SEO Audit Checklist: A Practical Step-by-Step Guide

An SEO audit is only useful if it ends in a short list of fixes, ranked by impact. This checklist works through a site in the order problems depend on each other: a page that can't be crawled can't be indexed, and a page that isn't indexed can't benefit from a better title. Work top to bottom, write down what you find, and don't start polishing meta descriptions until the foundations pass.

Before you start#

Gather the basics so you aren't guessing:

• Access to Google Search Console for the property, especially the Page indexing report and URL Inspection.
• A list of your most important URLs: home page, top categories, key product or service pages, top articles.
• The platform the site runs on, since many issues are template-level and a fix to one template fixes thousands of pages.

Start with a quick pass through the SEO Checker on the home page and two or three key templates. It gives you a first view of status codes, tags and headings before you go deeper. Google's SEO starter guide is a good reference for the fundamentals behind each check.

1. Crawlability#

Can search engines reach your pages?

• robots.txt exists and is intentional. Open /robots.txt and read every rule. Look for leftover Disallow: / from a staging environment and rules that block CSS or JavaScript needed to render pages. Test specific URLs with the Robots.txt Checker.
• robots.txt is not being used to hide pages from search. Google states that robots.txt "is not a mechanism for keeping a web page out of Google" (see Google's robots.txt introduction). Use noindex or authentication for that.
• Important pages return 200. Check that key URLs don't return 4xx, 5xx or soft-404 content.
• Pages are reachable through links. A page that nothing links to may never be found, even if it's in the sitemap.
• Server stability. Frequent timeouts or 5xx errors slow crawling down.

Our robots.txt guide covers rule syntax and common traps.

2. Indexability#

Once a crawler reaches a page, will it be indexed?

• No accidental noindex. Check the robots meta tag and the X-Robots-Tag header on key templates.
• Canonicals point to the right URL. Each page's canonical should be itself or its true preferred version: same protocol, same host, no redirect.
• One version of the site. http versus https and www versus non-www should all redirect to a single version.
• The XML sitemap is clean. It should list only canonical, indexable, 200-status URLs. Validate it with the Sitemap Checker. Google's sitemap overview explains when a sitemap is most useful, for example on large sites, new sites with few links, and sites with a lot of media.
• Search Console's Page indexing report matches your expectations. Read the reasons given for excluded pages and decide which exclusions are intended.

For sitemap structure, see how to create a sitemap.

3. On-page elements#

• Unique, descriptive titles on every indexable page, led by the topic.
• Meta descriptions that summarize each important page. They're used for snippets, not rankings, but they affect how the result reads.
• One clear main heading that matches the page's purpose, and a logical H2/H3 structure.
• Image alt text that describes meaningful images.
• No duplicated titles or descriptions across a template.
• Open Graph tags so shared links display correctly.

Details and examples are in meta tags for SEO.

4. Technical health#

• HTTPS everywhere with a valid certificate on every hostname, and no mixed content. See SSL certificate errors.
• Redirects are direct. Old URLs should redirect in one hop to the final destination, with permanent redirects for permanent moves. Chains and loops waste crawl time and lose clarity. See 301 vs 302 redirects.
• Broken internal links fixed. Every internal link to a 404 is a dead end for users and crawlers. See how to fix broken links.
• Mobile rendering works. A viewport tag is present, text is readable and tap targets aren't crowded.
• JavaScript-rendered content is visible to Google. Use URL Inspection's rendered view to confirm that main content and links appear.
• Clean URL parameters. Faceted navigation and tracking parameters shouldn't create large numbers of crawlable duplicates.

5. Speed and Core Web Vitals#

Google's Core Web Vitals, as defined on web.dev, measure loading (LCP), interactivity (INP) and visual stability (CLS). The "good" thresholds are LCP within 2.5 seconds, INP of 200 milliseconds or less, and CLS of 0.1 or less, measured at the 75th percentile of page loads.

• Test key templates with the Speed Test, which uses Google's PageSpeed Insights API.
• Separate field data (real users) from lab data (a simulated load). Field data tells you whether there's a problem. Lab data helps you find the cause.
• Check mobile first. Mobile results are usually the weaker ones.

Our guide on how to improve website speed covers the fixes.

6. Structured data#

• The markup on each template is valid and matches visible content.
• There are no conflicting duplicate blocks from multiple plugins.
• Only relevant types are used. FAQ and HowTo rich results are no longer shown by Google.
• Search Console's enhancement reports show no new errors.

See the schema markup guide.

7. Internal links#

• Important pages get links from relevant pages, not just from the navigation.
• Anchor text is descriptive. "Robots.txt guide" tells users and crawlers more than "click here." See the anchor text guide.
• No orphan pages. Compare the URLs in your sitemap against the URLs found by crawling links. Pages that appear only in the sitemap need links.
• Click depth is reasonable. Key pages shouldn't sit many clicks from the home page.

8. Content quality#

This is the hardest part to automate and often the most important. Google's guidance on helpful content asks whether a page provides original information, substantial coverage and a satisfying experience for the people who land on it.

• Thin pages: near-empty tag pages, auto-generated location pages and placeholder products. Improve them, consolidate them or noindex them.
• Overlapping pages targeting the same query. Merge them and redirect the weaker page.
• Outdated content with wrong prices, dead products or old instructions.
• Clear authorship and sourcing where trust matters.
• Search intent match. If results for a query are all how-to guides and your page is a sales page, the format is the problem, not the keywords.

9. AI readiness#

AI assistants increasingly answer questions by retrieving and citing web pages, so check how your site treats their crawlers:

• Review your AI crawler rules on purpose. Decide which AI user agents to allow. Blocking search-type agents can keep you out of those products' answers. See how to allow or block AI crawlers.
• Don't expect special files to do the work. Google says there are no additional requirements or special optimizations for its AI features beyond normal SEO. An /llms.txt file is optional. See what is llms.txt.
• Describe your business plainly on your own site: what you offer, for whom, at what price. See how brands get cited by AI.

Turning findings into a plan#

Record each issue in a consistent format so it can be prioritized and assigned:

Issue:     Category pages carry noindex from staging template
Scope:     All /category/ URLs (template-level)
Evidence:  <meta name="robots" content="noindex"> on /category/shoes/
Impact:    High - pages cannot appear in search
Effort:    Low - remove tag from category template
Owner:     Dev
Verify:    URL Inspection shows "Indexing allowed"; Page indexing report

Then sort:

1. High impact, low effort first: accidental noindex, robots.txt blocks, broken redirects on key pages.
2. Template-level fixes next, because one change covers many URLs.
3. Content work last. It takes longer but builds on a sound technical base.

Re-run the same checks after each fix ships. An audit is a baseline you return to, not a one-time report.