Windows 11: Is it really less popular than Windows XP? (Spoiler: No)

Another day, another shocking press release from an IT company trying to build its name recognition by stirring up controversy.
Written by Ed Bott, Senior Contributing Editor

Here we go again.

A small IT management company, Lansweeper, is trying to boost its name recognition with yet another shocking press release about Windows 11 adoption. The headline is pure catnip for tech editors: "More PCs running Windows XP than Windows 11."

And it worked! Just look at these credulous headlines:


The topline numbers are themselves literally incredible. According to the company's press release, only 1.44% of PCs are running Windows 11 as of April 2022, six months after the release of Microsoft's latest operating system. Another 6.6% are running "End of Life Operating Systems," mostly Windows XP and Windows 7.

That would be a spectacular failure for Microsoft if those numbers were indicative of the actual adoption rates for Windows 11. But a closer look at the report suggests that those numbers don't really prove much of anything.

For starters, where does the data come from? According to a footnote in the press release, the company "scanned 10 million Windows PCs coming from a mixture of Lansweeper (20%, enterprise) and Fing (80%, consumer)."

So, let's unpack that. Lansweeper is an asset management tool designed to identify all "assets" on corporate networks, including servers, routers, workstations, printers, monitors, and associated software. Fing, which was acquired by Lansweeper in 2020, makes a network scanner of its own and says its "Device Recognition and Fingerprinting technologies give visibility over the entire digital environment, from Consumers and Industrial IoT to OT [Operational Technology]."

Neither of those products are exactly the sort of analytics tools that are likely to be found in a typical consumer's home. You need to be a network administrator, or at least play one in your local domain, to use this stuff.

The pretty chart that was included with this press release has a few additional red flags in it. For one thing, 9.15% of the Windows devices sampled are running Windows Server products. None of those would be eligible for an upgrade to Windows 11, nor would it make sense to replace a server OS with a workstation OS.

Also: One in three work PCs "not capable" of running Windows 11. Here are the upgrades they need

According to yet another pretty chart, this one on a company blog post, roughly 9% of the workstations surveyed didn't meet the minimum RAM requirements for Windows 11. Seriously? That number is set at 4 GB, which has been the standard for every consumer PC sold in at least the last seven years. If that number is to be believed, it means at least 1 million of the 10 million workstations they sampled are running with only 2 GB of RAM.

One possible explanation is buried in another confusing paragraph, which mentions in passing that only about two-thirds of virtual machine workstations have enough RAM to meet Windows 11 standards. (Spoiler: You don't have to buy extra hardware to increase the RAM on a virtual machine. You just change a setting in the VM's configuration file.)

There isn't enough data to figure out what any of that means, but it sounds like this sample is skewed far more toward business networks' dedicated industrial IoT apps and OT controllers on physical and virtual workstations -- ones that don't need the kind of hardware found in a consumer or small business PC.

So what percentage of the PC installed base today is actually running Windows 11? Microsoft might know, but they're not saying. The last number I found at the company's press site boasted of "1.4 billion monthly active devices running Windows 10 or Windows 11."

One problem with anyone trying to measure device populations is that Windows 11 and Windows 10 both report themselves with similar version numbers. An up-to-date installation of Windows 11 is version 10.0.22000.613, while the latest Windows 10 is version 10.0.19044.1586. That makes it difficult for web-based analytics frameworks to tell the difference. Indeed, the US government's Data Analytics Program, for example, shows that 93.55% of its traffic over the last 90 days (representing 1.8 billion visits from Windows-based devices) came from devices running Windows version 10; there's no separate figure for Windows 11 traffic.

Also: Windows 11 FAQ: Our upgrade guide and everything else you need to know

This company is not the first to try to boost its reputation by flinging around numbers of questionable provenance. One Windows ad network sent out a press release last month insisting that Windows 11 adoption was up to 19.3%, which is ludicrously high.

The biggest driver of Windows 11 adoption will, of course, be in new PC sales. OEMs have sold more than 150 million new PCs in the six months since Windows 11 was released. Most of the consumer PCs, it's safe to argue, came with Windows 11 preinstalled, while most of the business PCs shipped with Windows 10 (enterprises typically upgrade by replacing hardware, not by allowing employees to download upgrades from Microsoft's servers).

Either way, that's tens of millions of new Windows 11 PCs. 

If history is any guide, those PC replacement cycles will push the installed base of Windows 11 to roughly equal status with Windows 10 in about three years. But you won't learn that from a press release.

Editorial standards