Technical SEO audits that actually move rankings: five checks worth doing first

A technical SEO audit tool will happily generate a report with two hundred flagged issues. Almost none of them move the needle on their own, and working through the list top to bottom burns a lot of hours on things like missing alt text on decorative icons while the page that's actually losing rankings goes untouched.

We run technical audits as part of most marketing engagements, and over time the list of checks that reliably correlate with real ranking or traffic movement has stayed short. These five are where we start, before anything else.

1. Is the page indexed at all, and under the URL you think it is

Before optimising anything, confirm Google actually has the page in its index, and check what URL it's indexed under. A stray `noindex` tag left over from staging, a canonical pointing at the wrong page, or a redirect chain that resolves somewhere unintended will silently cap every other improvement at zero. This is a five-minute check in Search Console's URL Inspection tool and it catches more real damage than almost anything else on a typical audit.

2. Does the canonical URL match the URL people actually land on

Duplicate content from URL parameters, trailing slashes, or www/non-www inconsistency splits ranking signal across multiple versions of the same page instead of consolidating it onto one. We check that every internal link, the sitemap, and the canonical tag all agree on one consistent URL form - not three different opinions about whether a trailing slash belongs there.

3. Is the content on the page the content Google actually renders

Content injected client-side after page load is sometimes indexed slower, or missed entirely, depending on how it's rendered. We diff the raw HTML source against what actually shows on screen - if the two disagree meaningfully (a price, a key paragraph, a product description that only appears after a JavaScript fetch resolves), that's a real risk worth fixing before anything cosmetic.

4. Are Core Web Vitals failing on real user data, not just a lab test

A PageSpeed Insights lab score and real Core Web Vitals field data (from actual visitors, via the Chrome UX Report) frequently disagree - a page can score well in a clean lab run and still fail in the field because of a slow third-party script that only loads for some visitors, or a network condition the lab test doesn't simulate. We prioritise field data because that's what Google's ranking systems actually use, and it tells you whether real visitors are having a bad experience, not just whether a synthetic test run was.

5. Does the internal linking actually reach the pages you want ranked

• Pull a crawl and check click-depth from the homepage - a page four or five clicks deep gets crawled less often and passes less internal authority.
• Confirm your most commercially important pages are linked from somewhere with real traffic, not just from a sitemap file nobody clicks.
• Look for orphaned pages: indexed, but with zero internal links pointing to them - they're relying entirely on external signals to rank.
• Check that navigation and footer links use real anchor text, not just "click here" or a bare icon with no accessible label.

A fifty-item audit checklist treats every issue as equally urgent. In practice, five checks explain most of the ranking problems we actually find.

What should you check first in a technical SEO audit?

Start by confirming the page is actually indexed under the URL you expect - a stray noindex tag or misdirected canonical caps everything else at zero. Then check canonical consistency, whether Google renders the same content a visitor sees, real-user Core Web Vitals rather than lab scores, and whether your important pages are reachable in a few internal clicks from the homepage.