The technical SEO audit I run before touching anything

Most SEO work fails quietly. Someone rewrites the titles, adds a blog, builds a few links — and nothing moves, because the site was never eligible to rank in the first place. Before I change anything on a site I run the same nine checks, in the same order. Most of them take minutes.

1. Is it indexed at all?

A site: search gives a rough count. Search Console's coverage report gives the truth, and the reasons: excluded by noindex, blocked by robots.txt, crawled and not indexed, discovered and not crawled. A staging noindex that shipped to production is the single most common cause of a site that "does not rank for anything", and it is a one-line fix that nobody looks for because it is too obvious.

2. Does robots.txt block anything that matters?

Usually it is fine. Occasionally it disallows /wp-content/, which stops Google fetching the CSS and JavaScript it needs to render the page, and a page Google cannot render is a page it judges on the raw HTML. On sites with a search or filter system, the opposite problem: nothing is blocked and the crawler is spending its budget on ten thousand filter combinations.

3. Do the canonicals point where you think?

I check the canonical on a handful of page types — home, a category, a product or service, a paginated page. Self-referencing is the default and the right answer in most cases. What I am looking for is the site that canonicalises every paginated page back to page one, or the migration that left every canonical pointing at the old domain.

4. Are the redirects one hop, and permanent?

Four URL variants exist for every site: with and without www, over HTTP and HTTPS. Three of them should 301 to the fourth in exactly one hop. Chains lose a little authority at each step and cost real time on a mobile connection. I test each variant by hand rather than trusting the server config.

5. Is the sitemap true?

A sitemap that lists redirects, 404s, noindex pages or drafts is a sitemap telling Google you do not know what is on your own site. Every URL in it should return 200 and be indexable, lastmod should reflect an actual change, and the file should be referenced from robots.txt.

6. Does the page render without JavaScript?

Not because Google cannot execute JavaScript — it can — but because rendering is a second, slower, queued pass. If the main content, the internal links and the title exist in the server response, indexing is immediate and reliable. If they are assembled in the browser, everything about indexation gets slower and less certain.

7. What do the Core Web Vitals say in the field?

Lab tools tell you what a page can do. The Chrome UX Report tells you what it did, for real visitors, on their real devices. The two frequently disagree, and when they do the field data is the one Google uses. INP is the one that catches people out since it replaced FID: it measures every interaction, not just the first, so a heavy page-builder site can pass FID historically and fail INP now.

8. Is the structured data valid, and is it about the right thing?

Two separate questions. Valid but wrong is common: an Organization block on every page of a personal site, a Product block on a category listing, a FAQPage on a page with no visible questions. I validate the syntax, then check that each type actually describes what the page is.

9. Can you reach every important page from the homepage in three clicks?

Internal links are the part of SEO you fully control, and the part most often left to the theme's menu. A page that is only linked from a sitemap file is a page you have told Google not to care about. I map the click depth of the pages that matter commercially and fix the ones that are five levels down.

Then, and only then

After those nine, content and links are worth doing, and they work. Before them, they are a slower and more expensive way of finding out that the site had noindex on it.