site audit

A site audit crawls your site the way a search engine would, following internal links and recording what it finds on each URL. Tools like Semrush run over 140 on-page and technical checks per page, then sort results into three severity tiers: errors (most harmful), warnings, and notices.

What a crawl actually checks

Findings usually cluster into a few buckets:

• Crawlability and architecture: robots.txt blocks, orphan pages, deep crawl depth, redirect chains.
• Indexability: accidental noindex, non-canonical URLs sitting in the sitemap, 404 and 301 status problems.
• Performance: failing Core Web Vitals, uncompressed assets, render-blocking scripts.
• Content and markup: duplicate titles, thin pages, broken or invalid structured data.

Audit tools now add AI-search checks too, things like llms.txt, semantic HTML, and answer-shaped content, because the same clean architecture that helps Googlebot also helps AI crawlers reach and quote your pages.

Two cautions. A site audit is a snapshot, not a fix: the crawler hands you a list, and someone still has to judge which items move rankings versus which are cosmetic noise. And a “site audit” is broader than an on-page or content audit. It looks at the whole property’s technical health, not one page. That whole-site view is why it sits at the front of most SEO engagements.