Skip to content
SeoData

Technical SEO audit checklist and what to check first

A technical audit can produce hundreds of warnings, and fixing them in the order they appear wastes weeks. This technical SEO audit checklist puts the checks in priority order, so you start with problems that block search engines entirely and finish with refinements that only matter once the basics are solid.

Last updated October 20, 2026

Close detail of a printed website audit page on a desk with handwritten priority numbers in the margin and a blue pen resting on it

How to use this checklist

Work from the top down. Each group depends on the one before it. A fast page that cannot be crawled earns nothing, and perfect structured data on a page that is excluded from the index is wasted effort. Run a full crawl first with one of the SEO audit tools you trust, export the issues, then walk through the groups below and mark each check as passed, failed or not relevant.

Before you start, agree on scope. Decide whether the audit covers the whole domain, one subfolder or one language version, and whether staging or test hosts are excluded. Note which page templates exist, such as homepage, category, product, article and landing page, because most technical issues are template issues and the fix is made once per template.

Record the date and the number of crawled pages. You will compare the next audit against this baseline, and the change between audits is often more useful than the audit itself.

1. Crawling and access

If search engines cannot reach your pages, nothing else matters. Check these first.

  • robots.txt loads, returns a 200 status and does not block important sections or the resources needed to render pages.
  • Important pages return a 200 status. Note every 4xx and 5xx response and the pages that link to them.
  • Redirects are single hops. Chains and loops waste crawl budget and dilute signals.
  • The site has one preferred version. HTTP, HTTPS, www and non-www all resolve to the same host with a single redirect.
  • Server responses are stable. Repeated timeouts or 5xx errors during the crawl point to hosting problems.

Compare the number of crawled pages with the number of pages you expect the site to have. A crawl that finds far fewer pages usually means blocked sections or links that only appear after scripts run. A crawl that finds far more usually means parameters, session IDs or calendar pages generating endless URLs.

2. Indexing signals

Once pages can be crawled, confirm they are allowed into the index and that you send consistent signals.

  • No important page carries a noindex tag or an X-Robots-Tag header by mistake.
  • Every indexable page has a self-referencing canonical, or a canonical pointing to the correct preferred URL.
  • Canonicals do not point to redirected, broken or noindexed pages.
  • The XML sitemap lists only indexable pages that return 200, and it is referenced in robots.txt.
  • Parameter URLs, filters and internal search results do not create large numbers of thin duplicate pages.
  • Thin or near-duplicate pages are merged, improved or excluded on purpose.

If you have access to the site's search console data, compare the pages you want indexed with the pages actually indexed. Gaps in either direction are the most direct sign of an indexing problem, and they often explain traffic drops that look like ranking losses.

Structure tells search engines which pages are important. Internal links are the part of SEO you control completely.

  • Important pages are reachable within three or four clicks from the homepage.
  • No valuable page is orphaned, meaning it has no internal links pointing to it.
  • Navigation and category pages link to the pages you want to rank.
  • Internal anchors describe the target page instead of saying click here.
  • Internal links point to final URLs, not to redirects or broken pages.
  • Pages that earn many external links pass value to commercial pages through contextual internal links.

A simple way to review structure is to list your twenty most important pages and count the internal links pointing to each one. Pages that matter commercially but receive only one or two links from deep inside the site are easy wins. Add links from navigation, related content blocks and relevant articles.

4. On-page technical elements

These checks are quick to fix in bulk and often show up across thousands of pages at once.

  • Every indexable page has a unique title tag that starts with its main topic.
  • Meta descriptions are unique and describe the page. Missing ones are less urgent than duplicated ones.
  • Each page has a single, descriptive H1.
  • Images have meaningful alt text where they carry information.
  • Language and region tags are correct on multilingual sites, and hreflang pairs point to each other.

To check a single page in detail before rolling a fix across the site, a website SEO checker gives a fast page-level report.

5. Page speed, rendering, mobile and security

Speed problems rarely stop indexing, but they affect user experience and can hold back pages that compete closely.

  • Core page templates load their main content quickly on a mid-range mobile device.
  • Layout does not shift as images, fonts and ads load.
  • The page responds quickly to the first tap or click.
  • Large images are compressed and served at the size they are displayed.
  • Critical content is present in the HTML or renders reliably, rather than appearing only after heavy scripts run.

Test templates, not single URLs. Fixing the product template or the article template fixes every page built on it.

Mobile and security belong in the same pass. Most crawling happens with a mobile user agent, so the mobile version is the version that counts.

  • Mobile pages contain the same main content, links and structured data as desktop pages.
  • Text is readable and tap targets are not crowded on small screens.
  • The whole site loads over HTTPS with a valid certificate.
  • No page mixes secure and insecure resources.

6. Structured data and search features

Structured data helps search engines understand a page and can make it eligible for richer results.

  • Markup matches the visible content of the page.
  • Required properties are present for each type you use, such as Article, Product or BreadcrumbList.
  • Breadcrumb markup reflects the real site hierarchy.
  • No markup describes content that is not on the page.

Validate markup on one page per template rather than page by page. Errors in structured data almost always come from the template, so one fix clears the issue across the whole section.

7. Prioritize, fix and track

At the end of the audit you will have a long list. Turn it into a short plan.

  • Rank issues by impact and by the number of important pages affected, not by the number of warnings.
  • Fix crawling and indexing blockers first, then structure, then page-level elements, then speed refinements.
  • Group fixes by template so one change resolves many pages.
  • Re-crawl after each release and compare with the baseline to confirm the fix and catch regressions.

Assign every fix an owner and a target date. Many audits fail not because issues are hard, but because the list sits in a document nobody owns. Developers respond better to a short ticket with the affected template, three example URLs and the expected result than to a long export of warnings.

After the first full audit, schedule a lighter crawl every month. New releases, plugin updates and content migrations introduce fresh problems, and catching them within weeks is far cheaper than finding them a year later.

An AI action plan can sort audit findings by expected impact so the first tasks on your list are the ones that matter most. Watch rankings for key pages with position tracking after major fixes, so you can connect technical changes to search results.

Analyze domain

SeoData crawls your site, groups issues by priority, compares each audit with the previous one and turns the findings into an action plan ranked by impact. Enter a domain to start your audit.

Analyze domain