How to use the diagnostic
Enter your homepage address and press Scan website. There is nothing to install and nothing to verify. The scan takes about a minute on most sites, and you can watch each stage as it runs.
- Give it your homepage, not a deep page. The scan starts there and follows your own links and sitemap outward, so starting at the top gives it the best picture of the site.
- Use the address people actually visit. If your site lives at
www, enter thewwwversion. The scan follows redirects and will tell you where it ended up, but starting at the real address avoids reporting your own redirect as a finding. - Read the top priority first. Findings are ordered by how much they cost you, not by the order they were checked.
- Check the evidence line before you change anything. Every finding shows what was actually read from your site. If the evidence does not match what you see, tell me, because that is a bug worth fixing.
- Save the report with the Save as PDF button if you want to hand it to a developer. The saved copy includes the guidance panels, even the ones you left collapsed on screen.
How to read the results
Findings are grouped into four states, using the same idea Search Console uses. The count cards at the top tell you how many of each you have.
Critical
Something is stopping search engines from reaching or indexing your pages. This is the only state worth dropping other work for. A blocked robots.txt file, a noindex tag on your homepage, or a server that will not answer all sit here. A site with a critical finding usually cannot rank at all until it is fixed, which is why they are rare and why the tool is careful about claiming one.
Warning
Something is working against you but not blocking you. A missing sitemap, a title that gets cut off in results, broken internal links. Worth fixing, in your own time, in the order shown.
Passed
The check ran and found nothing wrong. These are listed rather than hidden on purpose, so you can see what was actually inspected instead of guessing at what a clean report covered.
Not enough data
The check could not run, so no conclusion is offered. A request that timed out or was refused by a firewall lands here. This state exists because the alternative is worse: a scan that treats a failed request as proof of a problem will tell you your sitemap is missing when it is sitting there working perfectly.
What each category covers
- Crawl and indexing. Whether the site responds, whether it is served over HTTPS, what robots.txt permits, whether a sitemap exists and can be read, and whether anything on the page tells search engines to stay away. This is the part that decides whether the rest matters.
- On-page. Titles, meta descriptions, headings and canonical tags across the pages sampled. These shape how your pages are understood and how they look in results.
- Links. A capped check of internal links for ones that lead to a missing page.
- Images. Whether images carry alt text, which matters for accessibility first and image search second.
- Structured data. Whether schema markup is present and which types were found.
The rules this tool follows
A free scanner that guesses is worse than no scanner, because you spend real time chasing findings that were never true. These are the rules it works to.
- It samples, and it says so. Up to 25 pages, chosen by following your sitemap and your own internal links, with the pages a business usually cares about ranked first. The scope line at the top of every report tells you exactly how many pages were read.
- Absence is only claimed when it is proven. A missing sitemap or robots.txt is reported only on a clean not-found response. A timeout, a refusal or a server error is reported as not enough data.
- A blocked scan is never reported as a blocked site. If a firewall or CDN refuses the request, you get one honest finding saying so. Search engines are usually allowed through where unknown crawlers are not, and this scan has no way of knowing what Google sees, so it does not pretend to.
- Every finding carries its evidence. The status code, the tag that was read, the pages affected by name. Nothing is asserted without showing what it came from.
- It is polite to your server. Requests are capped and spaced, one domain can be scanned once every ten minutes, and the crawler identifies itself honestly rather than pretending to be a browser.
What it cannot tell you
This is a technical SEO audit, and technical problems are only one reason a site loses traffic. No crawler can judge whether your content answers the question someone typed, whether you are targeting terms you could realistically win, whether a competitor has simply pulled ahead, or what your Search Console data says about queries you already appear for.
So a clean report is genuinely useful information, but it is not the same as good news. If the scan finds nothing and your rankings still are not recovering, that narrows it down: the problem is not on the surface, and a proper website traffic loss audit means reading your own Search Console data alongside the competitive picture. That is what the SEO Audit and 90-day roadmap is for.
Common questions
Do I need to sign up to run it?
No. The scan runs and shows your top finding without an email address. Seeing the full diagnosis asks for one, because that is the point at which this stops being free work with nothing in return.
What happens to my email address?
It is stored with your scan so I can follow up if you ask me to, and it is not shared with anyone or added to a mailing list unless you tick the box. Diagnostic records are deleted after twelve months. The privacy policy covers the detail.
Can I scan a site I do not own?
Technically yes, since it only reads pages that are already public. It is most useful on a site you can actually change.
Why did it only scan some of my pages?
The cap is deliberate. Twenty-five well chosen pages surface the same technical problems as a full crawl on almost every site, without spending an hour of your server’s time to prove it. Problems that only appear on one obscure page are the kind a manual audit is for.
It says my site is blocked, but it loads fine for me.
That means a firewall let you through and refused this scan, which is normal and not necessarily a problem. To find out what Google actually sees, use the URL Inspection tool in Search Console, which reports Google’s own fetch rather than anyone else’s.