Choosing a content management system commits a project for several years: build cost, maintenance cost, availability of skills, capacity to evolve. This guide compares WordPress, Drupal and PrestaShop on the criteria that actually decide, and offers a decision grid usable in a scoping meeting.
The criterion that decides: content structure
Before any feature comparison, a single question separates the families of tools: is the content structured or editorial?
- Editorial — articles, pages, a few taxonomies. The need is to publish quickly and often.
- Structured — nested types, relationships between entities, approval workflows, per-role permissions, demanding multilingual requirements. The need is to model correctly.
An editorial project run with a modelling tool costs a lot for nothing. A structured project run with an editorial tool ends as an accumulation of extensions no team wants to take over.
The comparison
| Criterion | WordPress | Drupal (CMS / 11) | PrestaShop |
|---|---|---|---|
| Onboarding | Unbeatable | Acceptable since Drupal CMS | Average |
| Content modelling | Extensions (ACF and equivalents) | Excellent, native | Catalogue-oriented |
| Fine-grained permissions | Limited without extensions | Native, granular | Native (employee profiles) |
| Multilingual | Extension (Polylang, WPML) | In core | In core |
| Ecosystem | Considerable | Narrower | E-commerce specialised |
| Cost of a small project | Low | High | Medium |
| Robustness at scale | Depends on discipline | Superior | Good on the catalogue |
| Availability of skills | Very wide | Narrow and expensive | Average |
WordPress: the default choice, and what weakens it
WordPress remains the rational choice for most editorial projects: speed of delivery, low entry cost, an immense talent pool. Its weakness is known and treatable: the accumulation of extensions.
Two practices change the trajectory of a WordPress project over five years:
- Manage dependencies rather than installing from the admin — a project structure such as Bedrock makes deployment reproducible and history readable.
- Count extensions as debt: every activated extension loads resources on every page and becomes a breaking point at each version upgrade.
The question is not to pick a side between WordPress and Drupal, but to match the tool to the real nature of the project.
Drupal: relevant, with conditions
Drupal changed face with Drupal CMS, the ready-to-use distribution released in January 2025, its Recipes and its Experience Builder. Modelling, permissions and multilingual support remain its structural strengths.
Its weakness is not technical but human: recruiting an experienced Drupal developer costs more and takes longer. That point weighs as much on a quote as on taking over an existing project, and arises before the technical decision.
PrestaShop: the e-commerce case
For a shop whose catalogue is the core of the business — variants, combinations, pricing rules, carriers, accounting — a native e-commerce tool keeps the advantage over an extension grafted onto an editorial CMS. The real cost lies in major version upgrades and third-party module maintenance, which must be budgeted from the outset.
Decision grid
| Situation | Default choice |
|---|---|
| Editorial site, fast launch, constrained budget | WordPress |
| Publication with daily output and multilingual needs | WordPress with a well-managed multilingual extension |
| Institution, public sector, university | Drupal |
| Editorial application with complex modelling | Drupal |
| Shop whose catalogue is the business | PrestaShop |
| Shop attached to a content site | WordPress with an e-commerce layer |
| No recurring maintenance budget | WordPress, and nothing else |
Questions to ask the client before deciding
- Who will maintain the site in three years, and on what budget?
- Does the content have relationships between entities, or only categories?
- How many distinct roles take part in publishing?
- Is multilingual a translation, or different content per market?
- What is the tolerance for downtime during version upgrades?
The answers to those five questions determine the choice more reliably than any feature comparison.
Articles in this section
- Drupal in 2026: still a reasonable choice?
- Node.js 26 enables the Temporal API by default
- CloudPanel: self-hosted WordPress without a heavyweight panel
This page is updated as articles are published in the Back-end section.