SEO audit tool, defined
An SEO audit tool is software that inspects a website for problems that affect how search engines crawl, index and understand it. It discovers your pages by following links, records what each URL returns, checks the result against a rule set (status codes, redirects, titles, canonical tags, indexing directives, links, images, speed) and reports the issues by URL and severity.
- It simulates a search engine crawler; it does not see Google's index or rankings.
- Its findings are hypotheses. Search Console tells you which ones Google has actually run into.
- A health score is the vendor's formula, not a Google signal.
- The value is in the triage: a long report usually holds a handful of issues that matter.
An SEO audit is the job: a review of a site's technical health, on-page setup and content, ending in a list of things to fix. An SEO audit tool does the part of that job a person cannot do by hand, which is checking every URL on a site the same way, every time. A person can review 20 pages in an afternoon. A crawler checks thousands in minutes and catches patterns, such as one template that puts a noindex tag on every blog post, that nobody would spot one page at a time.
What the tool cannot do is decide what matters. It flags a missing meta description with the same confidence as a page accidentally blocked from Google. That judgement is still the auditor's, which is why this guide spends as much time on reading the report as on what produces it.
How an SEO Audit Tool Works
Google Search works in three stages: crawling, indexing and serving. An audit tool imitates the first stage and the start of the second, then applies its own rules where Google would apply its own systems.
- 1
Discover URLs
The tool starts from your homepage (or a list of URLs, or your XML sitemap) and follows every link it finds, the way Googlebot discovers pages. Pages nothing links to are only found if the tool also reads your sitemap.
- 2
Fetch each URL
It requests each page and records the response: status code (200, 301, 404, 500), redirect target, response time, headers, and whether robots.txt allows the crawl.
- 3
Render the page (if switched on)
Many sites build links and content with JavaScript. Google renders pages with a recent version of Chrome, so a good audit tool can render with a headless browser too. Without rendering, it sees only the raw HTML and can miss content and links.
- 4
Extract the SEO elements
Title, meta description, headings, canonical tag, meta robots and X-Robots-Tag, hreflang, structured data, internal and external links, images and their alt text, word counts and page weight.
- 5
Run the rules
Each element is checked against a rule: missing title, duplicate title, canonical pointing to a redirect, noindex on a page in the sitemap, link to a 404, redirect chain longer than one hop, image without alt text. Different tools ship different rule sets.
- 6
Report and prioritise
Issues are grouped by type with the affected URLs, usually labelled by severity, often rolled into a health score. The good tools also show where each problem is linked from, so you can fix the source, not just the symptom.
What an Audit Tool Checks
Rule sets differ by vendor, but nearly every audit tool covers these areas. The exact checks are where the tools compete.
| Area | Typical checks |
|---|---|
| Crawlability | robots.txt blocks, links that are not crawlable <a href> elements, pages too deep in the link structure, orphan URLs found only in the sitemap |
| Status codes and redirects | 4xx and 5xx pages, internal links to redirects, redirect chains and loops |
| Indexing signals | noindex tags and headers, canonical tags pointing to redirected, broken or noindexed URLs, sitemap URLs that are not indexable |
| On-page elements | missing, duplicate or very long titles and meta descriptions, missing H1, image alt text |
| Duplicate content | exact and near-duplicate pages, URL parameters creating copies, HTTP and HTTPS or www and non-www versions both live |
| International | hreflang tags without return links, wrong language codes |
| Performance | page weight, slow responses, and Lighthouse lab metrics such as Largest Contentful Paint and Cumulative Layout Shift when the tool connects to PageSpeed Insights |
| Structured data | markup that fails to parse or is missing required properties |
For hands-on fixes, see our guides to finding and fixing broken links, duplicate content and noindex tags.
What an Audit Tool Cannot See
A crawler sees your site. It does not see Google. Google is blunt about this in its guidance on third-party tools: they do not have access to Google's internal ranking data and cannot guarantee performance. Each of the questions below needs a different source.
| Question | Can an audit crawler answer it? | Where the real answer is |
|---|---|---|
| Is this page actually indexed? | No. It can only see whether the page looks indexable. | Search Console Page indexing report, URL Inspection |
| Which canonical did Google choose? | Sees only the canonical you declared. | URL Inspection: Google-selected canonical |
| Does Googlebot get the same page? | Crawls as its own bot, from its own location. | URL Inspection live test |
| Are real users getting a fast page? | Lab data at best: one simulated load on one device. | PageSpeed Insights field data and the Core Web Vitals report |
| Which pages earn clicks, and which lost them? | No. | Search Console Performance report |
| Did the visitors convert? | No. | Google Analytics 4 |
| Why does a competitor outrank me? | No. Third-party tools have no access to Google's ranking data. | Your own judgement, informed by the pages that rank |
Speed is the subtle one. PageSpeed Insights separates lab data (one simulated page load on one device and network, useful for debugging) from field data (what real Chrome users experienced). Most audit tools report lab numbers. Google's Core Web Vitals assessment uses field data for Largest Contentful Paint, Interaction to Next Paint and Cumulative Layout Shift, so a page can score well in a crawler and still fail for real users, or the other way round.
The Main Kinds of Audit Tool
They differ in where the crawl runs, how deep it goes and how often it repeats. Most teams combine one crawler with the search engines' own tools.
Desktop crawlers
Screaming Frog SEO Spider and Sitebulb's desktop app run on your computer and give the deepest view of a site's technical state. Screaming Frog's free version crawls up to 500 URLs per crawl; JavaScript rendering and saved crawls are paid features.
Cloud site audits inside SEO suites
Semrush, Ahrefs, Moz and similar suites crawl from their servers on a schedule and track issues over time next to rank and backlink data. Less configurable than a desktop crawler, easier to run every week.
Single-page checkers
Lighthouse, built into Chrome DevTools and run by PageSpeed Insights, audits one page at a time. Its SEO category checks basics such as a title, meta description, successful status code, indexability, crawlable links with descriptive text, valid robots.txt, image alt text, hreflang and canonical tags.
Search engines' own tools
Google Search Console and Bing Webmaster Tools report what the search engine itself found. Bing Webmaster Tools also includes Site Scan, an on-demand crawl that checks a site for common technical issues.
For a product-by-product comparison with features and trade-offs, see the best SEO audit tools. To run an audit with free tools only, follow the DIY technical SEO audit.
Reading an Audit Report: A Worked Example
A software company crawls its 1,400-URL site after a redesign. The tool reports more than 1,500 issues and a health score in the sixties. Here is how the report breaks down, and what each line is actually worth.
Illustrative example, not real data
| Issue flagged | URLs | Priority | Why |
|---|---|---|---|
| Blog posts carry a noindex tag after a template change | 180 | Critical | Confirm in the Page indexing report under Excluded by noindex tag. If the posts should rank, fix the template today. |
| Internal links to pages returning 404 | 41 | High | Visitors and crawlers hit dead ends. Fix the links at the source pages the tool lists, or redirect URLs that have backlinks. |
| Canonical tag points to a redirected URL | 14 | Medium | Sends a confused signal. Point the canonical at the final 200 URL. |
| Internal links pass through redirect chains | 96 | Medium | Wastes crawl and slows users. Update links to the final URL during normal maintenance. |
| Duplicate title tags | 120 | Check | Mostly paginated archive pages. Fix only on pages you want to rank. |
| Images missing alt text | 540 | Low | Matters for accessibility and image search. Fix on key templates first. |
| Missing meta description | 312 | Low | Google primarily builds snippets from the page content anyway. Write them for important pages. |
| Low word count | 220 | Ignore | Tag pages. Google says it has no preferred word count; decide whether these pages should exist, not how long they are. |
Four lines out of more than 1,500 issues carry nearly all the risk, and the biggest one by count (alt text) is not among them. The noindex template alone could take the blog out of Google as the posts are recrawled, and it is the kind of change that ships in a redesign without anyone noticing.
Before fixing anything, the auditor checks Google's side. The Page indexing report shows whether the noindexed posts have started dropping out. The URL Inspection tool on two sample posts shows the noindex Google saw and when. The Performance report, filtered to the blog folder, shows whether clicks have moved yet. That turns a crawler warning into a confirmed problem with a measured cost, which is what gets it fixed first.
After the fix, the auditor re-crawls the affected URLs, requests indexing for the most important posts, and watches the Page indexing report until the excluded count falls. The low-priority lines go into normal maintenance, and the health score climbs as a side effect.
One-Off Audits vs. Ongoing Monitoring
A one-off audit is a snapshot. It is the right tool before and after a migration, a redesign or a platform change, when you need to compare the whole site against a known baseline.
Between those moments, sites change with every release and every new page, and the problems that matter most (an accidental noindex, a broken template, a redirect loop) can appear on any Tuesday. Two habits cover that gap. Schedule a recurring crawl, monthly or weekly depending on how often you ship. And watch Google's side continuously: the Page indexing report and organic clicks show the problems Google has actually run into, often before a scheduled crawl would.
Murkuz covers Google's side of the audit
Murkuz is not a crawler. It connects your Search Console, Analytics 4 and Bing Webmaster data to Claude or ChatGPT, so you can check a crawler's findings against what Google actually did. Inspect_URL returns Google's index status and chosen canonical for a URL, Live_Traffic_Drops shows which pages and queries lost ground, and Landing_Page_Diagnosis looks at one page end to end. It reads your data and never changes your site. See the Search Console connector.
Common Audit Mistakes
Chasing a perfect health score
The score is the vendor's formula. Google does not see it. Ten minutes on a critical indexing issue beats a day of clearing low-priority warnings.
Treating every flag as a problem
Rules are generic. A missing meta description on a tag archive or a long title on a product page may not matter at all. Judge each issue by the pages it affects.
Auditing without rendering a JavaScript site
If your links or content load through JavaScript, a raw-HTML crawl reports a site that does not exist. Switch rendering on, or compare raw and rendered HTML on a sample.
Never checking Google's side
A crawler shows what is on your site. Only Search Console shows what Google indexed and how it ranked. Confirm the big findings there before you prioritise them.
Running one audit a year
Sites change with every release and every new post. Crawl after major changes and on a schedule, and keep an eye on Search Console in between.
Further reading
The most useful material on how audit tools work and where they fall short, starting with Google's own documentation. Each one was read in full before it was added here.
Official docsGoogle Search Central
In-depth guide to how Google Search worksCrawling, indexing and serving, including how Google renders pages with a recent version of Chrome. An audit tool is a simulation of the first two stages, so this is the model to hold it against.
Official docsGoogle Search Central
Google Search's guidance on third-party SEO tools, services and adviceGoogle's own view of what third-party tools can and cannot tell you, including that they have no access to Google's ranking data. Short, and the best antidote to chasing a tool's score.
Official docsSearch Console Help
Page indexing reportWhat Google actually indexed and the reasons it did not index the rest. This is the report that confirms or dismisses what a crawler flags.
Official docsSearch Console Help
URL Inspection toolHow to see Google's indexed version of one URL, the canonical Google selected, and a live test of the page. Use it to check any single issue an audit tool raises.
Official docsGoogle Search Central
Understand JavaScript SEO basicsHow Google processes JavaScript pages in crawl, render and index phases. Explains why an audit that does not render JavaScript can miss links and content on modern sites.
TutorialScreaming Frog
How To Crawl JavaScript WebsitesA practical walkthrough of crawling with rendering switched on and comparing raw and rendered HTML, from the makers of the most widely used desktop crawler.
Frequently Asked Questions
What is an SEO audit tool?
An SEO audit tool is software that checks a website against search engine best practices and lists what is broken or missing. Most crawl the site the way a search engine bot does, following links from page to page, recording what each URL returns and running a set of rules against every page: status codes, redirects, titles, canonical tags, indexing directives, internal links, images and speed. The result is a list of issues by URL, usually sorted by severity.
What are SEO audit tools used for?
They find technical problems that stop pages from being crawled, indexed or understood: broken links, redirect chains, accidental noindex tags, missing or duplicate titles, wrong canonical tags and slow pages. Teams run them before and after a migration or redesign, after large releases, and on a regular schedule to catch problems that new content and code introduce.
Is Google Search Console an SEO audit tool?
Partly. It does not crawl your site on demand, but it shows Google's side: which pages are indexed and why others are not, the canonical Google selected for a URL, Core Web Vitals from real users, manual actions and security issues. A crawler shows what is on your site; Search Console shows what Google did with it. Most audits need both.
What is the difference between a site crawler and an SEO audit tool?
A crawler collects data: it fetches every URL it can find and records status codes, headers, links and page elements. An audit tool runs rules on that data and reports issues with a priority. Most audit tools include their own crawler, and desktop crawlers such as Screaming Frog also flag issues, so in practice the two overlap.
Are free SEO audit tools enough?
For a small site, often yes. Search Console, Bing Webmaster Tools with its on-demand Site Scan, PageSpeed Insights and Lighthouse are free, and the free version of Screaming Frog crawls up to 500 URLs per crawl. Larger sites, JavaScript-heavy sites and agencies usually need a paid crawler for more URLs, JavaScript rendering, saved crawls and scheduling.
Does a better audit score improve rankings?
Not by itself. Health scores are each vendor's own formula, not something Google uses. Google states that third-party tools do not have access to its internal ranking data and cannot guarantee performance. Fix the issues that stop important pages from being crawled, indexed or used, and treat the score as a rough progress meter.
How often should you run an SEO audit?
Run a full crawl before and after any migration, redesign or major release, and on a fixed schedule in between: monthly for sites that publish or ship often, quarterly for small sites that rarely change. Between crawls, watch the Page indexing report and organic traffic in Search Console, because they show the problems Google has actually noticed.