Drupal carries a reputation forged in the version 6 and 7 era: powerful, rigorous and punishing to approach. In 2026 the question of its relevance rests on different ground. The project has shipped a ready-to-use distribution, an AI-centred roadmap and a rebuilt site-building experience. What remains is knowing which projects it still fits.
Drupal CMS, the notable development
The event is not Drupal 11 as such, but Drupal CMS, released as version 1.0 on 15 January 2025 and stemming from the Starshot initiative announced by Dries Buytaert at DrupalCon Portland 2024. It is a ready-to-use download built on the Drupal 11 core, aimed at an audience the project had historically overlooked: marketing managers, content creators and site builders with no development skills.
Two building blocks carry the change: Recipes, configuration packages that install a complete feature in one operation, and Single Directory Components on the theming side. Experience Builder, a visual editing tool, completes the set. Official details are published on the initiative page on drupal.org.
Drupal CMS no longer aims to replace WordPress on small editorial sites. It targets the structured project that an accumulated WordPress install will not sustain.
The 2026 bet: AI and commerce
The 2026 roadmap is explicit: eight structural AI capabilities in Drupal CMS, with agents able to assist configuration, and a marked investment in commerce — e-commerce recipes, express checkout, Experience Builder applied to product pages. The project is therefore not merely closing an ergonomics gap: it is betting that AI-assisted site building will be a differentiator.
The promise of “agents that configure the site” nevertheless belongs to the category judged on delivered projects, not on announcements.
Drupal or WordPress: the comparison
| Criterion | Drupal (CMS / 11) | WordPress |
|---|---|---|
| Beginner onboarding | Acceptable since Drupal CMS | Unbeatable |
| Content modelling | Excellent, native | Plugins (ACF and equivalents) |
| Fine-grained permissions | Native, granular | Limited without extensions |
| Multilingual | In core | Plugin (Polylang, WPML) |
| Extension ecosystem | Narrower | Considerable |
| Cost and time for a small site | Higher | Lower |
| Robustness on a large project | Superior | Depends on team discipline |
The deciding criterion. When content is structured — nested types, relationships, editorial workflows, per-role permissions, demanding multilingual requirements — Drupal saves time over the life of the project. For classic editorial content with a need to iterate fast on a tight budget, WordPress remains the default choice.
Where Drupal still fits
- Institutions, public sector, universities: governance, permissions and multilingual support available as standard.
- Complex editorial applications where modelling outweighs time to market.
- Long-lived projects with an in-house technical team responsible for maintenance.
- Strong security and audit constraints, where a narrower ecosystem becomes an advantage.
Where it does not
- A blog, brochure site or small e-commerce store to launch in two weeks.
- A client with no recurring maintenance budget.
- A need for a niche extension that only exists in the WordPress ecosystem.
- A team that will not invest in learning the tool.
The main risk is not technical
Drupal’s structural weakness is the talent pool. Recruiting an experienced Drupal developer costs more and takes longer than finding a WordPress integrator, and that reality weighs as much on a quote as on taking over an existing project. Drupal CMS lowers the barrier to building; it does not necessarily lower it for maintaining a bespoke site. The question belongs alongside hosting decisions, where a well-managed self-hosted foundation beats bargain shared hosting.
What to take away
Drupal remains a defensible choice in 2026, provided it answers a genuine structuring need. Drupal CMS finally gives it a usable entry point, and the roadmap shows a project looking forward. It is not, however, a default tool: it earns its place when content is structured, the horizon long and the team ready to invest.
I worked with Drupal 6, hook_menu() and a learning curve closer to a cliff than a slope. That memory long served as my argument against the tool; it no longer holds. What still holds is the talent question: I raise it systematically with a client before proposing Drupal, because that is where projects break — not in the code. — Simon Janvier
