A technical SEO audit should answer whether the right public pages can be discovered and understood. It should also leave a clear repair list. A green automated score cannot establish that a page deserves to rank or that the business will receive more enquiries.
Use this checklist on a few important templates before repeating it across a larger site.
Confirm discovery and indexing intent
Start with the page's purpose. Is it intended to appear in search, or is it a private account, internal search result, or duplicate view?
For an intended public page, check its response and visible content, internal links that lead to it, and any indexing directives. Review a verified property's URL Inspection result when you have Search Console access. A public-page audit alone cannot tell you everything Google has indexed.
A robots.txt rule controls crawling; it is not a reliable mechanism for keeping a URL out of search or protecting private content. Google's robots.txt introduction explains those limits. Private material needs actual access controls.
Check the preferred URL
Inspect redirects and canonical links. The declared canonical should represent the intended equivalent page, not an unrelated homepage. Keep internal links consistent with the preferred URL.
Canonical declarations are signals, not commands that guarantee Google's selected URL. Review Google's canonicalization guidance when dealing with duplicates, parameters, or migrated content.
Read the page as a prospective customer
Check that the title and main heading describe the actual service or answer. A meta description should be a useful summary, not a string of repeated phrases. Ask whether the content helps someone choose the next step.
Google can generate a search title and snippet from several page signals, so your title and description are not guaranteed verbatim search copy. Its SEO starter guide explains the basics.
Test links in the body, navigation, and main call to action. Confirm that the page makes sense on a narrow screen. Image alternatives should reflect the image's purpose; an empty alternative may be appropriate for a purely decorative image.
Separate technical findings from editorial judgments
An automated tool can detect that a title exists. Someone must judge whether it describes the page. A crawler can find a broken link. Someone must choose a useful replacement. An image-alt check cannot certify the site's accessibility.
Use this repair record:
Page and intended search purpose:
Observed problem and evidence:
Technical or editorial decision needed:
Owner and proposed change:
How the published result will be verified:
Search Console follow-up, if relevant:
Prioritize confirmed access or indexing-intent mistakes on important pages, then fix the individual content and navigation problems that obstruct understanding.
Keep AI-search work grounded
Google's guidance for AI features says ordinary SEO fundamentals still apply and no special AI schema is required. Clear answers and useful evidence are sensible editorial work; they do not guarantee an AI citation.
Avoid generating many near-identical pages solely for keyword variations. Improve a page when you can add a useful explanation, example, or decision aid.
Verify the change and preserve the date
Recheck the actual published URL after a repair. Preserve an article's original publication date and update its modification date only when the content was substantively revised.
Run a free page snapshot for an initial technical review. Use the maintenance report template to explain the completed work without promising a ranking outcome.