How do you perform a technical SEO audit?

Start with evidence, not opinions

I usually begin with a crawl, sitemap export, Google Search Console data, server status checks and a manual look at key templates. The goal is to see the site the way crawlers see it. If the homepage looks polished but important service pages are four clicks deep, the audit needs to catch that.

Then I compare systems: what the sitemap says should exist, what the crawler finds, what Google indexes, what gets impressions, and what receives internal links. Gaps between those systems often reveal the real problem.

The audit areas

A useful audit covers status codes, redirects, canonicals, robots directives, sitemap quality, duplicate pages, thin templates, JavaScript rendering, Core Web Vitals, mobile rendering, structured data, internal links, hreflang if needed, pagination, faceted navigation and log files when scale justifies it.

Each issue needs priority. A missing alt attribute is not equal to a canonical loop on every category page. I separate blockers, growth constraints and hygiene tasks. Developers need that order or the sprint turns into random cleanup.

Deliverables that actually help

The best deliverable is a fix backlog: issue, affected URLs, evidence, business impact, recommended implementation, owner and priority. For larger sites, I add template-level examples and acceptance criteria so the fix can be tested.

After implementation, I re-crawl. Without verification, an audit is just advice. Technical SEO only counts when the site changes and search engines can process the change.

My working checklist

• Crawl the site and export index data.
• Compare sitemap, crawl, index and traffic layers.
• Prioritize blockers before hygiene issues.
• Re-crawl after implementation.

Related SEO questions

This answer belongs to the same SEO wiki cocoon. Start from the SEO Wiki hub, then use these related answers to move sideways through the cluster.