The animated ranking of internet users from 1990 to 2024 begins in a world nobody would recognise. In 1995 the United States had roughly 25 million people online — more than ten times any other country. The internet was, in any meaningful sense, American.

By 2024 China has about 1,296 million people online and India 942 million, against 322 million in the United States. China passed the US in 2008 and has led every year since; India took second place in 2019.

This is a count, and that is the whole point

The obvious objection is that China and India have far more people, so of course they have more users. That is exactly right, and it is why the chart is worth looking at rather than a reason to dismiss it.

A ranking of internet penetration — the percentage of a population online — looks completely different. It is led by small, wealthy, highly connected countries, most of them in northern Europe and the Gulf. Neither China nor India comes close to the top.

Both charts are correct. They answer different questions. Penetration tells you how thoroughly a country is connected. Absolute count tells you where the users are — and for anyone building a product, deciding which languages to support, or trying to understand which platforms will dominate, the count is the number that matters. The fact that Indonesia, Brazil and Pakistan sit high on one list and nowhere near the top of the other is not a contradiction. It is the difference between depth and scale.

The overtake happened faster than the infrastructure suggests

What makes the Indian curve steep is that it largely skipped the desktop era. Broadband to the home was never going to reach a billion people quickly. Cheap smartphones and, from 2016, extremely cheap mobile data did.

The result is an internet population that is overwhelmingly mobile-first, often mobile-only, frequently arriving on a device with limited storage and an intermittent connection. Anyone whose mental model of "an internet user" was formed in a country where people got online through a computer on a desk is picturing a minority of the current world total.

What the number does not capture

"Internet user" is a looser definition than it appears. It generally means someone who has used the internet in the past few months, by any means. It does not distinguish between a fibre connection and a shared handset with a few hundred megabytes a month. Two countries with similar user counts can have very different levels of actual access.

Nor does it say anything about what is reachable. A user behind extensive filtering is counted the same as one who is not, which means the ranking measures connection but not the same internet.

Those caveats matter for interpretation. They do not undo the headline: the centre of gravity of the online world moved to Asia, it moved within about fifteen years, and the country that invented the thing is now well behind.