For two decades, WordPress has been the default answer to “how do I make a website?” In 2026, that question has new answers. You can describe a site to Lovable, Bolt, or v0 in plain English and watch it ship in minutes.
Elementor’s Angie generates native WordPress widgets from prompts. Astro downloads have doubled year-over-year. Some developers are loudly declaring the CMS era over.
So is it time to write the obituary?
The honest answer: no — but the headline writers aren’t entirely wrong either.
What the numbers actually say
WordPress peaked in mid-2025 at 43.6% of all websites globally. As of April 2026, that figure sits at 42.5% — a drop of roughly one percentage point, and the first meaningful decline since W3Techs started tracking the platform in 2011.
On a base of nearly 600 million sites, even one point represents millions of properties. The next-closest competitor, Shopify, holds about 5.1%.
WordPress still has roughly nine times the reach of its nearest rival.
So the platform isn’t collapsing. It’s plateauing — and for the first time, slipping. The interesting question is why.
The case for “RIP”
Three forces are pulling people away from WordPress, and AI is the accelerant under all of them.
Vibe coding has eaten the simple-website market. Andrej Karpathy’s 2025 tweet introducing the term “vibe coding” has been viewed over 4.5 million times, and it captured something real. Tools like Lovable, Bolt.new, and v0 by Vercel can take a prompt and produce a deployed full-stack app — frontend, backend, database, auth, hosting — in a single browser tab. For a brochure site, a landing page, or an MVP, the WordPress workflow of “buy hosting, install WordPress, pick a theme, configure plugins, manage updates” looks slow and heavy by comparison.
Performance is becoming a real liability. The average WordPress page loads in around 3.4 seconds, well above Google’s 2.5-second Core Web Vitals threshold. Modern alternatives like Next.js average 0.8 seconds; Webflow lands around 1.4.
When Google’s rankings increasingly reward speed, the gap between a plugin-laden WordPress install and a static Astro site is no longer just an aesthetic preference — it’s an SEO penalty.
Maintenance fatigue is real. Every WordPress site owner knows the cycle: a plugin updates, breaks something, conflicts with another plugin, leaves a security hole. Managed AI builders handle patching invisibly.
As one Joost de Valk quote making the rounds put it bluntly: for the millions of sites that are some pages and a blog, you don’t need a CMS — you need a website.
Add to this the governance turbulence of the past 18 months — the WP Engine lawsuit, the Automattic layoffs, BlackRock marking down its Automattic shares by 63.5% — and you can see why some long-time WordPress professionals are quietly migrating clients to other platforms.
The case for “long live the king”
Now the other side, which is just as honest.
WordPress still powers more of the web than the next nine platforms combined.
WooCommerce processed an estimated $35 billion in gross merchandise volume in 2025 and runs roughly a third of all online stores. The plugin ecosystem — over 60,000 free plugins — solves problems that AI builders haven’t gotten to yet: membership portals, LMS systems, advanced multilingual support, complex booking flows. The community is vast; every error message has a Stack Overflow answer.
And here’s the dirty secret of vibe coding tools: they get you about 70% of the way to a real production app. The remaining 30% — edge cases in auth, payment validation, bespoke business logic, SEO for single-page apps that Google can’t easily crawl — is exactly where non-technical founders hit the wall. A weekend prototype isn’t the same thing as a business that runs for ten years.
WordPress is also adapting rather than dying. Elementor’s Angie operates as an agentic AI inside the WordPress dashboard, executing tasks across the whole site rather than just spitting out code snippets. SeedProd, 10Web, Divi AI, and Hostinger’s AI builder all bring prompt-to-page generation into the WordPress workflow itself. The platform that “AI is killing” is busy absorbing AI as fast as it can.
What’s actually happening: bifurcation, not death
The most accurate read on 2026 is that the market is splitting.
The bottom of the market — five-page brochure sites, simple portfolios, quick MVPs, throwaway campaign pages — is leaving WordPress for AI builders and static-site stacks. These users were always over-served by a full CMS. They never needed plugin architecture, custom post types, or a database. They needed a website. Lovable, Bolt, Durable, Hostinger AI, and Astro-plus-Claude-Code setups serve them better, faster, and cheaper.
The middle and top of the market — content-heavy publishers, e-commerce stores running real inventory, businesses whose non-technical staff need to update the site daily, anything requiring the long tail of WordPress plugins — is staying. And for these users, AI is making WordPress better, not obsolete. Generating a custom widget by describing it in chat is a genuine productivity leap.
This is why the headline number — 42.5% — is misleading in both directions. The decline is real, and it represents a structural shift in who reaches for what tool. But “WordPress is dying” misreads the shape of the change. It’s not dying. It’s losing the bottom of its market while consolidating the middle and top.
So should you say RIP?
If you’re building a single landing page this weekend, yes — skip WordPress and use an AI builder.
You’ll be done before you’d have finished installing a theme.
If you’re building a content business, an online store, or anything a non-technical client needs to maintain over years, no — WordPress with AI tooling on top is still the most pragmatic choice. The plugin ecosystem alone is worth more than any single AI builder currently offers.
The “RIP WordPress” framing is mostly a developer’s hot take dressed up as an industry trend.
The actual story is more interesting: the web is finally getting tools matched to different jobs instead of forcing one CMS to be everything for everyone.
That’s not a death. It’s a maturing market — and for users on every side of it, that’s good news.
