Method & disclaimer
How these numbers are made
No website hands out its analytics. Every traffic estimate you find anywhere, including this one, is inferred from something else. Here is exactly what Site Figures infers from, and where it stops being reliable.
Where the numbers come from
Site Figures ranks a domain using public datasets that record how much real browser traffic sites receive, then converts that position into a traffic figure using published distribution data. A worldwide ranking is cross-checked against national ones, which is what keeps a large national site from being mistaken for a global one.
Where a site publishes its own analytics openly, through privacy-friendly tools such as Plausible, Simple Analytics, GoatCounter or Umami, we read those published figures and show them instead of an estimate, clearly marked as real. This is uncommon, but when it happens the numbers are exact rather than inferred.
Most free traffic tools rank sites by backlinks, which counts how many other sites link to a page rather than how many people visit it. Site Figures does not.
About the link profile
The link figures count how many separate networks and IP addresses link to a domain, not how many individual links exist. That is deliberate. A thousand links from a single server say far less about a site than ten links from ten independent networks, and network counts are much harder to inflate artificially.
There is no free source for a total backlink count, so you will not find one here. Networks and IP addresses are what the public data provides, and they are the more meaningful figure anyway.
Site Figures does not list which sites link to a domain. The public link graph that would make that possible runs to billions of connections and tens of gigabytes per release, which is beyond what this tool is built to handle. For a full backlink list you need a dedicated tool with its own crawler.
What these figures cannot do
- They are not analytics. Treat them as directional. Do not use them for valuations, invoices, reporting or anything with money or a contract attached.
- The very largest sites are the least precise. Public rankings cannot separate the biggest sites on earth from large national ones, so for those the range shown is wide and the true figure tends to sit toward its upper end.
- Browser mix matters. Sites whose visitors use unusual browsers are systematically misjudged. This is accounted for in the range but cannot be removed.
- Small sites fall off the map. Below a certain level of traffic a site is not included in the rankings at all. Then only an upper bound can be given, which is still information, but not a number.
- Visitors are derived, page views are measured. Page views come straight from the ranking. Visitors are calculated from them using an estimated pages-per-session figure for that type of site, so they carry an extra assumption.
- Earnings are illustrative. They are page views divided by a thousand, multiplied by the RPM you choose. Real ad revenue depends on topic, country, ad placement and season, and can differ by an order of magnitude.
Data sources and licences
- Chrome UX Report — monthly site rankings by page loads, CC BY-SA 4.0, via the crux-top-lists and crux-top-lists-country projects.
- Majestic Million — link-based ranking, used only to refine a position inside a known band, CC BY 3.0.
- Tranco — a research ranking combining several independent lists.
- Common Crawl — open web crawl, used for crawl footprint and trend.
- Internet Archive — Wayback CDX API, used for domain age and archive activity.
- Wikimedia — the pageviews API, used for calibration.
- Natural Earth — public domain map geometry, via the world-atlas project.
Privacy
No account is needed and nothing about you is stored. Lookups are cached by domain so that popular queries do not need recomputing; that cache holds the result, not the requester. The RPM you pick is kept in your own browser.
If you accept the cookie notice, Google Analytics records which pages are visited so we can see what the tool is used for. IP addresses are shortened before storage and nothing is shared with advertisers. Decline and no measurement cookies are set at all; the tool works exactly the same either way. Your choice is remembered in your own browser, so clearing your browser data resets it.
Errors and corrections
If a figure looks clearly wrong, it usually means the site sits in an awkward spot: a very large brand in the top band, a site whose audience is outside the covered countries, or one that has recently grown or shrunk faster than the monthly data can follow. The range shown is the honest answer to how much room there is for that.