Your website rated with one number
Paste an address and the test returns a 0–100 score built from six areas: speed, SEO, AEO, GEO, technical health and accessibility. The number is not an estimate — it is a weighted average of more than fifty concrete checks, and every one of them shows how it went and why.
The score exists so things can be compared: a site before a change and after it, your site and a competitor, one category against another. It is not the goal in itself — the findings underneath it are.
- overall score
- 0–100
- checks in the report
- 50+
- cap on a blocking finding
- 35
A weighted average of the seven report categories.
Each with a status, a weight and its own recommendation.
noindex, a crawl ban, missing HTTPS or a server error.
Steps
How to get your website rated
The test runs on a public address; nothing to install, nothing to sign in to.
- 01
Enter the site address
The domain is enough. The test works on any publicly reachable page, competitors included.
- 02
Let the test finish
Usually a minute or two. The longest part is the speed measurement at Google, which runs for mobile and desktop separately.
- 03
Read the score and its confidence
The score states how many categories could be measured. An unmeasured category is left out of the average.
- 04
Work the findings by weight
The report orders categories and checks so that whatever moves the score most sits at the top.
How the score is calculated
Every check has a status and a weight reflecting its severity. A category score is the weighted average of its checks; the overall score is the weighted average of the categories. Measurable metrics such as LCP or CLS are not converted with hard cut-offs but on a ramp, so even a small improvement shows up in the score.
- SEO — 25%
- Indexability and on-page signals: title, H1, canonical, Open Graph, robots.txt, sitemap, links.
- Speed — 25%
- Core Web Vitals from PageSpeed Insights, measured separately for mobile and desktop.
- GEO — 13%
- llms.txt, AI bot rules, entity profiles, contact and author, content freshness.
- Technical — 13%
- Viewport, favicon, custom 404, HSTS, broken links and other technical health.
- AEO — 12%
- FAQ and HowTo schema, question headings, direct answer, paragraph length.
- Accessibility — 12%
- Landmarks, form labels, accessible names, tab order.
How to read the final number
Above 85 means a site with no serious shortcomings — what remains is detail. Between 60 and 85 you will usually find two or three things worth fixing that measurably help. Below 60 something is systematically wrong, and the report shows what.
Watch out for blocking findings. A noindex directive, a crawl ban in robots.txt, missing HTTPS or an error response mean the site is effectively absent from search, and the score is capped at 35 points. Without the cap such a site could score ninety, because everything else passes.
When a category cannot be measured, a dash appears instead of a number and it drops out of the overall average. The report always states how many measured categories the score rests on.
Comparing with competitors and over time
The test works on any public address, so you can run it on three sites in a row and compare the numbers. The methodology is identical every time, which is what makes a comparison worth anything.
One caveat for tracking over time: the speed part naturally fluctuates between runs, because lab measurements depend on machine load at Google. SEO, AEO, GEO, technical health and accessibility are deterministic by contrast — the same HTML gives the same result.
FAQ
What people ask about this
The same answers you would get from us by email — no marketing hedging.
What is a good website score?
Above 85 the site is in good shape and only details remain. Between 60 and 85 there are usually two or three specific things whose fix shows immediately. Below 60 something is systematically wrong. Treat the number as a signpost to the findings rather than a goal in itself — ninety points with empty content brings nobody in.
Why is my score low when the site looks good?
Most checks concern things you cannot see: robots directives, structured data, accessible names, headers. A beautiful site can carry a noindex, missing alt text and broken internal links. That is exactly why this is measured rather than eyeballed.
Is the rating objective?
It is repeatable and traceable, which is what matters in practice. The category weights are our choice and open to argument, but they are published and every number shows which checks produced it. Nothing is generated by a language model, so two runs over the same HTML give the same result.
Can I save the report or pass it on?
Yes, the report is ready to print to PDF straight from the browser — it has its own print layout with category cards and aligned figures. Nothing is written to a database; the result exists only in your browser.
How many tests can I run?
As many as you like; there is no cap and no account anywhere. The only practical limit is on the speed part, because Google PageSpeed Insights has its own request quota; once it runs out, speed is not measured and the report comes back without it.
Related guides
SEO check
Indexability, title, H1, canonical, Open Graph, robots.txt and sitemap — with a fix for every finding.
Speed test
LCP, INP, CLS and TTFB from PageSpeed Insights, mobile and desktop apart, with every metric explained.
Does AI recommend you
AI bot rules, llms.txt, entity profiles and content shape — everything that decides a citation in an assistant answer.
All that is left is the address.
The audit takes about a minute, nothing is stored and all it needs is a public URL. Feel free to run it on a competitor.