I’ve made my living from WordPress for twenty years. I’d love to tell you it’s thriving. I can’t, and the most honest evidence I have is my own behavior.
I still build on WordPress. I still reach for it for the jobs it’s best at. But it’s no longer my automatic first choice for everything, and that’s new. For a blog, a brochure, most small sites, I’ll build on Astro. For anything with dashboards or heavy custom logic, an AI build. For a store, Shopify. I’m biased toward WordPress. I know it better than any tool on earth. And I still keep picking other things, because for those jobs they’re now better.
When the person with every reason to default to a tool stops defaulting to it, that tells you something the headline numbers won’t for a few more years.
The tool of the day changed
Ten years ago you could walk into a cafe in any of the towns where people went to start businesses, and half the laptops had WordPress admin open. People building content sites, blogs, client work. WordPress was the tool of the day.
Walk into the same cafe today and you won’t see a single WordPress dashboard. You’ll see terminals, ChatGPT, and people posting straight to Instagram and X. The builders moved to AI and code. The publishers moved to social. Either way, nobody’s opening WordPress. The leading edge sits in those cafes, and it left.
Decline you can’t see in the numbers yet
People will tell you WordPress isn’t declining, and they’ll point at the stat: still about 42% of the web. They’re right about the number and wrong about what it means.
Installed base is a lagging indicator. Millions of existing sites don’t migrate, so share stays high out of pure inertia long after the story has changed. The leading indicator is the flow of new projects, and that flow is moving away.
WordPress hasn’t lost its install base. It’s lost the default. The share just hasn’t caught up.
The people it was built for left
Remember what WordPress was for. It was built to democratize publishing: give anyone an easy way to set up and put their voice online. For fifteen years that mission had millions of takers. Bloggers, solo publishers, small business owners who wanted a place of their own.
Two things happened to those people.
Google stopped sending them traffic. The Helpful Content updates gutted independent content sites, and AI Overviews are taking the clicks that survived. The independent blog as a business didn’t slow down, it got switched off.
And the people who just wanted a voice went to social media. No hosting, no theme, no plugin to learn. You open an account and post. The original WordPress constituency didn’t abandon the tool. The reasons they needed it disappeared.
AI came for the on-ramp
Then AI came for the part WordPress was built on: the on-ramp. WordPress won the long tail because it was the easiest way to get a site up. The famous five-minute install.
That wedge is gone. Tell an AI builder what you want and you have a site. No install, no theme shopping, no plugin rabbit holes.
There’s a deeper version of this problem too. WordPress’s core is twenty years of procedural PHP, and that architecture is hard for AI tools to read and work with. So at the exact moment building moves to AI, WordPress is one of the harder things to build with it. As Ryan Hellyer put it, the architecture that was always clunky is now actively costing us, because the tools meant to help can’t navigate it.
Social media ate publishing a decade ago. That’s old news. The new news is that AI is eating the thing that made WordPress the default in the first place, and the codebase isn’t ready to ride the shift. It’s the bigger threat, and the one I don’t see WordPress moving fast enough to answer.
What’s left is real but narrower
This isn’t death. Plenty of WordPress will be here in ten years. It’ll just be a smaller, different thing.
Enterprise will keep choosing it for control and auditability, for owning the code under sites that have to satisfy regulators or internal policy. Commerce will keep a serious WooCommerce contingent for the same reasons: people who want to extend a store in ways a closed platform won’t allow. Those are real, and they may even grow.
But notice what they share. They’re the cases where control beats convenience. WordPress stopped winning on convenience, which is exactly what it used to win on.
What happened to the room
I wrote about WordCamp Europe earlier this week, and the same shift shows up there. I’ve watched these conferences for over fifteen years. They used to be a melting pot: people genuinely excited to learn, and a thick layer of owner-operators there to talk shop, promote their plugins, make deals, size up the competition. For a small business owner in the space it was worth every euro.
Walk the hallway today and the independents are mostly gone. They didn’t lose interest. The ecosystem consolidated. The plugin shops and hosts that used to sponsor as scrappy independents got rolled up into a handful of big companies. The enthusiasm you meet now is often an employee’s enthusiasm. Real, but it’s a good job with travel, not a business with everything riding on it. That’s no knock on the people. It’s the predictable result of a roll-up. When the owner-operators get acquired, the room fills with their acquirers’ staff.
It shows in the program, too. As one writer put it after this year’s edition, the schedule has become a closed loop: contributors choosing talks from contributors, built for the WordPress economy and not the people who actually use WordPress.
And try engaging from the bottom. I’ve been building in WordPress more these past months than I have in years, and the experience is its own answer. A few years back, WordPress.org stripped the active install growth charts out of the plugin directory, one of the only signals a plugin developer has to tell whether their work is growing or dying. There’s a request to bring them back, marked high priority, sitting on WordPress’s own tracker. It was opened four years ago. It’s been reopened, poked at, and ignored. No code, no decision, no real answer. Small thing on its own. But it’s the texture of the whole experience: reasonable asks from the people who fill the directory, met with silence from the top.
That’s the pattern the plugin authors and small businesses know well. Leadership spent years not listening to the people who made WordPress valuable, and the friction pushed a lot of them out long before it was visible. The WP Engine saga was the part everyone saw. It was the tip of something that had been grinding for years.
The people who love it are drawing up fork plans
The clearest sign of the moment isn’t coming from WordPress’s critics. It’s coming from the people inside who still care about it. Joost de Valk, who built Yoast into a household name in the space, called for breaking the status quo and ending the project’s one-man rule. Ryan Hellyer wants to revive BackPress as a compatibility layer so WordPress can be rebuilt on a modern foundation. Malcolm Peralty laid out the case for splitting it in two: a frozen WP Classic for the enormous base that already runs it, and a modern WP Next for everyone else.
You don’t write “the case for the split” about something healthy. You write it about something you love that you think is stuck. When the loyalists start drawing up fork plans, that isn’t noise. That’s the symptom.
Said by someone still in it
So that’s my honest read, from someone with every reason to read it the other way. WordPress didn’t fail. The web moved, the audience left, the on-ramp got eaten, and the independents who built the ecosystem got bought or moved on. What’s left is real but narrower, and a lot less interesting to the kind of person who made WordPress what it became.
I’m not writing this as someone who left and is now throwing stones. That’s the easy genre this year, and it’s not me. I still build on WordPress. I still use it for what it’s good at. That’s exactly why this is hard to write. After twenty years, watching the ground shift under a tool you depend on, and still depend on, is a strange place to stand.
It’s also just where I am.
Related
Leave a Reply