Methodology

1 Why RankProof exists #

RankProof helps new developers, people who build and run a site on their own, freelancers and small teams make websites that are correct: accessible to everyone, easy for search engines and AI assistants to read, fast and secure. The checks that larger companies pay for should not depend on a budget, so every tool is free, needs no account and explains what it checks, why it matters and how to fix what it finds.

2 Who runs RankProof #

RankProof is run by a sole proprietorship registered in Greece. Its legal name, registration numbers and registered office are in the “Provider details” section of the terms. You can reach us through our contact form; there are no sales calls, newsletters or accounts behind it.

3 Why it is free #

RankProof has no accounts, plans or paywall. It is funded by display advertising, and ads never change a tool's result or what a guide recommends. To keep the tools free, each tool has a daily limit per visitor. Audit reports are deleted after 7 days and technical logs after 30 days.

4 How the tools measure #

Most of the 42 tools run in your browser. When you enter a URL, our server fetches the page once, because browsers cannot read other sites directly, and your browser analyses the HTML the site returned, without running its scripts. Each tool page says what that tool checks and, under "What it does not do", what it leaves out.

A few tools run on our servers. The website audit loads the page in a real browser, so content built by scripts is included. The page speed test asks Google's PageSpeed Insights for a Lighthouse run and for real-visitor data where Chrome has enough of it. The keyword tools combine public signals, such as video results, encyclopedia page views, the domains in today's top 10 and a public link-authority score; they show an order of magnitude, never an invented monthly search count.

5 Where AI is used #

AI models do three jobs here: they write the suggested titles, summary and opening lines of the content brief, they classify the intent behind a keyword, and they compare the meaning of pages in the content gap tool. A model never produces a score that is shown as a measurement: scores and figures come from data.

6 How the monthly studies work #

Each month we measure a sample of popular websites in every EU country, drawn from the public Chrome UX Report list. Our crawler names itself and obeys robots.txt. We publish totals and shares only, never the name of a site, and each study page describes its sample, method and limits. The studies are listed on the “Research” page.

7 Sources we cite #

For accessibility we work from the official texts: the European Accessibility Act on EUR-Lex, the EN 301 549 standard and the W3C's WCAG 2.1. For search, crawling and structured data we cite Google's and other operators' own documentation. Guides and tool pages link the source of each requirement, so you can check it yourself.

8 How the texts are written and translated #

The substance of every page comes from our own measurements and from the primary sources it links: requirements, thresholds and figures are taken from those, not from a model. AI helps us draft parts of the guides and tool descriptions and translate them from English into Greek, German, French, Spanish and Italian. The operator reviews every text before it is published.

9 How to report an error #

If a tool gives a wrong result, or a guide or a translation contains a mistake, tell us through the contact form. We check every report, fix confirmed errors and update the date on the page.

10 Updates and dates #

Tool pages and guides show the date of their last real change, and we never change a date without changing the content. The studies run every month, and each one shows the month it measured.