Technical SEO Audit: How to Run One Step by Step
Short answer List the pages that must rank, crawl the whole site with a free crawler, then read the results in this order: errors and blocks, redirects and duplicates, what Google actually indexed, phone and HTTPS, speed. Rank each problem by damage times whether it hits a key page. Fix blocks on key pages first; most crawler warnings can wait.
Want help applying this to your business? Talk it through on a free 30-minute call.
Book a callTechnical SEO audit: the steps in order
A technical audit answers one question: can Google reach, load and store the pages that bring you customers? Google's own bar has three parts:
"Googlebot isn't blocked." "The page works, meaning that Google receives an HTTP 200 (success) status code." "The page has indexable content." (Google Search Central: Google Search technical requirements)
Googlebot is Google's crawler, the program that visits pages. A status code is the three-digit reply a server gives: 200 means OK, 404 means not found. The same page warns that "indexing isn't guaranteed" even when all three hold. So the audit finds what breaks the bar, then ranks the fixes.
| Step | Tool (free) | What you're looking for |
|---|---|---|
| 1. List your key pages | A spreadsheet | The 5–20 pages that must rank: home, each service, each location |
| 2. Crawl the site | Screaming Frog SEO Spider, free up to 500 URLs | Every address, its status code, redirects, noindex tags, canonicals |
| 3. Read the crawl | The crawl export | Errors, blocks, chains, duplicates |
| 4. Compare with Google | Search Console: Page indexing, URL Inspection | Key pages Google left out, and why |
| 5. Phone and HTTPS | Lighthouse, Search Console HTTPS report | Content missing on phones, pages not served securely |
| 6. Speed | PageSpeed Insights, Core Web Vitals report | Key pages slow for real visitors |
| 7. Rank the fixes | The spreadsheet | One list, worst damage on key pages first |
Speed comes last on purpose. Google says it "always seeks to show the most relevant content, even if the page experience is sub-par" (Google Search Central: Understanding page experience). A page Google can't index loses everything; a slow page loses a little.
Before you crawl: write down the pages that matter
The key-page list decides what's urgent. A broken link on a 2019 blog post and a broken "Book now" page are the same warning in a crawler, but not the same problem. Next to each key page, note its clicks for the last three months from Search Console's Performance report. That's your baseline to compare against after fixes. If you haven't set up Search Console, start with Google Search Console for a small business.
Step 2: Crawl your website
A crawler follows every link on your site, as Googlebot does, and lists each address it reaches. Screaming Frog's SEO Spider is the common choice:
"The Screaming Frog SEO Spider is free to download and use for crawling up to 500 URLs at a time." (Screaming Frog SEO Spider user guide: General)
Most service-business sites fit. Paste your homepage, click Start, and export the results to a spreadsheet when it finishes. The guide notes a £199-a-year licence removes the limit and unlocks settings; a small site rarely needs it.
A crawl shows only pages that something links to. Google puts it plainly: "Every page you care about should have a link from at least one other page on your site" (Google Search Central: Link best practices). A key page that's missing from your crawl is an orphan: nothing links to it. That's a finding in itself.
Step 3: Read the crawl: errors, blocks, redirects, duplicates
Filter the export by status code first, then check the indexing columns. What each finding means:
| Finding | What it means for Google | Fix |
|---|---|---|
| 5xx (server error) | Google slows crawling; indexed pages are kept for a while, "but eventually dropped" | Your host or developer, now |
| 404 on a page that should exist | Google drops the address from its index | Restore the page or 301-redirect it to its replacement |
| 404 on a page you meant to delete, no replacement | Normal; Google simply forgets it | Remove internal links pointing to it |
| Page looks empty or like an error but returns 200 | Google may flag a "soft 404" | Return a real 404, or add the missing content |
| noindex on a key page | Google won't show it | Remove the tag (often left over from a staging site) |
| Key page blocked in robots.txt | Google can't crawl it, and can't see a noindex on it either | Remove the Disallow rule |
| Redirect chain (A to B to C) | Google follows up to 10 hops but advises "redirecting to the final destination directly" | Point A straight at C; update internal links |
| Internal links to redirects or 404s | Wasted clicks for visitors and crawlers | Change the link to the final address |
| Same page at http/https or www/non-www | Duplicates competing | 301-redirect all versions to one |
| Canonical points to the wrong page | Google may index the page you didn't want | Set the canonical to the page itself |
The 404 and 5xx rows come from Google Search Central: How HTTP status codes and network errors affect Google Search. The redirect row comes from Site moves with URL changes.
Two rules trip owners up. First, robots.txt doesn't hide pages: Google says it's "not a mechanism for keeping a web page out of Google" and points to noindex or a password instead (Introduction to robots.txt). Second, the two clash:
"If the page is blocked by a robots.txt file or the crawler can't access the page, the crawler will never see the noindex rule" (Google Search Central: Block search indexing with noindex)
A canonical is a tag that names the main version of a page when several addresses show the same content. Google ranks a redirect and a canonical tag as strong signals and a sitemap as "a weak signal" (How to specify a canonical URL). Duplicates aren't a penalty: "Having duplicate content on your site is not a violation of our spam policies" (SEO Starter Guide). Fix them so the right address ranks, not out of fear.
Step 4: Compare the crawl with what Google indexed
The crawl shows what exists. Search Console shows what Google kept. Check each key page with the URL Inspection tool; Google's starter guide says "To check how Google sees your page, use the URL Inspection Tool in Search Console."
Don't chase a fully indexed site. The Page indexing report's help page says:
"Don't expect every URL on your site to be indexed." (Search Console Help: Page indexing report)
It adds that the goal is "the canonical version of every important page indexed", and that sites with fewer than 500 pages probably don't need the report at all. For a small site, inspecting the key pages one by one is enough. If a key page shows "Crawled - currently not indexed", see what that status means.
Sitemaps: an XML sitemap is a list of your page addresses made for search engines. Google says you might not need one if your site has "about 500 pages or fewer" and is well linked (Learn about sitemaps). If you have one, it should list only live, canonical pages. Any sitemap address your crawl flagged as a 404, redirect or noindex is a cleanup item.
Step 5: Check phones and HTTPS
Google ranks the phone version of your site:
"Google uses the mobile version of a site's content, crawled with the smartphone agent, for indexing and ranking." (Google Search Central: Mobile-first indexing best practices)
So text, links or reviews that your theme hides on phones may not count. Compare a key page on your phone with your laptop. Google's Mobile-Friendly Test is gone: Google retired it, with Search Console's Mobile Usability report, on December 1, 2023, pointing to Lighthouse instead (Google Search Central Blog, April 2023). Lighthouse is built into Chrome: open the page, open DevTools, choose the Lighthouse tab and run it (Chrome for Developers: Introduction to Lighthouse).
For security, Google's self-check asks "Are your pages served in a secure fashion?" and points to Search Console's HTTPS report (Understanding page experience). Any key page that loads over plain http goes on the list.
Step 6: Check speed on key pages only
Google says "Core Web Vitals are used by our ranking systems" (Understanding page experience). Core Web Vitals are Google's three measures of loading, response to taps, and layout jumps, taken from real visitors. Check your key pages in PageSpeed Insights, and treat a failing page that gets calls as worth fixing. The thresholds and what usually slows a small site are in Core Web Vitals: does speed affect ranking?.
Turn the audit into a prioritized fix list
Score each finding on two things: how bad it is, and whether it touches a key page. Then work top to bottom.
| Priority | Type of problem | Examples |
|---|---|---|
| 1. Today | Blocks Google from a key page | noindex, robots.txt block, 5xx, 404 or password on a key page |
| 2. This week | Sends visitors or Google to the wrong place | Redirect chains, links to 404s, wrong canonicals, http and www duplicates |
| 3. This month | Hurts people using the page | Content hidden on phones, http pages, failing Core Web Vitals on key pages |
| 4. When convenient | Cosmetic, or on pages nobody visits | Long titles, missing alt text on decorative images, warnings on old tag pages |
What to skip: a crawler's overall "health score" is that vendor's own scale, not Google's. Crawl budget is a non-issue for most service sites; Google's guide on it is for sites with 10,000+ fast-changing pages or 1 million+ pages (Crawl budget management for large sites).
After you fix, inspect the key pages again in URL Inspection and compare clicks to your baseline. Google warns that "Some changes might take effect in a few hours, others could take several months" (SEO Starter Guide).
Example
These scenarios are illustrative.
A 40-page plumbing site after a redesign
The crawl finds the old /services/drain-cleaning/ address returns 404, and the new drain page isn't linked from the home page. Clicks to drain cleaning fell from 120 a month to 30. Priority 1: 301-redirect the old address. Priority 2: link the new page from the home page. The 60 "missing alt text" warnings wait.
A dental clinic with a staging leftover
URL Inspection shows the new implants page "Excluded by 'noindex' tag". The developer left the tag from the test site. Removing it is a five-minute fix and the only Priority 1 item; everything else is Priority 3 or 4.
An accountant with 450 blog tags
The crawler reports 900 duplicate meta descriptions, all on tag archive pages that get no clicks. They're Priority 4. The real finding is one redirect chain on the tax-return service page, which goes three hops; pointing it straight at the final address is Priority 2.
FAQ
How long does a technical SEO audit take for a small site?
For a site under 100 pages, plan an afternoon: the crawl takes minutes, reading it and checking key pages in Search Console takes most of the time.
How often should I run a technical SEO audit?
Google sets no schedule. Run one after any redesign, platform move or domain change, and when key pages lose clicks; otherwise a monthly look at Search Console catches most new problems.
Do I need a paid tool?
Not for most service businesses. Search Console, PageSpeed Insights, Lighthouse and the free crawler cover a site under 500 addresses.
Should I fix every warning the crawler shows?
No. Fix anything that blocks or misdirects a key page. Many warnings are on pages that get no visits and change nothing.
Is a technical audit the same as a full SEO audit?
No. A technical audit checks whether Google can reach and store your pages. It doesn't judge whether your content matches what customers search, your reviews, or your links from other sites.
Different for you?
- Traffic fell right after a new site went live: start with redirects. See traffic dropped after a website redesign.
- Your site doesn't appear on Google at all: run the shorter checks in website not showing up on Google first.
- A big drop with no site changes: it may be an algorithm change, not a technical fault. See traffic dropped after a core update.
- You'd rather have the audit and fixes done: see /seo/.