The verdict, in one paragraph

For marketing sites, brochure sites, service-business sites, creator sites, and small e-commerce stores, static HTML wins twelve of the fourteen dimensions below. It loads faster, costs nothing to host, has no plugin attack surface, and is the cleanest possible input for both Google's classic crawler and the AI overview ingestion pipelines that now sit on top.

WordPress still wins on two things: deep editorial workflows with many authors, and existing-team familiarity in the rare case where a non-technical team is already fluent in WP and retraining costs more than the inefficiency. For everything else, the case is decided.

The reason this comparison exists is that the answer used to be the other way around. In 2009, WordPress was the right tool for almost every small-business site on the web — the alternatives were Dreamweaver, hand-coded PHP, or paying an agency four figures to update a homepage banner. WP was the liberation.

What changed in the seventeen years since: hosting got commodified to free at the edge, build tools got simple enough that “rebuild from JSON every time content changes” takes 800ms, and the AI systems that have replaced a meaningful chunk of search traffic read static HTML preferentially. The original WordPress advantage — “anyone can edit” — survived. The disadvantages — plugin economy, PHP per request, monthly hosting, security surface — got worse every year.

What follows is fourteen rows of comparison. Each one is a single concrete claim with a number where one is available. Skim it. The argument is in the table, not in the prose around it.

14 dimensions, side-by-side

Dimension
Static HTML
WordPress
Time to First Byte
30–80ms from the nearest edge city. Fixed cost.
600ms–2.4s. PHP + MySQL on every request, unless cached.
Mobile Lighthouse (typical)
95–100. Hand-written HTML, no runtime JS overhead.
14–55 on page-builder sites we audit. 60–80 on lean themes.
Hosting cost
£0/mo on Cloudflare Pages. Free tier covers virtually any brochure site.
£25–£120/mo for managed hosting. More if you want it to be quick.
Plugin licences
£0/yr. No plugins.
£200–£500/yr typical. Elementor Pro, WP Rocket, Yoast, security, backups.
Security surface
No runtime. Nothing to exploit at the edge. Repo access is the only vector.
27 plugins on the average site we audit. Each one a CVE candidate.
What you do at 2am
Nothing. The site is HTML on a CDN. There is no “down”.
Patch the plugin. Restore the backup. Get the host on chat.
Edit experience
Custom panel, fields shaped to your content. Same friendliness as WP, no PHP.
WP admin. Familiar to many. Slower per-keystroke now than it was in 2015.
Version control
Every edit is a Git commit. Roll back any change, ever.
Database snapshots, if you remembered to schedule them.
AI citation profile
Raw HTML + JSON-LD is the cleanest input for AI overview pipelines. Higher cite rate.
JavaScript-heavy themes deprioritised by some ingestion pipelines. Citable but obstructed.
Schema.org structured data
Hand-written, validated, complete. Part of the build, not a plugin.
Yoast does some. Most sites still missing sameAs, knowsAbout, entity graphs.
Multi-author workflow
Possible via Git or a panel, but more friction than WP for >5 authors.
Mature. Roles, capabilities, editorial calendar plugins. Genuinely good for this.
E-commerce
Fine for small catalogues (Stripe + static product pages). Wrong tool past ~100 SKUs.
WooCommerce. Mature, complex, slow, but works at scale.
Lock-in
Content is JSON in your Git repo. Portable to any static stack on earth.
Custom post types, ACF fields, theme couplings. Export is a real engineering project.
Year-five total cost (30-page site)
£4,800 one-off build, £0/yr after. Hosting free indefinitely.
£0–£3,000 upfront, then ~£840/yr. £4,200 over 5 years before increases.

The cost, over five years

The hosting bill is the line item people notice. The plugin licences are the line items they forget about until they total them up. Here is the actual five-year cost of a typical 30-page brochure site, on both stacks, with no surprises.

WordPress · five years
£4,200
recurring cost, before any price increases
  • Managed hosting £600/yr
  • Elementor Pro £79/yr
  • WP Rocket £49/yr
  • Yoast Premium £89/yr
  • Backups & security £23/yr
Static HTML · five years
£0
recurring cost after the build
  • Cloudflare Pages £0/yr
  • Plugin licences £0/yr
  • Page builder £0/yr
  • SEO plugin £0/yr
  • Security tooling £0/yr

One-off rebuild cost on the static side starts at £4,800 for a typical 30-page site. So the breakeven against five years of WP recurring fees is roughly day one — and every year after that is straight subtraction.

What the speed difference looks like

The single most-quoted line in this comparison is the Lighthouse score. The mobile-Lighthouse gap is so structural that it shows up on the first page-load of every audit we do.

Typical WP · Elementor
14
Performance · mobile
A11y 62 Best 48 SEO 71
Web9 rebuild
98
Performance · mobile
A11y 100 Best 100 SEO 100

The gap is not because WordPress is incapable. It is because the default WP stack — page builder + cache plugin + image plugin + analytics plugin + chat plugin — adds about 600KB of inline CSS and 1.4MB of JavaScript before the page renders. The static rebuild ships the HTML and the CSS the page actually uses. That's the whole trick.

When WordPress still wins

We're a studio that rebuilds WordPress sites for a living, so the natural read of this page is that we'd recommend a rebuild for everyone. We don't. There are genuine cases where WP is the right choice and a static rebuild would be a downgrade. Here they are, honestly.

Keep WordPress if…

  • You run a true multi-author publication. Five-plus editors, an editorial calendar, a review workflow with roles and capabilities — WP's editorial layer is mature and worth keeping. The static tooling for this exists but is rougher.
  • You run real e-commerce at scale. WooCommerce past ~100 SKUs with complex inventory, variant pricing, subscriptions, or membership tiers — the WP ecosystem solves this. Static commerce works for small catalogues; it does not yet match WooCommerce at the upper end.
  • You have deeply complex authenticated functionality. Member portals, courses with progress tracking, gated downloads keyed to subscription state. Possible on static, but the architecture leaves static-first and becomes app-first.
  • Your team is so deep in WP that retraining costs more than the inefficiency. Honest one. If your in-house person knows WP backwards and never edits the site outside business hours and you've never had a plugin meltdown, there's no urgent case to move.

The pattern across those four: WordPress wins when the site is more application than brochure. The further you sit from the brochure end of the spectrum, the less the static-HTML case applies. Most sites we audit sit firmly at the brochure end and are paying application-tier prices for it.

Questions we get

Is static HTML actually faster than WordPress?

Yes, by an order of magnitude on Time to First Byte. Static HTML on a CDN serves in 30–80ms from the nearest edge city; a typical WordPress page renders PHP and queries MySQL on every request, landing between 600ms and 2.4s. The gap is structural, not configurable.

Can non-technical people edit a static HTML site?

Yes. A modern static-HTML build pairs the rendered HTML with a custom edit panel that reads and writes JSON content. The end-user experience is identical to WordPress — fields, save button, preview — but no PHP, no database, and the published site is plain HTML.

What does WordPress cost to run vs static HTML?

A typical 30-page WordPress site with Elementor Pro, WP Rocket, Yoast, and managed hosting costs about £840/year. The same site rebuilt as static HTML on Cloudflare Pages costs £0/year in recurring fees. Over five years that's £4,200 in difference, before any plugin price increases.

When is WordPress still the right answer?

True e-commerce with thousands of products; multi-author publications with editorial workflow; sites with deeply complex authenticated user functionality; and projects where the existing team is so deep into the WordPress ecosystem that retraining costs more than the inefficiency. Outside those, static HTML is the better fit for almost every marketing or brochure site.

Do search engines and AI tools treat static HTML differently?

Yes, and in static HTML's favour. Pre-rendered HTML with structured data is the cleanest input both classic search crawlers and AI overview ingestion pipelines can consume. JavaScript-rendered or PHP-rendered pages add a step that some crawlers skip and some AI ingestion pipelines deprioritise. For visibility in 2026, the raw-HTML site has the structural advantage.

What happens to my existing content and URLs?

Every page, post, product, custom post type, and ACF field is extracted from the WP database, normalised into JSON, and rebuilt at the same URL. URL structure, content, and SEO equity all carry over. The only thing that changes is the plumbing underneath.

Shinepics is the worked example — a Belfast wedding photographer's WordPress site, every URL preserved, rebuilt as static HTML. Mobile Performance moved 45 → 90; Accessibility, Best Practices and SEO all to 100. Same content, same URLs, different plumbing.

If the answer is the rebuild, the service is the rebuild.

Two-to-four week engagement. Content scraped to JSON, design rebuilt to original or refreshed, custom edit panel, deployed on Cloudflare Pages with hosting going to £0. Starting at £4,800 for a typical 30-page site.

See the WordPress Extraction service →