DIY Technical SEO Audit: A 10-Point Checklist You Can Run Yourself

A useful technical SEO audit checks ten things, in this order: indexing, canonicals, sitemap, robots.txt, redirects, mobile rendering, speed, internal links, structured data and duplicate content. The order matters more than the list. Most audits start with speed scores because they are easy to generate, and a site with a canonical error is not helped by a faster-loading page nobody can find.

You do not need a paid crawler for a first pass. Google Search Console, the free tier of a crawler like Screaming Frog, PageSpeed Insights and your browser’s view-source are enough to find almost everything that genuinely costs you traffic. What follows is the sequence, with what to look for and what to ignore at each step.

1. Is your content actually indexed?

Open Search Console → Pages. Compare the indexed count against the number of pages you believe you have published. If indexed pages are a small fraction of your real total, stop the audit here and fix that first; nothing else on this list will produce traffic until it is resolved.

Read the reasons under “Why pages aren’t indexed” in order of count. “Discovered – currently not indexed” usually means a crawl-budget or quality signal issue. “Crawled – currently not indexed” means Google saw the page and chose not to store it. “Alternate page with proper canonical tag” is often fine. “Excluded by noindex tag” is either correct or a disaster, depending on which pages are in the list — check it page by page, because an accidental sitewide noindex is one of the few errors that can erase a site’s traffic overnight.

The two patterns that cost the most are worth naming. First, a site relaunched on a different hostname — moving from www to non-www, or http to https — while Search Console still reports on the old property. Second, a staging site’s noindex rule shipped to production. We cover the diagnosis in detail in crawled but currently not indexed.

2. Do your canonical tags point where you think?

View the source of five pages — homepage, a post, a category, a tag archive and a paginated page — and find the rel="canonical" line. Each should point to itself, in the exact form you want indexed: right protocol, right host, no trailing-slash inconsistency, no query strings.

The classic failure is a plugin or theme setting that canonicals every page to the homepage. It is rare but catastrophic, and it looks like nothing is wrong until you check the source. Spend the two minutes.

3. Does your sitemap load, and does it tell the truth?

Open your sitemap index URL in a browser. Three things to confirm: it returns XML rather than an error or an HTML page, every URL inside uses your canonical hostname, and it contains only pages you want indexed. A sitemap listing noindexed pages, redirects or 404s sends contradictory instructions and wastes crawl budget.

A WordPress quirk worth knowing: if your SEO plugin’s sitemap returns a 404 while the plugin is clearly active, the usual cause is stale rewrite rules rather than a broken setting. Visiting Settings → Permalinks and saving without changing anything regenerates them and commonly fixes it on the spot.

4. What is your robots.txt blocking?

Load yourdomain.com/robots.txt and read every line. You are looking for rules that block more than intended — a Disallow: /wp-content/ will stop Google loading your CSS and images, which breaks how it renders your pages. You are also checking that your sitemap is declared, and deciding deliberately whether AI crawlers are allowed.

User-agent: *
Disallow: /wp-admin/
Allow: /wp-admin/admin-ajax.php

Sitemap: https://yourdomain.com/sitemap_index.xml

That is a complete, correct robots.txt for most WordPress sites. If yours is substantially longer, read the extra lines carefully — long robots files accumulate rules nobody remembers adding, and each one is a chance to block something you need crawled.

5. Are your redirects clean?

Test a handful of old URLs with a redirect checker and note the number of hops. Every extra hop adds latency and, in Google’s own guidance, risks diluting the signals you are trying to pass. Rewrite multi-step chains as single 301s to the final destination. Also check for redirect loops, and for 302s used where a move is permanent — a temporary redirect tells Google to keep the old URL indexed.

6. Does the mobile version contain everything?

Google indexes the mobile version of your site. If your mobile template hides content, collapses sections that never load, drops structured data or omits internal links present on desktop, then as far as Google is concerned that content does not exist.

Test it properly: use Search Console’s URL Inspection tool, run a live test, and open the rendered HTML tab. Read what is actually there rather than trusting a visual check. Then confirm tap targets are large enough to use — a 15-pixel icon is technically present and practically broken, and these failures show up in behaviour data long before anyone calls them SEO problems.

7. Which speed numbers are worth chasing?

Run PageSpeed Insights on your homepage and two interior pages, and read the field data at the top rather than the lab score below it. Field data comes from real visitors; the lab score comes from a simulated device and moves around between runs. Chasing the lab number is the single most common waste of effort in technical SEO.

Each of those three has a predictable main cause, which makes them unusually tractable. Slow LCP is nearly always a large unoptimised hero image or a render-blocking stylesheet. Poor INP is nearly always too much JavaScript running on load. Layout shift is nearly always images and ads without declared width and height. Fix the cause rather than the score.

8. Can every page be reached by a crawler?

Crawl your site with the free tier of any crawler and look for orphan pages — URLs in your sitemap that no internal link points to. An orphan page is findable by Google only through the sitemap, which is a weak signal, and it almost never ranks. Equally, check your click depth: pages more than three clicks from the homepage are crawled less often and treated as less important.

Fixing this is editing, not engineering: add links from relevant existing posts, from category hubs, and from your most-visited pages. The full method is in our guide to internal linking and site structure.

9. Is your structured data valid?

Run one page of each template through the Rich Results Test, then open Search Console → Enhancements. Fix everything marked Invalid before anything marked Warning — warnings are missing recommended fields, errors mean the markup is being discarded. The detail is in our schema markup guide.

10. How much of your site is duplicated?

Sort your crawl by title tag and look for repeats. Identical titles across many URLs usually mean tag archives, filtered views, session parameters or paginated series are generating near-duplicate pages. Decide for each group: noindex it, canonical it to a single version, or block the parameter. Doing nothing is the only wrong answer, because Google will pick a canonical for you and may not pick the one you want.

What should you ignore?

Automated audit tools generate hundreds of findings, most of which do not matter. A short list of things not to spend your afternoon on: meta keywords, which no search engine has used for over a decade; keyword density figures; a perfect 100 PageSpeed score; “too many internal links” warnings on a genuinely useful navigation; missing alt text on decorative images; and W3C HTML validation errors that do not affect rendering.

A tool that reports 400 issues is reporting its thresholds, not your priorities. The ten steps above cover the problems that actually suppress traffic. Everything else is housekeeping you can do when there is nothing important left.

How often should you repeat this?

Quarterly for a stable site, and immediately after any of four events: a theme change, a migration or hostname change, a plugin update that touches SEO or caching, and a redesign. Those four cause most of the technical regressions that go unnoticed for months, because nobody thinks of a visual change as an indexing risk.

Keep the output as a dated one-page note: what you checked, what you found, what you changed. The value compounds — next quarter you will be comparing against a record rather than starting over, and you will catch the slow drift that single audits always miss. For the content side of the same work, see the on-page SEO elements that matter and SEO explained in simple steps; for the strategy layer, ranking factors in the GEO and AEO era. More in our technology and digital marketing section.