WordPress marketshare is "shrinking"

Joost de Valk writes,

There’s no more denying it: if you look at W3Techs CMS market share numbers, WordPress’ market share is shrinking, losing 0.4% market share since February.

This is a sensationalist article based on faulty stats and anecdotes, though I wanted to comment on some of the points.

As the author notes that the main source for these stats is the now-defunct Alexa, and he plans to switch to a different source soon. Those same stats also note that WordPress grew in marketshare by several percentage points before this slight drop.

Additionally, W3Techs mentions that they don’t count subdomains, which leaves out any platform that hosts their user’s site on a subdomain, and large orgs that also adopt the subdomain setup. This leaves out every single site without a custom domain that uses WordPress.com, Blogger, Tumblr, Wix, Webflow, and others.

If you look at cwvtech.report you’ll see that in the last year, sites on Wix and Squarespace on average have improved their site speed more than WordPress sites. WordPress has a performance team now, and it has made some progress. But the reality is that it hasn’t really made big strides yet, and in my opinion, really should.

When comparing WordPress Core Web Vitals (CWV) against silos like Wix and SquareSpace, it’s important to differentiate between self-hosted (WordPress.org) and WordPress.com, which may be difficult to measure. WordPress.com does a lot to improve performance by default, most sites have JetPack features including image optimization, and their server speed is exceptional, with their own page-caching setup. WordPress.org sites do not come with any of these niceties out of the box, and it’s up to the developers or users to take care of optimization. Slow shared hosting and the added bloat of page builder plugins is common with WordPress.org sites, and negatively impacts CWV.

There have been efforts to improve the performance of self-hosted sites, but a major performance improvement that added WebP image conversion by default was recently rejected by the community, whose principle concern was the storage space required by generating another set of image sizes in a different format. Why is storage space a problem? Cheap, and usually slow, shared hosting.

WordPress’ full site editing project is not done yet. Anecdotally, more and more people are having a hard time deciding how to build their site on WordPress. Wix and Squarespace are simply way simpler tools to build a site. As they improve their SEO tooling, there’s less and less reason to switch over to WordPress.

I fully agree with this one. My client projects for the last 6 months were built using Webflow, as I needed rapid turnaround and low maintenance that WordPress just doesn’t provide. I don’t want to use Webflow, as I can’t easily switch later (although full code/assets export and CMS to .csv is available), but building sites with it is much faster, and the client’s experience managing their content is way better. I’ve spent a lot of time, and not a small amount of money, making it easier for me to build and maintain WordPress sites over the years, and yet I still opted for Webflow.

The primary challenge with WordPress for me is hosting, as I have high standards for performance, but I also have to balance performance with cost, both financially and time spent on maintenance. I still haven’t found a WordPress host or hosting setup that is the best balance of both. I recently migrated my sites from Kinsta to WordPress.com, and plan to write about it soon, but it still doesn’t fully meet my needs with pricing (one year up front) and features (no staging). I can’t start a new project and share a staging site with a client while keeping my initial investment minimal.

I may end up looking into VPSs again and using SpinupWP in the future.

I think WordPress, for the first time in a decade, is being out-“innovated”. Now I say “innovated” because Squarespace and Wix are not really doing anything that new. They’re just implementing best practices for both site speed and SEO. They are, however, rolling that out for all their users. So all of their users get better and better page speed performance and improved SEO. As a result more and more of their sites are doing well.

I read this and was surprised, as the CWV report doesn’t show a large difference in performance, except in Webflow’s case.

CWV chart for May 2022

The chart shows the percentage of sites with good CWV, between January 2020 and April 2022:

Why are Wix and Squarespace so close to WordPress, with a fraction of the sites? No matter the sample size, I’d expect a much bigger difference in stats from a fully-managed service like Wix and Squarespace than WordPress’s combined stats across WordPress.com and .org sites. These managed services have a lot of control over their user’s sites and designs, yet they aren’t making any large strides in performance. I would have expected their numbers to be much closer to 80%-90%.

I honestly expected Webflow to be at the bottom of this chart, as building sites with their tool is less abstracted from the actual code, and Wix/SquareSpace use a highly abstracted editor system with blocks and components, so Wix and SquareSpace have even more control over making their user’s sites faster.

Going by these stats, WordPress has 616,771 speedy sites with good CWV, while Webflow, Wix, and Squarespace combined have 97,150 speedy sites, which doesn’t mean anything but I thought it was funny.

If WordPress wants to maintain its market share or better yet, grow it, it’ll have to get its act together. That means it should focus on the performance of these sites across the spectrums of site speed and SEO. The Full Site Editing project is simply taking far too long. That’s causing the rest of the platform to lag behind current web trends. Without a drastic change in this approach I think WordPress will continue to lose market share for the next few years.

I’m all for better performance and SEO out of the box with WordPress. I want WordPress to get to the point where I don’t need to install the author’s bloated Yoast plugin or SEO Framework just to fix WordPress SEO issues. Even WordPress.com isn’t immune to the same issues and yet they tout their SEO on their sales pages. Both WP.com and .org create useless attachment pages for every media file uploaded, which clutters the sitemap and may lower your ranking.

Full Site Editing would be great once polished, and if WP.com offered more features geared towards freelancers and agencies, like staging and monthly billing for hosting, this would be the ideal setup for me over Webflow.