WordPress Themes Are Becoming Irrelevant (For Some of Us)

I spent the better part of a week evaluating WordPress themes for a site rebuild. I compared Elementor, Bricks Builder, Ollie, Kadence, and GeneratePress. I watched demos, read documentation, tested pattern libraries, poked at header builders. And somewhere in the middle of it, I had a thought I couldn’t shake: why am I doing any of this?

I already have AI tools that write CSS on request, build layouts from a description, and explain exactly what code to paste and where. The entire value proposition of a premium theme had quietly evaporated, and I hadn’t noticed until I was already three themes deep into a comparison I no longer needed to make.

What You Used to Pay a Theme For

The pitch for a premium WordPress theme was always about saving time on design decisions. You paid for someone else’s taste and expertise, packaged into a product. Specifically, you were buying:

Pre-built page patterns and templates so you’re not starting from a blank canvas
A visual builder so you can design without touching code
A header and footer builder with drag-and-drop controls
Typography systems, color palettes, and spacing presets that work together
A plugin ecosystem that plays nicely with the theme

That’s a legitimate bundle of value. If you don’t know CSS, can’t hire a developer, and need a professional-looking site in a weekend, a premium theme or page builder solves a real problem. Elementor Pro, Kadence Pro, Bricks Builder at $79/year, Ollie Pro: these products exist because for a long time, they were the best available tool for that job.

The problem is that the job has changed.

The Moment the Calculus Shifted

I was rebuilding The Good Life Collective, a community site I run. It’s not a complex site. I wanted a clean layout, a custom header, some specific typography choices, and a handful of reusable section patterns. Nothing exotic.

I started evaluating themes the way I always had: features list, demo quality, community reputation, price. But this time I kept catching myself thinking: I could just ask Claude to write that CSS. I could describe this layout and get the block markup in two minutes. I could build that header pattern myself with a bit of back-and-forth in a chat window.

The pre-built patterns that used to justify a theme’s price tag? I can generate something more specific to my actual design in about five minutes. The visual builder? I’m increasingly faster describing what I want in plain English than clicking through nested panels. The header builder with its seventeen configuration options? One custom HTML block and some CSS that AI writes for me.

Every feature that used to differentiate a premium theme from a free one is now something I can produce on demand, at zero cost, in minutes. The premium themes haven’t gotten worse. The alternative has just gotten dramatically better.

What a Theme Actually Needs to Do Now

After going through this evaluation, I landed on a much shorter list of requirements. A theme now needs to:

Load WordPress correctly without fighting the block editor
Add as little bloat as possible (no unnecessary scripts, no mystery CSS files)
Stay maintained so security patches actually arrive
Not lock you into proprietary patterns that break when you switch

That’s the entire list. Design decisions, pattern libraries, visual controls: all of that is now handled outside the theme. The theme is plumbing. It should be invisible.

This means the thinner the theme, the better. GeneratePress on the free tier is a strong choice. Twenty Twenty-Five, the default WordPress theme, is genuinely excellent as a starting point. Both are well-maintained, add almost nothing unnecessary, and stay out of your way. That’s the whole brief now.

The Paid Theme Market Has a Real Problem

I want to be fair here because I’ve recommended these products before and I still do for the right audience. The companies building premium themes aren’t doing anything wrong. They built excellent products for a genuine need. But the need is shifting fast.

When your core value proposition is “design decisions made for you,” and AI can make better-tailored design decisions on request in real time, the business case gets harder to defend. Not impossible, but harder. The market for “I need a website and I’m not using AI tools” will persist for a long time. Millions of people run WordPress sites and will continue to buy themes and builders. But for anyone actively using AI in their workflow, paying for a premium theme is increasingly hard to justify.

The $79/year Bricks license, the Elementor Pro subscription, the Kadence Pro upgrade: these are all buying you convenience. AI provides that same convenience for free, with more flexibility and no lock-in. That’s a structural problem for the theme market, not a temporary blip.

This Doesn’t Mean Themes Are Dead

I wrote a post earlier this year on the best WordPress themes for 2026 that recommends Ollie, GeneratePress, Astra, Kadence, and Bricks. That post is still useful, and those recommendations are still accurate for their intended audience: people who aren’t using AI-assisted development, people who want a turnkey solution, people who value visual builders over code.

If you run a WordPress site for a client who manages it themselves, a theme with a good visual builder still makes sense. If you’re handing off a site to someone who needs to add pages without touching code, pattern libraries and page builder UI matter a lot. The themes-are-irrelevant argument applies specifically to people in the workflow I’m describing: developers and power users who have AI tools in their corner.

For that group, the landscape has changed. The old comparison criteria (pattern quality, header builder flexibility, WooCommerce integration depth) have been largely superseded. What you’re evaluating now is code quality, maintenance track record, and how little the theme gets in your way.

The Editor Doesn’t Matter Either

Once the theme becomes invisible plumbing, the same logic extends to the WordPress editor. I’ve used TinyMCE (the classic editor) for over a decade. I never warmed to Gutenberg. But during this rebuild, I realized the editor debate is also becoming irrelevant.

If you write in Markdown (and many of us do), you can compose in whatever tool you prefer, whether that’s Obsidian, iA Writer, a plain text file, or even a conversation with an AI. Then you either paste it into WordPress (which converts Markdown to blocks automatically) or push it via WP-CLI without ever opening the admin panel. The writing environment and the publishing environment don’t need to be the same thing anymore.

The theme is plumbing. The editor is plumbing. What matters is the writing itself and how the finished site looks and performs. Everything in between is becoming a solved problem.

What I Actually Did

I ended up dropping Elementor entirely for the rebuild. After evaluating the alternatives, I went with a lightweight free theme as the base and have been building out what I need with AI-generated CSS and custom patterns. The result is cleaner, faster, and more specifically suited to what I actually wanted than any pre-built theme would have produced.

It’s also cheaper. And the code is mine, not dependent on a third-party builder’s proprietary markup that breaks when they push an update or raise prices.

The irony of spending a week evaluating themes only to conclude I barely needed one is not lost on me. But it was a useful week. I came out of it with a clearer picture of what a theme’s job actually is in 2026, and it’s a much smaller job than it used to be.

Grab GeneratePress free or Twenty Twenty-Five, describe your design to an AI, and build from there. The era of paying for someone else’s design decisions is coming to an end, at least for those of us with better tools at hand.