A while back, I was discussing rebuilding a website with my client. When going over the plan, I proposed a CMS-less system, since I’d be the one administering the site, and suggested instead a flat-file PHP template system. She said that we needed to use WordPress because of its popularity. “It’s popular for a reason“, she insisted.

I won’t lie, I rallied against WordPress for a long time, but the claim of powering almost half the web wasn’t the reason. In fact, I hadn’t heard that until only recently. But I still want to investigate this claim.

Sources

The top sources for this claim include WordPress, W3Techs, and TechRadar, well known for their journalistic integrity – wait, no, this is that listicle site that turned to semi-serious tech talk (and they still do listicles).

Anyways, out the gate, I’m discounting WordPress’ claims that they power 40% of the web. Regardless of their source for that, they’ve got an interest in making sure that they’re relevant. TechRadar’s source is W3Techs, so that’s what I want to look in to further.

Methodology

Reading through the W3Techs article, it’s clear that they aren’t counting *all* websites. They’re only counting the “Top 10 Million”, as defined by Amazon’s Alexa, discontinued in May of 2022. They whittle that list down further by removing sites that only show default pages, which is a huge mistake, in my opinion, because they later go on to say they count websites that have subdomains that use WordPress.

So, my website, for a long time, ran on a flat-file PHP system – not WordPress, but I did have 3 sub-domains running WordPress. I wonder if they would have counted my website? According to this article, yes, but I think that’s dubious – how do they discover a sub-domain? If it’s not linked to from the main site and not in the list, do they still find it?

“The Meaningful Web”

They say they whittle down this list because they want to measure the “meaningful web”, but, and excuse me for being pedantic, the claim “40% of the web uses WordPress” and “40% of the meaningful web uses WordPress” are two different statements.

I wonder – are they counting the deep web on there? Do they count websites that need to be logged in to to use? How about CDNs with a splash page?

I’m not faithful in this claim because of the methodology. Also, 10 million? That’s the number of sites they claim make up the “meaningful web”? I’d argue there’s a lot more than that. In fact, here’s one of them that says there are over 1 Billion! If we take that as fact, then WordPress is used by 40% of the sites that represent just 0.1% of the web – yes, 1/10th of a percent of the web. Even if we’re assuming the methodology is sound, which, again, I’d argue it’s not, that means that WordPress powers 40% of 1/10th of a percent of the web – just 0.04% of the web.

We Don’t Really Know

Of course, I’m not gonna claim that anyone is orders of magnitude off. There’s a middle ground here. Yes, a lot of those 1 billion websites don’t have a thing on them, or are used simply for a gateway or mail exchange. I could get just as critical about that claim – but I don’t think I need to.

All I’m trying to say is I don’t think we can truly know how much of the web is uses WordPress. It’s probably a very significant number – but it’s nowhere near the majority. I’d guess it’s actually somewhere between 1/4th to 1/3rd. Certainly still significant and an achievement. But darn it, I’m not here for good guesses, I’m here for accuracy, and W3Techs’ work isn’t convincing me.

Photo by Nicholas Cappello on Unsplash

No Comments...yet