Stack Decision
Drupal vs WordPress — Drupal is fast because of its caches, and they have to be configured
Your content model is complicated enough that somebody has suggested Drupal, and nobody has told you what it needs from a server or what the toolchain looks like day to day.
The short answer
Budget for the toolchain: Drupal expects Composer for dependency management, Drush at the command line and generous PHP memory, and its performance depends on internal page cache, dynamic page cache and BigPipe being configured rather than merely present.
WordPress serves the overwhelming majority of organisational sites well and reaches speed through server-level caching instead. Drupal is the right answer where structured content, editorial workflow and deep integration are written into the brief. Our hosting runs either.
By the HostingFast team · Reviewed 24 August 2026
NVMe
Storage, every plan
Free
Domain, first 12 months
99.9%
Uptime target, watched
Flat
Renewal behaviour
Entities, fields and views give Drupal an architecture capable of modelling almost any content relationship you can describe. That is exactly why public bodies and universities standardise on it, and exactly why an eight-page brochure site finds it exhausting.
It is also why the performance conversation is different. WordPress speed is largely a question of what sits in front of PHP; Drupal speed is largely a question of whether its own cache layers are configured correctly, and that is work somebody has to do.
Drupal's speed lives in its own cache layers
Internal page cache serves anonymous visitors. Dynamic page cache handles authenticated requests by caching the parts that do not vary. BigPipe streams the personalised fragments afterwards so the page appears to arrive quickly even when parts of it are still being assembled.
All three are configuration rather than magic. A Drupal site that feels slow is very often one where anonymous caching was disabled during development and never switched back on, or where a contributed module marks everything uncacheable.
Verify it rather than assume it. Check the cache-related response headers on an anonymous request, then repeat as an authenticated user. If both look identical, one of the layers is not doing its job and no amount of hardware will substitute.
What each one needs from the host
Drupal wants generous PHP memory, a database with room to work, and SSH with Composer and Drush available. Overdrive is shaped for exactly that: 500 GB of disk, 3.5 cores with 2.5 GB of RAM, both PostgreSQL and MySQL on hand, and native support for Node.js, Python or Ruby alongside.
Smaller Drupal builds run perfectly well on a standard plan, so do not over-provision on reputation alone. The requirement to check is whether your build's Composer operations and cache rebuilds complete comfortably, not whether the platform has an intimidating reputation.
WordPress asks less and gets more from the layer in front: LiteSpeed page caching, Redis object caching and AccelerateWP on the managed plans, with WordPress Toolkit handling staging and updates.
The command line is part of the product
Drupal expects Composer to manage dependencies and Drush for routine operations — cache rebuilds, configuration import and export, user management. Both are available at the command line here, which matters because a Drupal deployment that cannot run Composer is not really a Drupal deployment.
Configuration management is the property that makes this worth it. Configuration lives in files, moves through version control and imports on deploy, which turns environment drift into a diff rather than a mystery. Git is available alongside Composer.
WordPress has an equivalent in WP-CLI, also available over SSH. The two toolchains are different in shape but similar in importance: the site you can drive from the command line is the site you can automate and reproduce.
Testing the decision before you commit to it
Build a small version of your real content model on each. Not a demo — your entities, your fields, your editorial workflow. Then time an authenticated editing session, because that is the experience the people producing pages will live with every day.
Turbo carries 20 GB of NVMe, 25 sites and 50 MySQL databases with per-site PHP switching, which is enough to host both prototypes on one account and compare them under identical conditions.
Staging copies come with the plans, migrations are free, and upgrades between tiers happen in place from the client area. None of that decides the architecture question — it just means the experiment costs almost nothing to run.
Prototype the content model, not the brochure
Every platform demo is a brochure site, which is the one case where both candidates are indistinguishable. Build your real entities and your real workflow, then measure the authenticated editing experience.
Both PostgreSQL and MySQL are available on the business tier, so a database preference does not have to force a hosting compromise.
- Three Drupal cache layers, all configured
- Composer and Drush at the command line
- Configuration in files, moved by Git
- Time the authenticated editor session
Why HostingFast
Standard on every plan
Composer, Drush and Git
SSH access with the toolchain a Drupal deployment assumes, rather than a file manager and hope.
PostgreSQL and MySQL both
Available on the business tier, so a database preference is not a reason to change host.
PHP memory you control
Per-site version switching and limits set from the panel, which is what cache rebuilds actually need.
LiteSpeed for the WordPress path
Server-level page caching plus Redis and AccelerateWP on the managed WordPress plans.
Staging for either platform
Clone, run the update, import the configuration and only then ship it to production.
Node, Python or Ruby alongside
The business tier deploys them natively, which matters when a CMS is one part of a larger system.
Prices Side by Side
Our figures set beside the household brands
Typical intro and renewal pricing across the market, lined up against ours — including the renewal figure most comparison charts skip.
| Feature | HostingFastMost popular | Typical household-name host | Typical bargain host | Typical loss-leader deal |
|---|---|---|---|---|
| Starting price / mo* | $2.42/mo | $4–$6 | $2–$4 | $1–$3 |
| Renewal price / mo | $2.42/mo | $10–$15 | $8–$12 | $4–$6 |
| Entry-level plan renews at its joining price | ||||
| SSL as standard | ||||
| Migration done for you | ||||
| NVMe drives even on the entry tier | ||||
| Daily backups starting at tier one | ||||
| Real humans on support, 24/7 |
*The figure in our column is the cheapest annual-term plan we sell, drawn live from the catalogue behind our pricing page, so it cannot go stale. The other three columns give the ranges that shared hosts in each bracket tend to advertise: opening offers that usually want a one-to-four-year commitment, then climb the moment it runs out. Naming rivals and printing their prices is something we stopped doing, because a number we cannot verify on the day you read it does not belong here. Put us beside whichever host you are genuinely weighing up. Then read the renewal row harder than any other.
Quick Start
From order to online
- 1
Model your real content, briefly
Build the entities, fields and workflow you actually need on each candidate. A demo site makes both look identical; your content model is the thing that separates them.
- 2
Verify the cache layers are on
Check the cache headers anonymously and then as an authenticated user. If they look the same, a layer is disabled or a module is marking everything uncacheable, and that is the problem rather than the hardware.
- 3
Time the editor, not the visitor
Measure an authenticated editing session end to end — loading a node form, saving, previewing. That is the experience your content team lives in, and it is almost never what a public benchmark measures.
Built In
Loaded onto every plan
- SSH with Composer, Drush and Git available at the command line
- Per-site PHP version switching with memory limits you control
- PostgreSQL and MySQL both available on the business tier
- Node.js, Python and Ruby deployable natively on that tier
- LiteSpeed caching in front of PHP on every plan
- Redis object caching and AccelerateWP on the WordPress plans
- Staging environments for rehearsing a deploy and a config import
- NVMe SSD storage on every tier, entry plan included
- Daily backups with self-service restores from the panel
- Free migration of an existing site, run by our engineers
Frequently Asked
Questions we get asked again and again
How much PHP memory does Drupal need in practice?
More during cache rebuilds and Composer operations than during normal serving, which is the part that catches people out. Treat it as a sizing input: if operations only complete after repeatedly raising the limit, the build wants a larger tier rather than another configuration change.
Do I get Composer and Drush on shared hosting here?
Yes — SSH access with Git and Composer is part of the standard plans, and Drush installs through Composer like any other dependency. That matters, because a Drupal deployment that cannot run Composer is not really a deployment, it is a set of files.
Does Drupal work with server-level caching?
Yes, and it complements rather than replaces Drupal's own layers. Internal page cache, dynamic page cache and BigPipe handle what the application knows about; the server cache handles anonymous responses in front of that. Confirm which one answered by reading the response headers.
Can I use PostgreSQL for a Drupal site?
On the business tier, yes — both PostgreSQL and MySQL are available there. Drupal supports either. Choose on what your team operates confidently rather than on a benchmark, because operational familiarity affects uptime far more than the engine choice affects query time.
Keep reading
How to Install WordPress
The clean installation path, with the step that trips people up flagged in advance.
Drupal Hosting
Drupal with Composer, Drush and per-site PHP control available from the start.
Secure Hosting
Imunify360, account isolation and hardened defaults for builds that cannot absorb a compromise.
Moving to a new host? Read the checklist before you start.
A plain running order for a move nobody browsing the site will notice: what to copy across first, how to carry mailboxes over without dropping one message, when to repoint DNS and not a minute earlier, plus the pair of errors responsible for nearly every outage we get called in to fix.
Bring the toolchain with you.
SSH with Composer, Drush and Git, PHP memory you control, and both PostgreSQL and MySQL on the business tier.
View Drupal Hosting plans