Why we count our own visitors, and what we do not collect

Every company website wants to know the same things: how many people came, which pages they read, and where they came from. Most sites answer that by adding a third-party script that follows visitors across the web. We chose a smaller, plainer approach: the site counts its own visits, and we decided in advance what we would not collect.

What we count

For each page a visitor opens, the site records:

• The page and the time it was opened
• How long the page was actually in use, counted only while the tab is visible
• Where the visitor came from, such as a search engine, another website, or typing the address directly
• Device, browser and operating system, in broad categories like "mobile" or "Chrome"
• Country, state and city, approximate
• Clicks on links that matter to us: opening CirruSoul Arena, going to the contact page, e-mail, phone and WhatsApp links, and links to other sites

A visit ends after 30 minutes without activity. From these records we can say how many people visited, how many were new, how many reached the contact page, and which blog posts people actually read to the end.

What we do not collect

• No IP addresses are stored. The address is used once, at the moment of the visit, to look up an approximate location, and is then discarded.
• No names, e-mails or accounts. A visitor is a random identifier kept in their own browser. Clear the browser's data and you are a new, anonymous visitor.
• No third-party scripts or advertising trackers. Nothing is sent to an analytics company, so nothing can be combined with data from other sites.
• No automated traffic. Search-engine crawlers, link previews and monitors are filtered out so they do not inflate the numbers.

Because there are no tracking cookies, there is nothing to ask permission for, and no banner to click away.

How the numbers reach us

Every night just after midnight India time, the previous day is closed and a summary is e-mailed to us: visitors, page views, the most visited pages, where visits started, which links were clicked, and the countries, states, cities, devices and sources behind them. The daily totals are kept permanently. The individual page views behind them are deleted after about 13 months.

About location

Location comes from a free IP-to-location database from DB-IP, which is accurate to the country most of the time and to the city only roughly. It is good enough to tell us whether readers are mostly in Uttar Pradesh or in the United States, and no more precise than that. We treat it as a hint, not a fact.

Why bother building it

Off-the-shelf tools would have taken less time. But we wanted a measurement we fully understood and could explain in one page, with no data leaving our own server. It also gave us a small, real project to build the way we build things for clients: simple, owned by the customer, and clear about its limits.

Want measurement like this for your own product, with data that stays yours? Get in touch.