Back to Knowledge Hub
SEO

Technical SEO Audit: Checklist and Prioritizing Fixes

5 min read
Technical SEO Audit: Checklist and Prioritizing Fixes

The real value of a technical SEO audit lies in prioritizing issues by business impact. Learn what evidence to look for when verifying indexing, JavaScript, redirects, and internal links.

A technical SEO audit checks whether search engine crawlers can reach, process and correctly interpret the important content on your site. Its results are only useful if every finding comes with the affected URL, evidence, a recommended fix and a way to verify it.

This is the list we use, on our own site as well. One thing I want to say up front: the number of issues a tool reports is not a priority list. Accidentally excluding one important service page can be more urgent than dozens of warnings that make no business difference.

Which URLs to check first

Start with the homepage, the important service or category pages, and a few typical content pages. For a webshop, include product pages and variants. On a multilingual site, take samples from every language.

In Search Console, also look for pages that lost impressions or clicks, and note the dates of redesigns, migrations and bigger releases.

The goal is a sample that covers the different page templates. If a template is broken, one fix repairs many URLs at once.

Is the page reachable and indexable?

Check the HTTP response, the robots.txt rules and the tags that affect indexing. A 200 response only shows that the server sent something, not that Google indexed the page.

robots.txt controls crawling, noindex controls indexing. If the crawler is not allowed to fetch the page, it may never see the noindex on it.

Look at the canonical URL Google chose and the processing status in URL Inspection as well. "Indexed" does not mean ranking well.

How to verify the fix: the right URL works, the important content is accessible, and nothing is excluded by mistake.

What does the crawler see on a JavaScript site?

This is the point most people skip, and we learned why it matters on our own site. Our site is built with JavaScript. When we fetched every URL as Googlebot, it turned out that for some of them the crawler received an empty shell, with no title and no text, while everything looked perfect in the browser. In the re-check after the fix, 532 of 533 URLs were fine, and we fixed the last one separately. We now run that check regularly.

Compare the raw HTML response with the page as built in the browser. Google can run JavaScript, but rendering is a separate step, and many other crawlers, including several AI crawlers, cannot do it, or only partly.

Check that the title, main heading, text and internal links are really in the rendered content, and that after navigating between pages the previous page's title or canonical does not stay behind. On sites built with AI-assisted development tools, check this URL by URL as well: correct metadata in the editor does not prove the crawler gets the same.

How to verify the fix: the right content appears both on direct load and after internal navigation, and the HTML the crawler receives contains it.

Are canonicals and redirects consistent?

The canonical is a signal about the preferred URL. Internal links, the sitemap and redirects should all support the same version of a URL.

For a permanent move we use 301 or 308, for a temporary one usually 302 or 307. What matters is the meaning: do we want the new address shown permanently, or is the old one still the target?

It is a misleading simplification that a 302 "passes no link value". Google treats permanent and temporary redirects as signals of different strength.

How to verify the fix: the old URL leads to the right new page, there is no loop, and internal links already point to the final address.

Do internal links and the sitemap tell the same story?

Important pages should be reachable from the menu or related content, through standard links that contain the target URL.

The sitemap lists the URLs you want indexed. Listing them does not guarantee indexing, but non-existent, redirected or excluded URLs cause confusion. On a multilingual site it is separate work to include every language version and make sure hreflang pairs point to each other both ways. On the five-language Agrider site, that means thousands of URLs.

For webshops, check the category to product relationship and how filters are handled. I wrote about that in the e-commerce SEO guide.

Which performance data to look at

Lab tests are useful for debugging. Real user data shows what happens on actual visits. The two can differ.

Core Web Vitals measure when the main content appears, how fast the page responds to interaction, and how stable the layout is. When fixing, look for the specific cause: a heavy hero image, too much JavaScript at start-up, elements loading without reserved space. On our own homepage we cut the mobile download from 4.4 MB to 1.25 MB, mostly through images and less code loading up front.

A better score on its own does not prove more revenue or better rankings. Next to speed, look at the main client journey and how useful the content is.

The right order of fixes

First, whatever prevents important pages from being reached and displayed: an accidental noindex, broken redirects, empty rendering, a broken enquiry form.

Second, contradictions between URLs, internal linking and the performance of important templates. Smaller issues come after that.

For every task, write down the expected behaviour, for example: "when the article is opened directly, its own title and full text appear". That is far easier to verify than a general "SEO fixed".

What to ask for at the end of an audit

A short summary for decision makers and an actionable task list: affected URL or template, evidence, proposed fix, owner, acceptance criteria.

Before a redesign, the audit results belong in the development brief. If several teams work on the site, SEO consulting helps agree on a shared order of work.

Frequently asked questions

Is an automated audit report enough?

It is a good start, but the findings need to be checked and interpreted in business terms. Not every warning is an error, and not every error matters equally.

Does a 200 status mean the SEO is fine?

No. A server can return empty or broken content with a 200. That is exactly what happened with the empty-shell pages described above.

Are robots.txt and noindex the same thing?

No. The first controls crawling, the second gives an indexing instruction.

Does a new website need an audit?

Yes, a check before launch is especially useful. When replacing an old site, redirects and previously important URLs need to be reviewed as well.

If you want us to look at a specific problem, share the affected URL and what you are seeing. Technical work is part of our SEO service.

Background reading: Google JavaScript SEO basics, Redirects and Google Search.

This is what we do every day: SEO service.

SEO

Running an agency? White-label SEO for agencies

About the author

Balint Somos
Balint Somos

Balint Somos is the founder and project lead of Somos Media. He has built websites and run SEO for more than ten years, and leads the development of Melini's international stores.

Free

Free website auditby email within 24 hours

Our team reviews your website and emails you what's worth fixing within 24 hours. Speed, SEO, conversion, technical issues. No fluff, just concrete next steps.

  • PageSpeed and technical analysis
  • SEO issues and opportunities
  • Conversion recommendations, prioritized

No commitment. We don't send spam.

Would you rather talk about your project?Request a consultation