged. All businesses, URLs and figures below are fictional UK examples, illustrative only—not Craig’s client work or measured results.

1. Set the scope before crawling

Agree which website, subdomains and page groups are included. List the pages that create enquiries or sales, planned launches and any recent changes such as a redesign or migration.

  • Confirm access to Search Console for the correct property.
  • Agree whether staging, logged-in areas and international versions are included.
  • Ask the owner which pages matter most commercially.
  • Check crawl settings with whoever runs the server.
  • Record the audit date, tools and settings.

A crawler only sees what it is allowed and configured to reach. Note missing access rather than reporting an unseen area as healthy.

When agreeing which pages matter, use the keyword research guide to connect genuine audience searches with existing destinations. Research helps set priorities; it is separate from diagnosing crawl or indexing faults.

If the audit supports a platform change, follow the website migration SEO guide to record what must survive and repeat acceptance checks after launch. Keep migration sign-off separate from later search performance.

2. Compare Google’s evidence with your crawl

Start with Search Console’s indexing, Crawl Stats and URL Inspection evidence, then crawl the public site with a suitable tool such as the Screaming Frog workflow. Each source answers a different question.

  • Search Console: what Google reports for its crawling and indexing.
  • Site crawl: what your configured crawler found through links and sitemaps.
  • URL Inspection: detailed evidence for selected individual URLs.
  • Server logs: verified crawler requests, when available; see the log file guide.

Do not combine counts from different tools as if they measured the same thing. A URL can be crawled without being indexed, and an indexed URL may be missing from a crawl because it has no internal links.

3. Check crawling and access

  • Read robots.txt and confirm it does not block important pages or resources needed to render them.
  • Check important pages return the expected status and are accessible without signing in.
  • Review server errors and host availability problems in Crawl Stats.
  • Find links to missing pages and unnecessary redirect chains.
  • For large sites, investigate duplicate or low-value URL spaces with the crawl budget guide.

Google warns that robots.txt is not a way to prevent indexing. If a page must stay out of results, use an appropriate method such as noindex or access controls.

For choosing permanent or temporary responses and reviewing replacement pages, use the redirects for SEO guide alongside the chain checks.

4. Check indexing and canonical signals

  • Sample important pages in URL Inspection.
  • Check that wanted pages do not carry accidental noindex rules.
  • Review canonical tags for redirects, missing pages or unrelated destinations.
  • Check sitemaps list canonical, indexable URLs that return a normal page.
  • Group duplicate URL patterns, such as tracking parameters or shop filters.

A canonical tag is a signal, not a command. Google may choose a different canonical, so compare its selected canonical with your intended one before deciding what to fix.

Use URL Inspection’s live test for representative templates and compare the rendered page with what visitors see. Check important content and navigation are present, and that links use crawlable anchor elements with useful text. The internal linking guide explains how to review orphan pages and anchors.

Use PageSpeed Insights for performance evidence and the Rich Results Test for structured data eligibility. A good score or valid markup does not guarantee rankings or a search feature.

For the implementation and maintenance workflow behind those checks, follow the schema markup guide: choose the right type, check existing output and verify that the data matches the visible content.

6. Turn findings into a prioritised plan

Clyde Home Supplies, Glasgow — fictional UK audit; illustrative only

  • High priority: a category template has an accidental noindex rule on 18 pages that receive enquiries. Remove it, then inspect three sample pages.
  • High priority: the main menu links to old URLs that redirect twice. Link directly to the final page.
  • Medium priority: filter URLs with tracking parameters appear in crawls. Review internal links and canonical choices.
  • Monitor: some images lack alt text. Improve useful product images during the next content update.
  • No action yet: a crawler warning reports long titles. Check whether titles are misleading before rewriting them.

The shop, page counts and findings are invented. No crawl, ranking change or revenue result has been observed.

For each issue, record affected URLs, evidence, likely impact, effort, owner and the check that proves it is fixed. Prioritise problems affecting important pages and access to search first. A warning repeated on thousands of unimportant URLs may matter less than one error on a key template.

7. Recheck fixes and measure carefully

After deployment, crawl the affected pages again, inspect samples in Search Console and confirm redirects, canonicals and indexability are correct. Separate “fixed in code”, “live on the website” and “processed by Google”.

Review indexing, clicks and enquiries over a consistent comparison period, noting seasonal changes and other site work. Search Console validation can help track a reported issue, but improved rankings after a fix are not proof the fix caused them. Report outcomes using the monthly SEO report guide.

Limitations and next step

An audit cannot guarantee rankings, and no tool score is a complete measure of technical SEO. Start with scope, compare Google’s evidence with a crawl and fix the issues blocking important pages first. For a specialist review, enquire about an SEO audit; this is an enquiry, not an instant booking.

For bulk checks on large lists of URLs or linking domains, my URL Profiler review shows one way to gather the data.

Sources and review notes

Google’s technical SEO and crawling-error guidance was checked on that date; other linked guides reuse previously checked sources. The audit order and fictional findings are editorial recommendations, not platform requirements or measured results.

Let's work together.

Whether you need an SEO partner, a consultant, a speaker for your event or training for your team — get in touch and let's chat.

Get in touch
Get in touch