What this tool actually checks
The scan fetches a bounded number of pages from the site you enter, plus robots.txt and its XML sitemap, then runs six categories of checks over what came back:
- Content quality & originality indicators — thin pages, placeholder text, near-duplicate bodies, and pages that are mostly navigation.
- Trust & essential information — whether About, Contact, Privacy Policy, Terms and Disclaimer exist, carry real content, and are reachable.
- Navigation & usability — broken internal links, empty links, unlabelled form fields and skipped heading levels.
- Technical crawlability & accessibility — HTTPS, titles, descriptions, canonicals,
noindex, heading structure and lang.
- Observable policy-risk indicators — publisher content depth, and whether your privacy policy matches what the site actually loads.
- Performance, mobile & accessibility — measured response time, document size and mobile viewport, with untested items marked as such.
It only follows same-origin links, honours robots.txt, never logs in, never submits a form and never touches a page behind authentication.
How to read the score
The score is a weighted total across the six categories, out of 100. It exists to order your work: fix the high-severity items with the most evidence behind them first.
Every deduction traces back to one documented rule, and every finding shows the evidence it was based on. Four rules keep the number honest:
- A check that did not run cannot cost points.
- A page the checker could not reach is inconclusive, never a defect. Timeouts,
403s and DNS failures are evidence about the connection, not the site.
- Measurements that need a browser or a third-party API are marked not tested rather than guessed.
- Questions needing judgement — ad placement, traffic validity, account status — are flagged manual review and left to you.
This is an independent product heuristic. It is not an official Google score, it does not predict approval, and it does not replace a human review.
What it cannot tell you
Being straight about the limits is part of the point:
- Whether Google has indexed you. The tool checks crawlability. Indexing is a separate decision made inside Google.
- Your AdSense account status. Approval, ownership verification and policy notices live only inside your AdSense account.
- Whether your traffic is valid. Invalid traffic cannot be detected from the outside.
- Visual and behavioural problems. Overlays, accidental-click layouts and colour contrast need a rendered browser. Core Web Vitals are reported as untested unless a real measurement is available.
- Whether content is original. Similarity between pages is reported as an overlap signal with a confidence level, never as a plagiarism verdict.
About the language switch
The entire interface is available in English and Arabic. Switching to Arabic sets the document language, mirrors the layout right-to-left, and keeps URLs, check identifiers and code snippets left-to-right so they remain readable.
Both languages are served from one translation dictionary, and the test suite fails if a key exists in English but not in Arabic — so a half-translated screen cannot ship.
Frequently asked questions
Does a high score guarantee AdSense approval?
No, and no tool outside Google can tell you that. Approval depends on factors this checker cannot see: your account history, the reviewer's judgement, your site's reputation and traffic patterns. What the score does give you is an ordered, evidence-backed list of real problems worth fixing.
Why did a page the checker could not open cost no points?
Because that would be a false accusation. A timeout, a 403 from a bot-blocking firewall, or a DNS failure describes the connection between our server and the site, not a defect of the site. Those checks are recorded as inconclusive and deduct nothing.
Is it safe to run this on any website?
Yes. Requests are restricted to public internet addresses. Loopback, private, link-local and reserved ranges are refused, including the 169.254.169.254 cloud-metadata endpoint, and each redirect destination is re-validated before it is contacted. Request sizes, timeouts, redirect depth, crawl depth and page count are all capped, and scans are rate-limited per address.
How is this different from the SEO Checker on this site?
The SEO Checker looks at search-engine signals on a single page. This tool crawls a bounded set of pages and adds content substance, essential-page discovery, policy-risk indicators and a weighted readiness score across six categories — a different question, and a deeper one.
Why does it only crawl a limited number of pages?
To keep a single audit cheap for our host and considerate to your server. The default is twelve pages, two levels deep, with a hard time budget. Reports are deleted after seven days. It is enough to find the structural problems that matter; it is not a full crawl, and the report says so.
How do I get a PDF of the report?
Use Print / save as PDF. The report has a dedicated print stylesheet: navigation and controls are hidden, every finding is shown expanded, and Arabic right-to-left text renders correctly because your own browser's PDF engine handles it. No third-party PDF service is involved and no page content leaves your browser.