Block Themes or Classic Themes: Choosing in 2026

The honest answer depends on who edits the site after you hand it over, not on which approach is newer. What each one costs, where block themes genuinely win, and when classic is still the right call.

The way this question usually gets asked is which approach is more modern, and the answer to that is not useful. Both are supported, both are actively developed, and both will still work in five years. The decision turns on something far more boring: who is going to edit this site after you walk away.

Get that answer first and the rest follows.

The actual difference

A classic theme is PHP. Templates are PHP files selected by the template hierarchy, styles are CSS you write, and layout is whatever you build. Changing anything structural means editing code.

A block theme is HTML templates plus a configuration file. The templates live in a templates folder as block markup, reusable pieces live in a parts folder, and a single configuration file declares your colour palette, typography scale, spacing, layout widths and element defaults. Those declarations generate CSS custom properties and expose the same controls in the editor interface.

The consequence that matters: in a block theme, the person editing content can also edit the template. In a classic theme they cannot.

That one difference decides most projects

If the site will be maintained by a marketing team that wants to build a new landing page on a Tuesday without raising a ticket, a block theme is the right answer and it is not close. Giving a competent non-developer the ability to assemble pages from your components, inside constraints you defined, removes an entire category of dependency. That is a real organisational improvement, not a technical preference.

If the site will be maintained by developers, or by nobody, the editing advantage is worth nothing and you are left paying its costs. A classic theme gives you direct control, simpler debugging and markup that is exactly what you wrote.

The awkward middle is a client who will edit occasionally and inexpertly. A block theme hands them the ability to break the header. Constraints in the configuration file reduce the blast radius considerably, and they do not eliminate it.

What block themes genuinely do better

Design tokens in one place. Declaring your palette, type scale and spacing centrally, and having those propagate to both the front end and the editor controls, is a genuinely better arrangement than maintaining them in CSS and hoping the editor matches. Changing a brand colour becomes one edit.

The editor shows the real thing. Editing a page and seeing it as it will appear, with your actual styles, removes a persistent source of client confusion.

Patterns. Pre-built sections that an editor can drop in and fill are the feature clients notice. A library of your own patterns, covering the layouts the site actually needs, is where most of the practical value sits.

Less custom code for common things. Query loops, navigation, post lists and archive layouts come as blocks rather than as PHP you write and maintain.

What they cost

Less control over markup. You get the markup the block produces. Most of the time that is fine and occasionally it is not what you wanted, and working around it is more effort than writing the template yourself would have been.

CSS you did not write. Block themes generate a substantial amount of CSS from your configuration, plus per-block styles, plus inline styles for individual block settings. It is manageable, and it requires attention. Measure what actually ships on a real page rather than assuming it is lean.

Plugin blocks are a weight risk. A plugin that registers blocks frequently loads its assets broadly. This is the same discipline problem described in Core Web Vitals for WordPress without plugin overload, and block themes make it easier to accumulate.

A real learning curve. Not the concepts, which are simple. The specifics: what a given configuration property actually controls, which settings are exposed where, why a style is being overridden. Budget for it on your first project and do not quote as though you will not spend it.

Client-side breakage. Somebody will eventually delete a template part and not know what they did.

You do not have to choose completely

The option most teams miss is that a classic theme can adopt a configuration file for global styles without becoming a block theme. You get centralised design tokens, consistent editor controls and a palette that matches your CSS, while keeping PHP templates and full control over structure.

For bespoke marketing sites built by developers and maintained by developers, this is frequently the best position. You take the part of the new system that is unambiguously an improvement and leave the part that mainly serves non-developer editing.

It is also a sensible staging post if you expect to move to a block theme later, because the design tokens carry over.

When classic is still clearly right

Highly bespoke layouts that do not decompose into blocks. If every page is structurally different and driven by custom fields, you will spend more time fighting blocks than you would writing templates.

Sites where markup is tightly constrained, by an accessibility requirement, a performance budget or an external design system you must match exactly.

Heavy custom post type work with complex relationships, where a PHP template expressing the query directly is simpler than assembling it from block settings.

Projects where a developer is the only editor, forever. The main benefit does not apply.

When block is clearly right

Marketing sites with a content team who will build pages.

Multi-site or multi-brand setups where a shared set of design tokens and patterns gives you consistency across properties.

Anything where the handover matters more than the build, which is most client work.

Sites where the layout vocabulary is genuinely repeatable, which is most brochure and content sites once you look honestly at them.

What is a waste of effort

Choosing based on which is newer. It is not an argument, and it is the reason a lot of unsuitable projects got rebuilt.

Rebuilding a working classic theme as a block theme with no editing requirement driving it. You are spending a budget to arrive where you already were.

Exposing every control to editors because the system allows it. Constrain the palette, constrain the type scale, lock what should not change. An editor given unlimited options will produce a site with nine shades of blue.

Assuming either choice decides your performance. It does not. Fonts, images, third-party scripts and plugin assets decide it, which is the argument in performance budgets for marketing websites.

Either way, accessibility is yours to get right. Blocks produce reasonable markup and they will still let you build a page with four H1 elements, so the discipline in accessibility and SEO applies equally.

Migrating an existing classic theme

If you decide to move, the work is more predictable than it sounds, provided you do it in stages rather than as a rewrite.

The first stage is the configuration file alone. Move your colours, type scale, spacing and layout widths into it and have your existing CSS consume those values. Nothing visible changes, your templates stay as they are, and you now have design tokens in one place. This stage is useful even if you stop here.

The second stage is converting template parts. Header and footer first, because they are self-contained and the result is immediately visible. A classic theme can use block template parts without becoming a block theme, so this is reversible.

The third stage is the templates themselves, one at a time, starting with the simplest archive and leaving your most bespoke page template until last. Keep the classic PHP file alongside until you are satisfied, because the fallback is free.

Build your pattern library as you go rather than afterwards. Patterns are what the editing team will actually use, and a block theme handed over without them gives people a blank canvas and no guidance, which produces exactly the inconsistent site you were trying to avoid.

What happens to your existing content

This is the question clients ask and it has a reassuring answer. Post and page content lives in the database independently of the theme, so switching does not touch it.

Two things do need attention. Content built with a page builder plugin is stored in that plugin’s format and will continue to depend on the plugin, which means a theme migration does not free you from it. And content written as classic editor HTML renders inside a single block rather than as native blocks, which works perfectly well and is not editable in the new way until converted.

Convert selectively. Your handful of important landing pages are worth converting by hand so the team can edit them. Two hundred blog posts are not, and leaving them as they are costs nothing.

Custom fields keep working. Widgets are the one genuine casualty, since block themes handle sidebars differently, so audit what is in your widget areas before you start rather than after.

How to decide, quickly

  1. Ask who will edit the site in a year, by name and role. If it is a non-developer who will build pages, lean block.
  2. Ask how repeatable the layouts are. Count the genuinely distinct page structures. Under about eight, blocks fit well.
  3. Ask how bespoke the markup needs to be. Hard constraints push towards classic.
  4. Ask whether anyone on the team has shipped a block theme before. If not, add real time to the estimate or take the hybrid route.
  5. If it is still close, build classic with a configuration file for global styles. You keep control, you get the design tokens, and the path forward stays open.

If you are weighing WordPress against a JavaScript front end entirely, that is a different question and headless WordPress with Next.js covers it, along with what migrating actually involves.

We cover both approaches hands on in the SEO and WordPress training in Pune, and if you would rather hand the decision over, get in touch.