I build my own CMS for a living. I also still recommend WordPress to half my clients.
That sentence usually gets me side-eye from both sides of the room. The WordPress people think I am being condescending because I went off and built something else. The not-WordPress people think I am being soft because I have not fully renounced the old religion. Both are wrong, and both reactions are exactly why I think it is worth writing this down once, properly, so I can link to it next time someone asks.
WordPress powers roughly forty percent of the web for a reason
That number gets thrown around as if it is purely inertia. It is not. It is the size of the surrounding ecosystem.
If a small business owner in any city on earth needs a website, there is a freelancer within driving distance who can build it on WordPress. There is a host who specialises in it. There is a theme that already looks roughly like what they want. There is a plugin for the contact form, a plugin for the booking calendar, a plugin for the cookie banner, and a YouTube tutorial for every single one of them. There is a backup tool, a staging environment, a migration plugin, and a support forum where someone has already solved whatever weird problem just came up.
No other CMS comes close. Not even in the same postcode. That is not nothing. That is the entire reason a non-technical person can have a working website by the end of the week without hiring a developer at all.
For the brochure site, it is still the right answer
Take the most common small business case. A roofer, a dentist, a bakery, a yoga studio. They need maybe seven pages, a contact form, a blog they will update twice a year, and a way to embed a Google Map. They have a budget that does not stretch to a custom build, and they need to be able to change the phone number on the homepage themselves when they move office.
WordPress plus a decent theme is the right answer nine times out of ten. It is not even close. Anything custom is overkill. Anything headless is a maintenance burden they cannot carry. A static site generator is great until they want to change something and there is no GUI. WordPress nails this entire category and it has been nailing it for fifteen years.
Gutenberg in 2026 is actually good
I am going to die on this hill. The block editor stopped feeling like a beta around 2024, and the full site editor has matured into something I genuinely enjoy using. Patterns are properly first class now. Block themes are stable. The query loop block does ninety percent of what custom post archives used to need a developer for. Style variations let a non-technical user rebrand a site without touching CSS.
There are still rough edges. The block library is still inconsistent across third-party blocks. The interactivity API is still finding its audience. But the core experience of editing a page in WordPress in 2026 is fast, visual, and predictable. People who left during the 2018 to 2022 transition years and have not come back genuinely do not know what the current product feels like.
The boring advantages still matter
Hosting is cheap. Three quid a month gets you a working WordPress site on managed hosting with backups and SSL. Support is everywhere. Every agency on earth has at least one person who has been working with it for a decade. The learning curve is shallow enough that a motivated non-developer can be productive in a weekend.
If I tell a client to use my CMS, I am also telling them that the only support engineer on earth is me. That is fine for the kind of clients I build custom things for. It is wildly inappropriate advice for someone who just wants a website for their pilates studio.
The honest pushback
None of this means WordPress is the right answer for everything. It is not.
Headless WordPress is a compromise nobody really wins at. The REST API is fine, the GraphQL plugins are fine, but you end up running two stacks and paying for both. If you are going headless, you almost always want a CMS that was designed to be headless from day one.
Structured content is the bigger one for me. The native model in WordPress is post first, structure second. ACF and Meta Box do an enormous amount of work to bolt structure onto a system that fundamentally thinks in terms of posts and pages. If your content has a strict schema, hard relationships, and downstream consumers who care about type safety, WordPress will fight you the whole way.
And then there is the security story, which is the awkward truth nobody likes to say out loud. The plugin ecosystem is also the security ecosystem. Almost every WordPress vulnerability that makes the news comes from a third-party plugin, not from core. Core has actually been pretty hardened for years. The risk is that a site with twenty-eight active plugins has twenty-eight separate codebases of varying quality, and any one of them can be the door. Good operational hygiene fixes most of this, but most sites do not have good operational hygiene.
My actual client list
Half my current clients are on WordPress. The other half are migrating off, and every single one of those has a specific reason. One needs strict typed content for a multi-language product catalogue. One wants an editor experience that ties directly to a custom backend that has nothing to do with web pages. One is consolidating four different systems into a single owned platform and a generic CMS does not fit the shape of the problem.
None of them are leaving because WordPress is bad. They are leaving because their requirements moved past the place WordPress is best at.
That is the framing I wish more people used. WordPress is excellent at the thing it is excellent at. When your project sits inside that envelope, use it. When your project sits outside it, do not. The mistake is dragging it into shapes it was not built for, or refusing to use it on a project where it would obviously be the right tool.
Final word
You can hate WordPress on Twitter all you want. You can write threads about how it is bloated and how PHP is dead and how the block editor is a war crime. None of that is going to change the fact that the web still runs on it, that small businesses still need websites, and that for an enormous category of work it remains the most pragmatic option on the table.
Build the right thing for the project in front of you. Sometimes that is a bespoke Go CMS with an MCP surface. Sometimes that is WordPress with three plugins and a theme from 2022. The job is to know the difference.