← Back to Blog

WordPress vs a Custom Website: Which Does Your Business Need?

For most UK small businesses running a handful of pages to generate enquiries, a custom website wins. It beats WordPress on speed, security, and total cost — once you count three years of plugin licences, security patching, and slow-page lost conversions. For a business that publishes high volumes of content daily and needs non-technical staff editing constantly, WordPress is still a reasonable choice. The two tools solve different problems, and most of the advice online skips straight past that.

WordPress now runs 40.2% of all websites and 58.8% of every site built on a known content management system (W3Techs, Usage Statistics of Content Management Systems, checked September 2026). That scale is real. But scale isn't the same question as "is it right for your business" — and the data on speed, security, and cost tells a more complicated story than WordPress's market share suggests.

Key Takeaways

  • WordPress powers 40.2% of all websites and 58.8% of the known-CMS market — but among the internet's 1,000 fastest sites, only 8% run any identifiable CMS at all (HTTP Archive, Web Almanac 2024).
  • Only 40% of WordPress sites pass Google's mobile Core Web Vitals test, up from 28% in 2023 — meaning most still fail it (HTTP Archive, 2024).
  • WordPress's plugin and theme ecosystem added 11,334 new vulnerabilities in 2025, a 42% year-on-year rise; 91% came from plugins, not WordPress itself (Patchstack, State of WordPress Security in 2026).
  • A 0.1-second mobile speed improvement lifted conversion rate by up to 8.4% across 37 brands in a Google-commissioned study (web.dev, "Milliseconds make millions").
  • For a lead-generation or service-business site, the decision usually comes down to editing frequency versus speed, security, and long-term cost — not "which platform is more popular."

What's the real difference between WordPress and a custom website?

WordPress is a content management system (CMS) — pre-built software that handles page storage, user logins, and a plugin architecture, onto which you add a theme and third-party plugins to get the features and design you need. A custom website is code written and structured specifically for one business, with no general-purpose CMS layer underneath — every line ships because that business needs it, not because a template might need it for someone else.

That difference sounds abstract until you look at what it does to a page's weight. A WordPress site typically loads its own core files, a theme's CSS and JavaScript, and separate code for every plugin — a contact form plugin, an SEO plugin, a page-builder plugin, a caching plugin — even on a page that doesn't use most of what they load. A custom build ships only the code that specific page actually needs.

This isn't a small distinction. It's the reason the rest of this comparison — speed, security, cost — comes out the way it does.


How big is WordPress, really — and does it matter?

WordPress runs 40.2% of all websites and 58.8% of every site using an identifiable CMS (W3Techs, checked September 2026). No competitor comes close; it's the default choice for a huge share of the web.

But market share measures adoption, not performance. Look at where the fastest sites on the internet actually sit: among the top 1,000 highest-traffic sites globally, only 8% run any identifiable CMS at all — WordPress or otherwise (HTTP Archive, Web Almanac 2024, CMS chapter). The sites with the largest engineering budgets and the highest performance stakes overwhelmingly choose custom-built systems.

Scale versus speed WordPress dominates the web, but not the fastest sites on it 0% 20% 40% 60% 80% WordPress: all websites 40.2% WordPress: known-CMS market 58.8% Any CMS: internet's top 1,000 sites 8%
Sources: W3Techs, Usage Statistics of Content Management Systems (checked Sept 2026); HTTP Archive, Web Almanac 2024, CMS chapter.

That's not a coincidence. Adoption and performance answer different questions, and small businesses need to answer the performance question for themselves rather than borrowing WordPress's popularity as a proxy for "the right choice."


Which is faster: WordPress or a custom build?

Custom-built sites are faster by default, because they only ship the code a page actually needs. WordPress sites can be made fast, but most aren't: only 40% of WordPress sites pass Google's mobile Core Web Vitals assessment, the industry-standard speed test, and that's actually an improvement — it was 28% in 2023 (HTTP Archive, Web Almanac 2024).

WordPress mobile Core Web Vitals pass rate Share of sites meeting Google's "good" speed threshold 0% 50% 100% 2023 28% 2024 40%
Source: HTTP Archive, Web Almanac 2024, CMS chapter.

The typical WordPress page weighs 2,047 KB on mobile and 2,252 KB on desktop in the median case (HTTP Archive, Web Almanac 2024) — before accounting for images. Every plugin adds its own CSS and JavaScript files, most of which load on every page regardless of whether that page uses the plugin's feature. A caching plugin, a security plugin, and a form plugin running together is a common combination, and each one is a separate piece of code the browser has to download and execute before the page becomes usable.

Why does 0.1 seconds of load time matter enough to build an entire buying decision around? In a Google-commissioned analysis covering 37 brands and more than 30 million sessions, a 0.1-second improvement in mobile page speed lifted retail conversion rate by up to 8.4% and average order value by up to 9.2%; lead-generation sites saw bounce rate improve by up to 8.3% for the same gain (web.dev, "Milliseconds make millions"). Speed isn't a technical vanity metric — it's directly attached to how many enquiries a site converts. Google's own Core Web Vitals guidance sets the "good" threshold at a Largest Contentful Paint of 2.5 seconds or faster (web.dev, Core Web Vitals), and a plugin-heavy WordPress build makes that threshold considerably harder to hit than a page with no CMS overhead to load first.

None of this means WordPress can't be made fast. It can, with a disciplined theme, a minimal plugin list, and ongoing performance maintenance. The point is that speed on WordPress is something you have to actively fight for against the platform's own architecture — on a custom build, it's the starting position.


Which is more secure?

WordPress itself is maintained carefully. The vulnerability comes almost entirely from what gets attached to it. In 2025, the WordPress ecosystem added 11,334 new vulnerabilities — a 42% increase year-on-year — and of those, 91% originated in plugins, 9% in themes, and just 6 in WordPress core itself (Patchstack, State of WordPress Security in 2026). Seventeen percent were rated high severity, and more than a third — 4,124 vulnerabilities — required an active protection rule because no patch was available yet.

Where WordPress's 2025 vulnerabilities came from 11,334 new vulnerabilities across the ecosystem 0% 50% 100% Plugins 91% Themes 9% WordPress core <1% (6)
Source: Patchstack, State of WordPress Security in 2026.

This lines up with Sucuri's most recently published Hacked Website Report, covering 2023 — the most recent year Sucuri has confirmed figures for — which found WordPress accounted for 95.5% of the infected CMS sites its analysts examined that year (Sucuri, Hacked Website Report). That figure reflects WordPress's dominant market share as much as any inherent weakness, but the mechanism matters: every plugin you install is third-party code with its own update schedule, and a site running a dozen active plugins is running a dozen separate attack surfaces that all need patching on time, indefinitely.

A custom website doesn't remove security risk entirely, but it removes this specific pattern. There's no marketplace of third-party plugins sitting on your site by default. The attack surface is limited to code someone actually reviewed for your business.


What does each one cost, really?

WordPress usually costs less to get started — that's accurate, and it's the main reason so many small businesses choose it. A theme, a page-builder plugin, and some hosting can get a basic brochure site live cheaply. The costs that don't show up in that initial quote are ongoing: plugin licence renewals, security monitoring, and developer time whenever an update breaks something (which, with 11,334 new vulnerabilities surfacing across the plugin ecosystem in a single year, happens regularly).

A custom build costs more upfront because every page is designed and coded specifically for the business rather than assembled from existing parts. What it doesn't carry forward is the recurring plugin-and-patch maintenance cycle, or the conversion loss that comes with a slower page. We've written a full breakdown of UK pricing for both approaches in our guide to what small-business websites cost in the UK and our guide to custom software costs in 2026 — the short version is that the gap narrows, and sometimes reverses, once you price in three years of ownership rather than just the build.

If your current site is already slow, cluttered with plugins, or difficult to update safely, it's worth reading our guide to the signs your website needs a redesign before deciding which direction to rebuild in.


Does the platform affect whether Google and AI tools find and cite you?

Partly, and mostly through speed and structure rather than the platform's name. Google's ranking systems and AI answer engines both reward pages that load fast and are built with clean, unambiguous markup — a plugin-heavy page that's slow to render is a disadvantage on both counts. We've covered this shift in detail in our comparison of GEO vs SEO — what changed in 2026 and our plain-English guide to what generative engine optimisation actually is.

Where the platform matters more directly is content structure. AI engines extract answers most easily from pages that state their point in the first sentence of a section, name their sources, and use clean heading hierarchy — a discipline we cover in how to write answer-first content for AI search. That structure is achievable on WordPress with the right theme and editorial discipline, but it's the default output of a custom build, because the template controls the structure rather than a page-builder plugin improvising it page by page. For the full playbook on making any site AI-search-ready, see our guide to optimising your website for AI search.

Neither platform automatically wins here. But a slow, plugin-bloated WordPress site is fighting both Google's speed signals and AI engines' crawl-and-extract preferences at the same time — and that's a self-inflicted problem, not an unavoidable one.


When does WordPress actually make sense for a UK small business?

WordPress is the right tool in a specific, recognisable set of situations:

  • You publish content daily or near-daily, and multiple non-technical staff need to log in and post without waiting on a developer — a busy news site, a multi-author blog, or a content-led membership business.
  • You're validating a business idea on a minimal budget and need something live in days rather than weeks, accepting the maintenance trade-off in exchange for speed to launch.
  • You need a specific, mature plugin ecosystem — WooCommerce for a straightforward online shop, for example, has a large community and a well-understood feature set that's hard to justify rebuilding from scratch for a simple product catalogue.
  • You already have in-house WordPress expertise and the discipline to keep plugins minimal, licensed, and patched on schedule — the platform's risks are manageable, just not automatic.

If none of these describe your business, the case for WordPress weakens considerably.


When does a custom website make more sense?

A custom build is the stronger choice when:

  • Your website's job is to convert enquiries, not publish volume. A service business, consultancy, or trades company typically has a handful of pages that need to load fast and be easy to update occasionally — not a publishing workflow for dozens of contributors.
  • Speed and conversion rate matter directly to revenue, and the data above on Core Web Vitals pass rates and conversion lift from marginal speed gains applies to your traffic.
  • You want your site to look and function differently from competitors running the same theme and plugin stack, rather than visibly assembled from the same off-the-shelf parts.
  • You'd rather pay once for a lean, purpose-built site than take on an open-ended maintenance commitment across a dozen third-party plugins with independent update schedules.

This is the case ART Digital Media builds for by default — see our bespoke web design service for how we approach it for UK small businesses specifically.


How do WordPress and a custom build compare side by side?

FactorWordPressCustom-built
Typical upfront costLowerHigher
3-year cost of ownershipOften higher (plugins, patching, dev time)Often lower once built
Default mobile speed40% pass Core Web Vitals (2024)No CMS overhead to overcome
Security surfaceEvery active plugin is a separate riskLimited to reviewed, purpose-built code
Non-technical daily editingStrong — built for thisDepends on the CMS layer chosen, if any
Design differentiationConstrained by theme/plugin stackFully custom
Best fitHigh-volume content publishingLead-generation and service-business sites

How do you actually decide?

Start with how your website earns its keep. If its job is to be updated constantly by non-technical people — a blog, a news outlet, a content business — WordPress's editing workflow and plugin ecosystem are genuinely hard to beat, and the maintenance overhead is a fair price for that flexibility.

If its job is to convert visitors into enquiries and stay that way for years with occasional updates, weigh the ongoing cost properly: plugin renewals, security patching hours, and the conversion rate you're losing every time a page misses the 2.5-second Core Web Vitals threshold. Priced honestly over three years, a lean custom build is frequently the cheaper option, not just the faster one.


Frequently Asked Questions

Can I move from WordPress to a custom website without losing my Google rankings?

Yes, if the migration is done properly. The main risks are broken URLs and lost content, not the platform change itself. Google doesn't rank pages because they're built on WordPress — it ranks them on speed, relevance, and authority signals like backlinks. A well-planned migration preserves URL structure (or uses correct redirects), keeps the same content and internal links, and can improve rankings afterwards if the new site is faster.

Is WordPress inherently insecure?

No — WordPress core itself had only 6 new vulnerabilities out of 11,334 across the ecosystem in 2025 (Patchstack, 2026). The risk comes almost entirely from third-party plugins and themes, which is why a minimal, well-maintained plugin list matters more for security than which CMS you chose in the first place.

Do I need to know how to code to update a custom website?

It depends how the custom website is built. Some custom websites include a lightweight content-editing layer for text and images, so non-technical updates don't need a developer. Structural changes — new pages, new features — typically go through the agency or developer who built it, similar to how a WordPress site needs a developer for anything beyond what its plugins already support.

Is WooCommerce still a sensible choice for a small UK online shop?

For a straightforward product catalogue with standard needs, yes — it's mature, well-documented, and has a large support community. Where it starts to strain is once you need custom checkout logic, non-standard product configurations, or tight performance requirements, at which point the plugin-stacking pattern that affects WordPress generally starts to apply to your shop too.

How much does WordPress maintenance actually cost compared to a custom website?

WordPress maintenance is an ongoing line item — plugin licence renewals, security monitoring, and periodic update work, because 91% of the ecosystem's 2025 vulnerabilities came from plugins that need patching on their own schedules (Patchstack, 2026). A custom website shifts that cost forward into the build instead of spreading it across years of patching — see our UK website cost guide for current figures on both.


Sources

#StatisticSourceYear
1WordPress runs 40.2% of all websites; 58.8% of the known-CMS marketW3Techs, Usage Statistics of Content Management Systems2026
2Only 8% of the internet's top 1,000 sites run any identifiable CMSHTTP Archive, Web Almanac 2024, CMS chapter2024
3WordPress mobile Core Web Vitals pass rate: 28% (2023) to 40% (2024); median page weight 2,047 KB mobileHTTP Archive, Web Almanac 2024, CMS chapter2024
40.1-second mobile speed gain lifted conversion up to 8.4%, AOV up to 9.2%Google/web.dev, "Milliseconds make millions"2020
5Core Web Vitals "good" LCP threshold: 2.5 secondsweb.dev, Core Web Vitals2025
611,334 new WordPress-ecosystem vulnerabilities in 2025 (+42% YoY); 91% plugins, 9% themes, 6 corePatchstack, State of WordPress Security in 20262026
7WordPress = 95.5% of infected CMS sites analysedSucuri, Hacked Website Report2023