WordPress market share 2026 reads 41.2%. That is the W3Techs figure for July, and it is the first sustained decline the project has posted in a decade. The same dataset puts WordPress at 59.1% of the CMS market. The peak, near 43.5%, came in early 2025. Every month since has given ground back.
This is not a one-month wobble. Growth had already levelled off from 2022, per the WPZOOM statistics roundup, so the current run is a change in direction rather than a change in pace. For agencies and platform teams, that distinction matters more than the headline number.
What the WordPress market share 2026 number actually measures
W3Techs does not crawl the web. It samples the top 10 million sites by traffic, then extrapolates. That methodology is consistent, which is what makes the trend line usable, and it is narrow, which is what makes the absolute number easy to misread.
Three distortions are worth naming.
- Headless installs. A site running WordPress as a headless CMS can serve a front end that looks nothing like WordPress. Detection depends on what the crawler can see. In practice, some of these sites are counted and some are not, and the split is not stable across years.
- Spam and zombie sites. Dead installs, abandoned subdomains and auto-generated spam all sit in the sample. They inflate the denominator and the numerator unevenly, because spam farms churn faster than real sites retire.
- Sample composition. The top 10 million sites is not the web. It is the web that gets traffic. A CMS can gain users in low-traffic segments and still lose share in this sample.
None of that makes the decline fake. It makes the decline a statement about a specific population, measured a specific way, on a specific schedule.
Share and installs are not the same number
Share is a ratio. Installs are a count. They can move in opposite directions, and often do.
If the total number of websites grows faster than WordPress installs, share falls even when installs rise. If a large block of low-quality sites is deindexed or goes dark, share can rise even when nobody launched anything. The W3Techs series tracks the ratio. It does not tell you how many WordPress sites exist, and it never has.
| Metric | July 2026 | What it answers |
|---|---|---|
| Share of all websites | 41.2% | How common WordPress is in the sample |
| Share of CMS market | 59.1% | How dominant WordPress is among detected CMSs |
| Peak share | ~43.5% (early 2025) | Where the ratio last topped out |
| Install count | Not published here | Nothing. This dataset does not report it. |
The CMS-market figure is the more useful one for competitive analysis, because it holds the denominator closer to constant. At 59.1%, WordPress is still the default answer to “what should we build this on” for most of the market. The 41.2% figure is the one that gets quoted in board decks, and it is the one most likely to be quoted without the methodology attached.
The release cycle did not stop
WordPress 7.0 shipped in May 2026 with more than 420 enhancements and fixes from over 875 contributors, according to the July 2026 news roundup. That is a healthy release by any historical measure. Contributor counts at that level are not what a project in retreat looks like.
So the honest read is this: the software is shipping, the share is sliding, and the two facts are not in direct contradiction. A mature platform can improve on every axis its maintainers control and still lose ratio points to a growing denominator, a changing sample, or both.
What we conclude, and what we do not
We conclude that the 41.2% figure is accurate for what it measures and that the trend is real within that measurement. Ten months of decline is a trend, not noise.
We do not conclude anything about why. The sources here document the number and the release activity. They do not attribute the slide to a competitor, a pricing change, an AI builder wave or a hosting shift. Anyone telling you the cause with confidence is adding a story to a dataset. Apex Digital’s 2026 review frames it as a maturing market; that is a framing, not a finding.
For teams making decisions on this, the practical advice is unchanged. Watch the ratio quarterly, not monthly. Watch the CMS-market figure alongside it. And when a vendor quotes 41.2% at you, ask which sample it came from.
FAQ
Is WordPress actually losing users in 2026?
The W3Techs share figure is falling. That is a ratio, not a headcount. This dataset does not publish install counts, so it cannot tell you whether the absolute number of WordPress sites is up or down.
Why does W3Techs report 41.2% when other sources say something different?
Different samples. W3Techs works from the top 10 million sites by traffic and extrapolates. Any source using a different crawl, a different site list or a different detection method will land on a different number. Compare trends from one source, not levels across several.
Does the decline mean WordPress is in trouble?
It means the ratio is down within one measurement. WordPress 7.0 shipped 420-plus changes from 875-plus contributors in May 2026. A project can ship hard and still lose share points. The two are separate questions.
