A technical SEO audit checks whether search engines can crawl, render and index a website's pages, and whether anything in the site's code, server or structure is holding rankings back. It covers status codes, robots.txt, sitemaps, canonicals, JavaScript rendering, Core Web Vitals and structured data. The output is a ranked list of fixes, each with evidence and an owner.
This is the technical layer of our complete SEO audit guide, which also covers on-page content, links and AI visibility. If you want the checks as a working list, use the technical SEO checklist.
What a technical SEO audit covers
A technical SEO audit covers everything between a bot requesting a URL and that URL appearing in the index. It judges whether content can be found, read and stored, not whether it is good.
| Area | Question | Main tool |
|---|---|---|
| Crawlability | Can bots reach every important URL? | Crawler, server logs |
| Indexability | Are the right URLs indexed, and only those? | Search Console Page indexing report |
| Canonicalization | Does every page have one agreed version? | Crawler, URL Inspection |
| Rendering | Is the content in the HTML bots receive? | View source, URL Inspection live test |
| Performance | Do templates pass Core Web Vitals? | Search Console, PageSpeed Insights |
| Structured data | Is schema valid and accurate? | Rich Results Test, Schema.org validator |
| AI crawler access | Can answer engines fetch the pages? | robots.txt, CDN and firewall logs |
How to run a technical SEO audit in 7 checks
Run the checks in order, since each depends on the one before.
1. Crawlability
Crawlability is the first check because nothing else matters if bots cannot reach the page. Crawl the site with a tool such as Screaming Frog or Sitebulb and look for:
- Internal links pointing to 4xx pages or to redirects.
- 5xx server errors, especially intermittent ones. Google's documentation on HTTP status codes and network errors says 5xx errors make Google's crawlers temporarily slow down, and that URLs which persistently return server errors are eventually dropped from the index.
- Redirect chains and loops. Collapse every chain to a single 301.
- Orphan pages that appear in the sitemap but receive no internal links.
- Important pages buried more than three or four clicks from the homepage.
2. Indexability
Indexability is where audits find their biggest wins. Open the Page indexing report and work through each "not indexed" reason. "Crawled, currently not indexed" usually signals thin or duplicate content. "Discovered, currently not indexed" often signals weak internal linking or, on very large sites, crawl budget limits.
Then compare three numbers: URLs in the sitemap, indexable URLs found by the crawl, and URLs Google reports as indexed. Explain every large gap.
Check directives too. A noindex meta tag or X-Robots-Tag header on a template can remove thousands of pages at once. Confirm every noindexed URL is meant to be, and that none of them sit in the sitemap.
3. robots.txt, sitemaps and canonicals
These three tell search engines what to crawl, what exists and which version counts. Read them line by line.
- robots.txt should return a 200, allow CSS and JavaScript, and block only what you mean to block. It controls crawling, not indexing.
- XML sitemaps should list only canonical, indexable, 200-status URLs. Google ignores
priorityandchangefreq, so do not spend time tuning them. - Canonical tags should be self-referencing on indexable pages and agree with internal links, redirects and the sitemap. Google's guide to consolidating duplicate URLs explains that canonical tags are a strong signal but not a directive. When your signals disagree, Google picks.
4. JavaScript rendering
Rendering is the check that modern sites fail most quietly. Google's JavaScript SEO basics describe crawling, rendering and indexing as separate phases, and recommend server-side rendering or pre-rendering because not all bots can run JavaScript.
Test three ways:
- Open the page source and search for a sentence from the main content.
- Run the URL Inspection live test and read the rendered HTML.
- Compare a crawl with JavaScript rendering on and off, and diff word counts and link counts.
If titles, canonicals, body text or internal links only exist after rendering, log it as a P1 finding and name the component.
5. Core Web Vitals and mobile
Performance is a ranking tiebreaker and a conversion factor. As of October 2026, Google's Core Web Vitals documentation defines "good" as LCP within 2.5 seconds, INP under 200 milliseconds and CLS under 0.1.
Use the Search Console Core Web Vitals report for field data grouped by template. Then use PageSpeed Insights or Lighthouse to diagnose why a template fails. Also confirm mobile parity: Google indexes the mobile version, so content, titles and structured data must match desktop.
6. Structured data
Check that structured data is valid, matches visible content, and uses the right types on the right templates: Organization on the homepage, Article on posts, Product on product pages. Test a sample of each template in the Rich Results Test and the Schema.org validator.
7. AI crawler access
AI crawler access is the newest check in a technical SEO audit. Confirm that robots.txt and your CDN or firewall allow the search bots you want in AI answers, such as OAI-SearchBot and PerplexityBot. OpenAI's crawler documentation separates OAI-SearchBot, used to show sites in ChatGPT search, from GPTBot, used for training, so you can allow one and block the other.
Technical SEO audit checklist: what to fix first
A technical SEO audit checklist is only useful if it ranks fixes. Use this order.
| Priority | Finding type | Example |
|---|---|---|
| P0 | Blocks indexing of revenue pages | Noindex on a template, robots.txt blocking /products/ |
| P1 | Wrong signals on many pages | Canonicals pointing to redirects, content missing from raw HTML |
| P2 | Slows crawling or ranking | Redirect chains, failing Core Web Vitals template, orphan pages |
| P3 | Hygiene | Single redirect hops, minor schema warnings |
Group tickets by template: fixing one template often resolves thousands of URLs.
Technical SEO audit services vs doing it yourself
You can run a technical SEO audit yourself on a small site with free tools: Search Console, PageSpeed Insights and a crawler's free tier. Our roundup of SEO audit tools compares the options. Technical SEO audit services make more sense for large sites, migrations, JavaScript-heavy builds or ecommerce faceted navigation, where log analysis and developer-ready tickets take experience.
Frequently asked questions
What is a technical SEO audit?
A technical SEO audit is a review of how well search engines can crawl, render and index a website. It checks status codes, robots.txt, XML sitemaps, canonical tags, JavaScript rendering, Core Web Vitals, structured data and AI crawler access. It ends with a ranked list of fixes for developers and does not judge whether the content itself is good.
How long does a technical SEO audit take?
A technical SEO audit of a small site of a few hundred pages takes one to three days. A site with tens of thousands of URLs, faceted navigation or a JavaScript framework can take two to four weeks. Most of that time goes into analyzing crawl data and server logs and writing developer-ready tickets, not into running the crawling tools.
What tools do you need for a technical SEO audit?
You need Google Search Console for index and performance data, a crawler such as Screaming Frog, Sitebulb, Semrush Site Audit or Ahrefs Webmaster Tools, and PageSpeed Insights for speed diagnosis. For structured data, use the Rich Results Test. Large sites also need server log access, which is the only way to see exactly what Googlebot requests.
What is the difference between a technical SEO audit and a full SEO audit?
A technical SEO audit covers crawling, indexing, rendering, speed and structured data. A full SEO audit includes all of that plus on-page relevance, content quality, backlinks, internal linking strategy and visibility in AI answers. Run a technical audit alone when the problem is clearly technical, such as after a migration or a sudden indexing drop.
How often should you run a technical SEO audit?
Run a full technical SEO audit at least once a year, and before and after any migration, redesign or platform change. Between audits, check the Search Console Page indexing and Core Web Vitals reports monthly. Large sites that ship code weekly benefit from a scheduled crawl that alerts on new noindex tags, canonical changes or spikes in errors.
If you want a technical SEO audit done for your site, AEO GEO Labs runs fixed-fee audits that include the AI search layer, for B2B and SaaS teams. See our services.