# Are you still on WordPress, or did Claude build your site? > Sam Parr asked LinkedIn whether anyone still uses WordPress now that Claude builds sites. 235 people answered. Who they are, where the room splits, what it gets right about themes and plugins, what it gets wrong about WordPress, and what half a million AI tool calls on 1,500 live sites say. Source: https://www.respira.press/blog/are-you-still-on-wordpress --- ![A paper card with the question: are you still on WordPress, or did Claude build your site?](https://www.respira.press/blog/wordpress-2026/wordpress-2026-kv.webp) One question, 235 answers, two days. On 21 October Sam Parr, who built The Hustle and sold it to HubSpot, co-founded the founders' community Hampton and co-hosts the podcast My First Million, [posted on LinkedIn](https://lnkd.in/p/dwcsSZpp) that he is a lifelong WordPress user, that Claude now does everything for him, that the theme and plugin business looks finished, and asked whether anyone still uses WordPress at all. Two days later: 204 reactions, 235 comments. I read all of them. Here is who answered, where the room splits, what it gets right, what it gets wrong, and what the numbers from 1,500 live WordPress sites say. ## Who answered Sam's audience is founders, so the thread is founders: people running one company and one site, with a developer or an agency owner every fifth comment and a few SEOs in between. That matters for reading it. Almost nobody in the thread edits a site they did not build, and almost nobody hands a site to a client who will edit it next year. Those two people, the editor who comes later and the client who owns the thing, are the ones WordPress exists for, and they are mostly absent from the room. ## Where the room splits Three camps, in the order of how loud they were. - **Claude built it, no CMS.** The top-reacted comment is a stack: a code repository, a static site generator, a CDN host. Several people repeated it with their own flavour, Astro or Next.js on Cloudflare or Vercel, a headless content tool where a client needs one. The pitch is speed, a small bill, and no plugin layer to patch. Most of these people run their own site. One of them, a technical SEO, added the caveat the others skipped: many of the sites he sees built this way are invisible to Google and to the AI engines, with broken rendering and no basic technical hygiene, and they all look alike. - **Still WordPress, leaner.** The developers and the agencies. Their line is consistent: AI made WordPress more useful, not less, because it writes the custom block or the forty-line function they used to buy a plugin for. One agency owner went from 25 plugins per site to about six. A plugin author who tracked 50 sites counted 34 active plugins on average and called most of it setup glue that AI now writes. What they keep is the core: logins and roles, a real editor, an update system, twenty years of security work, and a site the owner owns. - **Both: the AI works inside WordPress.** The smallest camp and the most specific one. A marketer connects Claude to WordPress through application passwords and tells it to build pages and optimise content. A founder's parents change text on their site by asking Codex. An agency owner runs the Elementor and WordPress connectors from Claude and Codex and keeps the big multi-user sites on WordPress because moving them is a risk he does not want. A consultant built a skill so Claude writes native blocks instead of raw HTML, for branding consistency and because she is, in her words, a little old-fashioned. One commenter named Respira as the way to get both worlds, which is where I come in, and I will keep that to one paragraph later. ## What the thread gets right **Themes sold layout, and layout is the cheapest thing to generate now.** That sentence appears in the thread almost word for word, and the market agrees. On 1 July 2026 Envato ended its exclusive author programme and moved every ThemeForest seller to a flat 50 percent fee, citing market conditions; one outside count puts items sold across ThemeForest and CodeCanyon down by more than 70 percent between 2018 and 2025 (an estimate from public counters, so hold it loosely). A theme was a layout plus a promise that someone else would keep it working when WordPress and PHP moved. Claude makes the first part nearly free. The second part did not go away; it moved to you. **The plugin market splits in two, and only half of it is in trouble.** The thread sorted this better than most analyst notes. Plugins that sold convenience, a slider, a visual effect, a small feature, lose to a prompt. Plugins that carry years of fixes in places where a mistake costs money, payments, forms, backups, security, memberships, stay, because nobody wants to vibe-code a checkout and then own the patches. A WooCommerce engineer put it plainly: the question looks different when the plugin handles payment failures and inventory sync. ## What the thread gets wrong **"RIP WordPress" came from people who run one site.** The share is slipping, not collapsing: W3Techs has WordPress at 40.0 percent of all websites in October 2026, down from a peak of 43.6 percent in January 2025, and still more than every other content system combined. A founder who replaced a ten-page site with a static build is right about his site and wrong about the web. The person missing from the thread is the one several commenters kept pointing at: who edits the site in six months? A non-technical owner changing hours and prices, a marketing hire adding landing pages, a second vendor taking over. For them the admin is the handoff, and WordPress is the handoff the most people on earth already know. One commenter said it best: most WordPress users were never going to write code, and AI does not change that; their constraint was time and not wanting to maintain anything. **"WordPress and security in one sentence" got a laugh, and the laugh is half right.** Most WordPress risk lives in plugins and themes, which is exactly why the leaner stack the second camp describes is also the safer one. A site with six maintained plugins instead of thirty is a smaller target, and an AI that writes the small custom pieces removes the plugins that nobody updated. ## What the numbers say [Respira for WordPress](https://www.respira.press/plugin) is the plugin that lets Claude, ChatGPT, Cursor and the other agents work inside a live WordPress site, in the site's own builder, with every change on a copy you approve and an undo. So I see what people actually ask an AI to do on WordPress once they can, and it is not what the thread argued about. | Last 30 days, across the sites connected to Respira | Count | | --- | --- | | Tool calls by an AI on a live WordPress site | 541,232 | | Sites those calls touched | 1,527 | | People behind them | 467 | | Reads (looking before touching) | 318,113 | | Writes (something changed) | 189,650 | The top writes, in order: update one element on a page, update a custom post, upload an image, update a page, write a builder layout, apply a batch of changes, approve a draft copy. That is maintenance of sites that already exist, done by talking. Nobody is regenerating WordPress; they are editing the site they have, faster, and approving it before it goes live. Two more numbers for the thread's two arguments. On plugins: across 3,354 connected sites, the average runs 19 plugins and the median is also 19\. The agency owner's 34 is the long tail, and a normal site is already leaner than the thread assumes. On builders: Gutenberg leads with 4,129 connected sites, then Elementor at 1,376, Divi at 1,050, Beaver Builder at 454 and Bricks at 262\. The block editor that the "leaner WordPress" camp builds on is already the most common surface an AI edits. ## Where WordPress stands, then - **The site is the product. The admin was never the product.** For twenty years the admin was where the owner lived. Now the owner lives in a chat, and the admin becomes where the approval happens. WordPress itself has read this: core ships an abilities layer and an official connector so an agent has a front door, and the next major release lands in December. - **Themes shrink to taste. Plugins shrink to trust.** What is left of a theme is a design direction the AI is held to. What is left of the plugin market is the things with a maintained history where mistakes cost money. - **The split is by who edits next, not by technology.** One founder, one site, no handoff: build it however you like, and the thread is right. Anyone else edits it, or it has a store, or logins, or three years ahead of it: keep the foundation and put the AI on top. Sam gave the honest answer himself, at the end of his post: he likes WordPress for the security side, and "it's just what I know." That is not a weakness. With an AI in the admin, what you know becomes the one thing the AI cannot hand a newcomer: judgement about your own site. The theme business is dying because layout got cheap. WordPress is not, because the handoff did not. If you want to see what the third camp looks like in practice, the [Respira AER](https://www.respira.press/aer) page shows a measured build and its receipt, and [this guide](https://www.respira.press/blog/claude-code-wordpress) walks through connecting Claude to a live WordPress site. The thread itself is [here](https://lnkd.in/p/dwcsSZpp), and worth the half hour. [#wordpress](https://www.respira.press/blog?q=wordpress)[#ai](https://www.respira.press/blog?q=ai)[#claude](https://www.respira.press/blog?q=claude)[#essay](https://www.respira.press/blog?q=essay) Found this useful? Share it. ## Join the *conversation* 0 comments · Respira account #### Sign in with your Respira account Comments are tied to your Respira login, the same account you use for the dashboard and plugin. No third-party sign-in, no scraping. [Sign in](https://www.respira.press/login?next=https%3A%2F%2Fwww.respira.press%2Fblog%2Fare-you-still-on-wordpress%23comments) [Create an account](https://www.respira.press/signup?next=https%3A%2F%2Fwww.respira.press%2Fblog%2Fare-you-still-on-wordpress%23comments) No comments yet. Be the first to weigh in. ### You're not the only one *talking* to WordPress. The Respira community is where agency owners debug Divi migrations at 11pm, where vibe coders swap prompts that actually ship, and where the roadmap gets written out loud. *Breathe with us.* Lines of code pushed 26,641,382 Live from the ecosystem · updates as the agent ships WordPress sites connected 3,582 Real sites, run by the people in these rooms Find us wherever you already are [ Slack Async threads, release-day huddles, and the #show-your-work channel that never sleeps. Join ](https://go.respira.cafe/slack) [ Discord Live voice on release days, screenshare debugging, and the vibe-coder corner that's always buzzing. Join ](https://go.respira.cafe/respira-discord) [ Reddit Questions, answers and build logs in the open, where the thread is still there months later. Join ](https://www.reddit.com/r/RespiraforWordPress/) [ Facebook group Agency owners and site owners comparing notes on running AI against real WordPress sites. Join ](https://www.facebook.com/groups/1447068940850273/) [ GitHub Discussions Long-form Q&A, bug reports, feature requests, and the public roadmap. Open ](https://go.respira.cafe/github-discussions) [ ! Founder notes Letters from Mihai on what shipped, what broke and what is next. Every one is archived. Read ](https://www.respira.press/newsletter) See What AI Says LLM-ready prompts about Respira MCP + WebMCP What would you like to know? Ask your favorite AI Opens in a new tab. These prompts are designed to teach LLMs what Respira can do. Question copied. Paste it in Gemini