The WordPress market share in 2025 is not 43.4% of all active websites on the internet

Why 43% of the Web Is a Misleading Number (and What’s Really Going On)

Skepticism is healthy when it comes to big statistics. One of the most repeated numbers in our industry is that “WordPress powers 43% of the internet.” It’s a headline used to make WordPress feel like the safe, default choice due to it's perceived market share.

We weren't buying it and that figure doesn’t hold up under scrutiny. With how many people openly dislike WordPress—or abandon it after trying—it’s hard to believe nearly half of the functioning web relies on it. The 43% number may be technically correct in terms of installs, but it doesn’t reflect reality. It blends legitimate business sites with abandoned projects, half-finished experiments, demo installs, staging sites, and hacked domains left to rot.

Take a closer look, and the actual number of real, effective WordPress websites is much smaller. And when you compare that against platforms like Wix, Shopify, Squarespace, or Webflow—which publish transparent, verifiable user counts—the WordPress dominance story starts to look more like a mirage. Automattic (the company behind WordPress.com) often repeats the W3Techs figure in its press releases, blog posts, and marketing. They frame it as validation of WordPress’s global dominance, even though it’s really just a detection stat — not proof of active usage.

Why WordPress Looks Bigger Than It Is

WordPress’s reputation as the internet’s default CMS is partly deserved—it’s been around since 2003, it’s free, and it has an enormous plugin ecosystem. But its inflated dominance comes from structural quirks:

• Bundled hosting installs: Many hosts auto-install WordPress with every new domain purchase. Thousands of those sites are never used.
• Low barrier to entry: It costs nothing to install, so people experiment and abandon.
• Zombie installs: Abandoned but still detectable by crawlers.
• No central reporting: WordPress.org has no way to distinguish between active, effective websites and broken test installs.

Contrast that with Webflow or Squarespace, where each live site represents a paying customer who is using it actively.

How the 43% Number Gets Used for Selling People on WordPress

The “43% of the web” figure doesn’t just live on W3Techs. It’s become a marketing talking point — especially for Automattic, the company behind WordPress.com.

On its own website and in press coverage, WordPress.com regularly boasts that WordPress powers more than 40% of the internet. The problem is, this figure is lifted directly from W3Techs’ detection statistics — which, as explained above, don’t distinguish between active, functioning sites and abandoned installs.

This recycling of a raw detection number creates a perception that nearly half of the functioning web runs on WordPress. In reality, it’s closer to saying, “We can still find WordPress code on nearly half of all websites, whether those sites are alive, broken, or abandoned.”

It’s a subtle but powerful difference. One makes WordPress sound like the beating heart of the internet. The other reveals a much murkier reality.

Examples: Making 43% of the Web More Dynamic with the WordPress Interactivity API

Where the 43% Number Comes From

The famous WordPress market share number for 2025 comes from W3Techs noting that "WordPress is used by 43.4% of all the websites,", a research firm that scans the visible web to detect which technologies power different domains. When their crawler sees WordPress code, it gets logged.

‍

Sounds authoritative—but there are major problems:

• W3Techs counts any detectable install, whether the site is active or not.
• A dormant domain with a WordPress install still counts.
• A staging site still counts.
• An abandoned blog from 2008 still counts.

Because WordPress is open-source and free, bundled by most hosting providers, and often available as a one-click install, the barrier to entry is nearly zero. That means millions of ghost installs inflate the “market share.”

In short: WordPress isn’t truly powering 43% of the functioning web. It’s powering 43% of everything that’s ever been set up with WordPress and left detectable which includes active and inactive websites.

How W3Techs Measures WordPress Usage (and Why It’s Misleading)

The famous “43% of the web runs on WordPress” statistic comes from W3Techs, which has become the de facto reference point for CMS market share. But to understand what that number really means, you have to look at their methodology.

• W3Techs only counts “the relevant web.”
  According to their own documentation, they exclude sites that show only a default server page, contain no useful content, or are pure duplicates. That still leaves millions of domains that are technically live but not actively maintained (W3Techs: About Our Site Information).
• Detection is based on code snippets, not site activity.
  Their system scans a site’s HTML, headers, or scripts for fingerprints of specific technologies. If WordPress code is detected—such as a theme directory, plugin file, or “generator” meta tag—the site is logged as a WordPress website (W3Techs: Content Management Overview).
• There is no distinction between active and abandoned sites.
  W3Techs explicitly states that their numbers reflect the presence of technology, not whether the site is updated, secured, or used in practice. In other words, a broken staging site with leftover WordPress code counts the same as an actively maintained business website (HostingAdvice interview with W3Techs).

This approach is useful for detecting broad technology adoption, but it creates a distorted picture of real-world usage. WordPress benefits disproportionately from this because:

• Hosting companies often bundle WordPress installs with new domains, many of which are never launched.
• Test environments and half-finished experiments are detectable and count toward WordPress’s share.
• Abandoned sites with outdated WordPress code still register in the statistics.

So while the 43% figure is technically correct in terms of raw detection, it’s misleading if you take it as fact that nearly half the internet is actively powered by WordPress. The methodology simply doesn’t separate functioning, active websites from inactive websites.

Source: Straight to the Facts: W3Techs Provides Extensive, Reliable Technology Data Without Bias or Ulterior Motives

What’s Actually Counted: WordPress.org vs WordPress.com

Another layer worth clarifying is what kinds of WordPress sites W3Techs includes in that 43% figure. It’s not just self-hosted WordPress.org installs—it also includes a subset of WordPress.com sites. But here’s the nuance:

• WordPress.org (self-hosted): Every detectable install is counted.
• WordPress.com with a custom domain: Counted as well, since the domain masks the fact that it’s running on WordPress.com.
• WordPress.com on a *.wordpress.com subdomain: Not counted. These free hobby blogs and personal sites don’t make it into the W3Techs dataset.

W3Techs themselves confirm this distinction: WordPress.com subdomains are excluded unless the site is mapped to its own domain. In practice, that means the “43% of the web” figure is built on:

1. Millions of self-hosted installs from WordPress.org.
2. A smaller slice of WordPress.com sites that use custom domains.

“We do count both the self-hosted, open source version of WordPress which can be downloaded at WordPress.org, and we also count WordPress sites hosted at WordPress.com or elsewhere. However, we count the hosted sites only if they are reachable via their own domain (not only as subdomain of wordpress.com)...”
- Source: W3Techs

It’s a subtle detail, but it matters. The stat doesn’t capture every blog on WordPress.com, and yet it still manages to lump in a vast number of installs—many of which are test environments, abandoned projects, or low-quality builds.

And that brings us to the bigger problem: even when you accept this combined methodology, it still doesn’t answer the real question—how many of these sites are actually alive and effective?

Are Staging Sites Counted by W3Techs?

• W3Techs only analyzes publicly available websites.
  Their crawler scans the "relevant web," which they define as sites accessible through their own domain and not just subdomains or server defaults. If a staging site is behind a password, blocked by robots.txt, or hidden from public DNS, it won’t be counted. (W3Techs: About Our Site Information)
• But if a staging site is publicly accessible—even accidentally—it can be counted.
  Many WordPress developers spin up staging environments at staging.domain.com or similar. If those subdomains are not password-protected and contain detectable WordPress code, they would qualify for W3Techs’ dataset.
• Robots.txt doesn’t prevent detection.
  W3Techs doesn’t rely on search engine indexing. They run their own scans. So even if staging sites are blocked from Google, they could still be detected if left public.

Why This Matters

This means that while properly hidden staging sites don’t inflate the 43% number, poorly secured ones do. And given how often staging installs are left exposed (with demo content or default themes), it’s safe to assume they make up part of WordPress’s inflated market share.

‍

The Reality of Inactive Websites

The web is littered with inactive or unfinished projects. Globally, there are about 1.12 billion websites, but only ~193.89 million (17%) are active—meaning updated and in use. The rest are parked, abandoned, or broken.

Source: How Many Websites Are There? (2025 Stats)

That ratio alone should make you question WordPress’s 43% headline. If WordPress has a 43% share of all websites, but 84% of the web is inactive, then only a fraction of those WordPress installs are truly doing anything meaningful.

Now add to that the DIY abandonment factor. Many DIY website projects never reach completion—often stalling due to lack of time, technical obstacles, or design challenges. Multiple agencies and independent developers report seeing numerous abandoned WordPress installs, especially when installations come pre-bundled with hosting (e.g. staging sites, placeholders, or templates never customized). So without hard-data we'll have to assume based on anecdotal evidence widely shared across web development communities that at least 25% - 50% of DIY website projects are abandoned before launch or never completed at all.

WordPress is the biggest victim of this pattern. Because it’s free, people spin up installs, poke around, and then quit when they hit a wall with themes, plugins, or hosting complexity. Every one of those sites still counts in the “43% of the web” stat.

Source: Web Builder Statistics 2025: 1 in 3 Have Built a Site, But Most Struggle to Finish

How do we classify an 'active website'?

To understand why this matters, we need to define what we mean by a “functioning” website. It’s not just whether a site exists—it’s whether it works for its owner and users.

An active website should:

• Be live, updated, and serving visitors.
• Have SSL and security updates applied.
• Provide a clear purpose: business, nonprofit, education, publishing, etc.
• Be usable and navigable (not broken layouts or demo content).

By contrast, the WordPress ecosystem is often filled with random installs.

• Demo content left in place.
• Unusable sites with broken themes and plugins.
• Placeholder projects that were never launched.
• Spam websites
• Hacked websites.
• Staging or demo sites.

Counting those sites as evidence of WordPress’s dominance is like saying every car ever manufactured is still “on the road.” You cannot take the number at face value.

Our Best Estimate of WordPress’s True Number of Active Websites

So what’s the real number? No one can pin it down exactly, but we can make a reasoned estimate by layering what we know.

• Global activity rate: Out of roughly 1.2 billion websites, only about 16–17% are active—sites that are live and maintained. Applying that ratio to WordPress’s 43% share cuts the pool down to roughly 80–90 million WordPress sites that might plausibly be active.
• Ghost installs and abandonment: Hosting companies auto-install WordPress on millions of domains, and countless DIY projects stall out before launch. Even conservatively, at least 30–40% of those installs never become functioning websites. That lowers the count to somewhere in the 50–60 million range.
• Maintenance attrition: WordPress sites demand constant updates and patching. Many are abandoned within a year or two of launch, or left to languish with broken layouts and outdated plugins. If we cut away the sites that fail to meet even basic standards of being live, secure, usable, and purposeful, we’re likely left with around 25–35 million truly functioning WordPress websites worldwide.

That works out to roughly 5–7% of the entire web—a huge number, but nowhere close to the myth of 43%. And this is where the comparison to other platforms gets interesting: while WordPress inflates its presence through detections and ghost installs, platforms like Wix, Squarespace, and Webflow publish transparent subscriber counts that reflect real, verifiable usage.

The Transparency Advantage of Other Platforms

Here’s where the comparison really matters. Platforms like Wix, Squarespace, and Webflow don’t get to pad their numbers with ghost installs. They’re SaaS products, which means every live site is tied to a paying subscription. If you stop paying, the site goes offline—it’s that simple.

• Wix: Over 270 million users, with 6.1 million paying subscribers. If you’re paying, the site is live.
• Squarespace: More than 4 million active paid subscriptions.
• Webflow: Over 3.5 million designers and teams building real projects.

These numbers represent active, functioning websites with owners invested enough to keep them online. There’s no inflation from abandoned experiments, no half-finished installs, no staging sites accidentally counted as production.

And that’s the key difference: Wix’s 6 million paying subscribers are more meaningful than WordPress’s “43% of the web” headline. One is verifiable proof of real usage; the other is a detection statistic that can’t distinguish between a thriving business website and a forgotten test install.

Why This Matters for Businesses

If you’re a business choosing a platform, you should care less about vanity stats and more about effectiveness.

• WordPress’s 43% share is an illusion—it’s not a reflection of active use.
• Platforms with transparent usage data give a clearer picture of actual adoption.
• Modern SaaS and headless CMS platforms often deliver better performance, security, and ROI than a bloated WordPress install.

In other words: don’t pick WordPress just because it’s “popular.” Pick a platform that works for your business.

Our Prediction: The Shrinking WordPress Share

As SaaS platforms and headless CMS options continue to grow, the gap between WordPress’s perceived dominance and its actual footprint will only get wider.

By our best estimate, the true share of functioning WordPress sites is closer to 5–7% of the entire web—far below the “43%” headline. And that slice will keep shrinking over the next decade as Webflow, Wix, Squarespace, and modern headless platforms like Storyblok and Contentful pull more businesses into ecosystems built on transparency, performance, and reliability.

WordPress’s open-source model isn’t disappearing, but its days as the default choice for the web are numbered. The future belongs to platforms that can back up their numbers with real, active users—not inflated detection stats.

How We Arrived at 5 to 7%

Before we close, it’s important to acknowledge that no one can know the exact number of functioning WordPress sites. What we’ve done here is take the best available data—global activity rates, W3Techs methodology, DIY abandonment research, and the realities of WordPress maintenance—and layered them together to create a more realistic picture.

These are estimates, not hard numbers. Our goal isn’t to present a perfect data set, but to encourage a more honest conversation about what WordPress’s “43% of the web” really means. When you strip away inactive installs, unfinished projects, and abandoned sites, the story looks very different.

We invite you to use your own judgment as well. Apply the same discernment you would to any bold statistic: ask how it was calculated, what it really represents, and whether it reflects reality on the ground. That’s the lens we’ve applied here—and we believe it leads to a clearer, more accurate understanding of WordPress’s place in today’s web.

One thing is undeniable: WordPress’s growth over the last two decades has been remarkable. No other CMS has achieved such widespread adoption, and its open-source model has empowered millions of people to publish online who otherwise might never have had the chance.

But impressive growth doesn’t erase the reality that its dominance is overstated, or that many of those installs aren’t truly active, effective websites. The challenge now is less about celebrating raw numbers and more about asking the harder questions: which platforms are delivering functioning, secure, purpose-driven websites in 2025 and beyond?