Is WordPress still worth it in 2026?
WordPress is worth it when someone owns the updates and you need a plugin with no equivalent. For most small business sites it is not. The numbers, and what happened when we moved our own and client sites off it.
Updated 11 November 2026
On this page
WordPress is still worth it in 2026 when someone owns it: a person or provider who applies core, theme and plugin updates every week, on a site that needs a plugin with no equivalent elsewhere. It is a poor choice for a small business site that nobody maintains. 91% of the 11,334 WordPress vulnerabilities found in 2025 were in plugins, and fewer than half of WordPress sites (48% in September 2026) pass Google's Core Web Vitals on mobile.
It is not dead. W3Techs has it on 39.5% of all websites as at 5 November 2026, though that share has fallen every month for a year. Everything below is dated November 2026 and sourced. The last third is our own experience: we hosted and maintained WordPress sites ourselves, then moved them off.
Is WordPress still worth it in 2026?
WordPress is worth it in 2026 if you have someone who will update it every week and a reason to stay, and not worth it if the site is a brochure or service site that nobody will look after. The software is fine. The cost sits in the upkeep.
Worth it when:
- someone is responsible for core, theme and plugin updates, and does them weekly;
- the site depends on a plugin with no equivalent elsewhere, such as a specific booking or membership plugin;
- an in-house team already knows WordPress, or editors will not leave wp-admin;
- there is a large content archive that a team publishes to daily.
Not worth it when:
- the site is a handful of service pages and a contact form;
- nobody is named as the person who updates it;
- it runs a page builder and a long plugin list to do what a few pages need;
- speed, security and a low monthly cost matter more than a plugin marketplace.
Doing nothing is also a valid choice for a stable site that is working, as long as you are not leaving updates undone on a live WordPress install.
| Measure, as at November 2026 | WordPress | Static-first build (Astro) | Hosted builders |
|---|---|---|---|
| Share of all websites | 39.5% | Not reported separately | Wix 4.2%, Squarespace 2.4% |
| Mobile Core Web Vitals pass rate | 48.3% | 71.1% | Wix 80.9%, Squarespace 71.0% |
| Median mobile page weight | 2,656 KiB | 1,326 KiB | Wix 2,703 KiB, Squarespace 3,488 KiB |
| Who applies updates | You, or a retainer | The developer | The vendor |
| Entry hosting | WP Engine from US$28/month (billed annually) | Cloudflare Workers paid plan, US$5/month per account | The builder's plan |
| Main risk | Plugins nobody updates | Needing a developer for changes | Less control, harder to move away |
Sources: share from W3Techs; pass rates and page weight from the HTTP Archive Core Web Vitals Technology Report, September 2026; hosting from WP Engine's Australian plans page (US$28/month, billed annually, as at November 2026) and Cloudflare's Workers pricing.
Is WordPress dying, or do people still use it in 2026?
WordPress is not dying. It runs 39.5% of all websites and 58.2% of sites with a known CMS (W3Techs, 5 November 2026). It is shrinking slowly: its share of all sites fell in every month from November 2025 to November 2026.
| Date | WordPress share of all websites |
|---|---|
| 1 Nov 2025 | 43.2% |
| 1 Jan 2026 | 43.0% |
| 1 Apr 2026 | 42.5% |
| 1 Jul 2026 | 41.5% |
| 1 Sep 2026 | 40.7% |
| 1 Oct 2026 | 40.2% |
| 1 Nov 2026 | 39.5% |
Source: W3Techs monthly history. Its share of sites with a known CMS peaked at 65.2% in 2022 and is now 58.2%. Read the fall with care: over the same year the "no CMS detected" bucket rose by 3.6 points, to 32.2%, so some of it may be detection or a move to custom-built sites rather than sites switching to another named CMS. On W3Techs' count, 68.1% of WordPress sites now run version 7.
What are the downsides of using WordPress?
The downsides of WordPress are plugin risk, heavier pages than a static build, and upkeep that never ends. Each has a number behind it, below. None of them is a flaw in the core software. They come from how most WordPress sites are assembled.
- Plugins carry the risk. The risk is in third-party code, covered in the next section.
- Pages are heavier than a static build. The median WordPress page is about twice the weight of the median Astro page (2,656 against 1,326 KiB on mobile), covered in the speed section.
- It is never finished. Core, theme, plugin and PHP updates have to keep working together, and an update can break a live site. We wrote up one routine update that did. You either pay someone to watch it or you accept that it will age.
Is WordPress safe from hackers?
WordPress core is safe in practice, and the plugins are where sites get hit. Patchstack recorded 11,334 new vulnerabilities in the WordPress ecosystem in 2025, up 42% on 2024. 91% were in plugins, 9% in themes and 6 were in WordPress core, all rated low priority.
| Patchstack figure, 2025 | Value |
|---|---|
| New vulnerabilities in the ecosystem | 11,334 |
| In plugins / themes / core | 91% / 9% / 6 issues |
| Not fixed by the vendor when disclosed | 46% |
| Median time to mass exploitation of heavily exploited flaws | 5 hours |
| Attacks blocked by the hosts Patchstack tested | 26% |
Source: Patchstack, State of WordPress Security in 2026, updated 25 February 2026. A plugin with a published flaw and no fix leaves you waiting on a developer you do not control, and a host's firewall will not catch most of it.
We saw what that looks like on a client site. In October 2026 we rebuilt the website of a Wollongong uniform and embroidery supplier, and it had been hacked. Its backup, dated 17 September 2026, showed 18 active plugins. The form plugin was 49 releases behind the current version, with five published unauthenticated flaws listed against that version between May and August 2026. The page builder had not been updated for about a year while WordPress itself kept auto-updating. In the form plugin's upload folder were nine files ending in .php, .phar and .phtml, uploaded in three batches of three between late August and early September, and those files ran. The pattern is the one Patchstack describes: the core was current and a plugin nobody updated was the way in. If you have a site like this, our security service covers clean-up.
Why is WordPress slow?
WordPress is slow when it builds every page on request from PHP and a database, and then loads a page builder and plugin scripts on top. The fixes are caching and fewer plugins, and both are effort the platform does not do for you.
Google's Core Web Vitals pass rate shows how this plays out across real visitors. Fewer than half of WordPress sites pass on mobile, and the weak point is loading rather than responsiveness: 55% of WordPress sites have good Largest Contentful Paint, against 90% for responsiveness (INP).
| Platform | Mobile pass, all sites | Mobile pass, Australia |
|---|---|---|
| Wix | 80.9% | 86.2% |
| Shopify | 76.2% | 76.5% |
| Astro | 71.1% | 60.5% |
| Squarespace | 71.0% | 52.3% |
| WordPress | 48.3% | 50.2% |
| Elementor sites | 36.3% | Not pulled |
| Next.js | 35.7% | 38.0% |
| TanStack Start | 19.3% | 15% |
Source: HTTP Archive Core Web Vitals Technology Report, September 2026, share of origins with good LCP, INP and CLS. TanStack Start detection began in August 2026 and its sample is probably weighted towards app-style sites.
Two points. First, the frameworks we use ourselves do not pass more often than WordPress as a group: Next.js and TanStack Start pass on fewer sites. We are not claiming otherwise. What the table supports is narrower: static-first builds that ship little JavaScript, such as Astro, pass on far more sites, and a build approach decides speed more than the framework name. Second, around 60% of WordPress sites use a page builder, according to the Web Almanac 2025, and sites built with Elementor pass on 36.3%.
| Median mobile page weight, September 2026 | Total (KiB) | JavaScript (KiB) | Images (KiB) |
|---|---|---|---|
| Astro | 1,326 | 399 | 346 |
| TanStack Start | 1,537 | 408 | 540 |
| Next.js | 2,119 | 802 | 399 |
| WordPress | 2,656 | 661 | 1,017 |
| Wix | 2,703 | 1,716 | 184 |
| Elementor | 2,871 | 693 | 1,078 |
| Squarespace | 3,488 | 1,460 | 1,048 |
| Shopify | 3,921 | 2,234 | 732 |
Source: HTTP Archive, same report. WordPress is about twice the median Astro page, and it is not heavier than the hosted builders. "Bloated" is accurate against a static-first build and not against Wix or Squarespace.
What does a WordPress site cost to run in 2026?
A WordPress site costs hosting, plugin licences and someone's time. WP Engine's entry managed plan is US$28 a month, billed annually, as at November 2026, before premium plugins and before whoever applies the updates. A static site on Cloudflare has no PHP or database to run, so hosting is a small cost: the Workers paid plan is US$5 a month per account.
On the other side of the ledger, summaries of "WordPress vs custom website" quote A$25,000 to A$150,000 for a custom build (Google AI Overview, retrieved 6 November 2026). That is not what a small business has to pay. Our website plan is A$298 a month, GST inclusive, on a 12 month term, with no upfront build fee, and it includes hosting and upkeep. See pricing for what is in it, and our cost of a website in Australia guide for the wider picture.
What should you use instead of WordPress?
Use the simplest thing that fits the site. A service or brochure site is best as a static-first build, either an Astro site or a custom one. A site with logins or an app needs a full-stack framework. A shop is usually better on Shopify. If you want no developer involved at all, a hosted builder such as Wix or Squarespace is a fair choice and passes Core Web Vitals more often than WordPress, as the table above shows. If editors will not leave wp-admin, a headless setup or staying put are both reasonable.
Search results for "WordPress alternatives for small business Australia" list builders only. A custom static site is the option they leave out. For where to go and how, see migrating off WordPress in 2026 and the WordPress migration checklist.
When is WordPress still the right choice?
WordPress is the right choice when it has an owner and a reason. That means weekly updates are someone's job, the site relies on a plugin with no equivalent elsewhere, an in-house team knows the platform, or a large archive is edited daily in wp-admin. In those cases the plugin marketplace and editing experience are worth the upkeep. It is software that needs an owner, and the trouble starts when nobody is that person.
What happened when we moved sites off WordPress?
We moved our own site and four client sites off WordPress in 2025 and 2026. Here is what changed on their home pages, measured on 6 November 2026 against the last archived WordPress version.
Why we moved
We hosted and maintained WordPress ourselves. Our own site and at least two client sites, a burger takeaway and a law firm, shared a WordPress server we ran. The costs were patching and updates, with things breaking after them; hacks and malware clean-up; server work, including PHP and Varnish tuning, backups and uptime; and pages that stayed slow despite caching. In our experience the sites were slow, with load times of five to six seconds in some instances. That is our recollection, not a lab measurement, and we have not published field timings.
We left for all four reasons together: speed and page weight, security and plugin upkeep, hosting cost, and control. We wanted the code in git on a stack we own, instead of a page builder.
A prospect we spoke to had the same shape of problem: 31 plugins, $180 a month, and a home page that took about six seconds to load.
The measured results
| Home page, mobile, own files | Before | After | Own JavaScript, before to after |
|---|---|---|---|
| A burger takeaway: WordPress, Elementor, WooCommerce, Gravity Forms, to Astro (May 2026) | 3,521 KiB | 1,252 KiB | 509 KiB in 10 files to 46 KiB in 2 files |
| A law firm: WordPress, Elementor, Gravity Forms, to Astro (May 2026) | 4,244 KiB | 1,841 KiB | 698 KiB in 7 files to 238 KiB in 7 files |
| A pest control business: WordPress, Elementor, 7 visible plugins, to Next.js (Feb 2025), then TanStack Start (Jun 2026) | 3,009 to 3,617 KiB | 1,711 KiB | 450 to 789 KiB in 5 to 33 files to 408 KiB in 21 files |
Method: Playwright with Chrome, mobile viewport, cold cache, scrolled to the bottom, decoded bytes of requests to each site's own host. "Before" is the last archived WordPress home page on the Wayback Machine (the takeaway and the law firm, 23 November 2025; the pest control business, March and September 2024), and archive gaps mean the before figures are undercounts. Counting everything on the page, the takeaway went from 4,709 KiB to 1,778 KiB, 2.6 times lighter, and its HTML document from 206 KiB to 66 KiB. The law firm's HTML went from 422 KiB to 74 KiB.
Two honest caveats. The pest control business's JavaScript is roughly level with its 2024 WordPress page, so its weight win is in the rest of the page, not in scripts. And on 6 November the law firm's page was heavier in total than the WordPress one, because a Google Maps embed added about 3.6 MB when scrolled into view.
Maps and our own site, fixed on 11 November
That map embed was a problem we had made, so we fixed it, along with the pest control and supplier sites. The maps now load only when someone clicks them. Measured on 11 November 2026, desktop viewport, home pages, transferred kilobytes before and after the deploy:
| Home page | Before | After | Saved |
|---|---|---|---|
| The pest control business | 1,887 KB in 70 requests | 1,326 KB in 38 | 561 KB, 32 requests |
| The law firm (two maps) | 2,373 KB in 65 | 1,743 KB in 40 | 630 KB, 25 requests |
| The uniform and embroidery supplier | 2,209 KB in 73 | 1,651 KB in 51 | 558 KB, 22 requests |
Run to run noise is about 30 KB. On our own site, the same day, we found dashboard code was being loaded on every marketing page. Moving it out cut JavaScript transfer on the home page from 1,589 KB to 1,150 KB, and on the pricing page from 1,581 KB to 1,141 KB. Our own site is not lighter than the WordPress one it replaced, and we are not claiming it is. It still carries the Google tags we need for analytics and Ads conversion tracking.
The security case, and the outcomes
The supplier rebuild above is the clearest one: the rebuilt site has no plugins, no database on the public site and no upload endpoint. The same is true of the other rebuilds: no plugin to go stale, no admin login to attack and no CMS database, because content is files in git. Some keep a small cache or form store, so we do not say "no database at all".
On results, the takeaway had 35% or more extra visitors in the three months after launch against the three before, and the pest control business had about four times the Google clicks over 4.5 months, though over half of those are searches for its name. Those came with new content and SEO work as well as a new platform, and we would not credit the platform alone. Three of the client sites also record forms and phone taps in our dashboard, and together they record more than a dozen enquiries a month, measured over different periods.
For the owners, what changed is that there is nothing to update, no plugin to go stale and no admin login to attack. Speed gains are real where we kept JavaScript low, as on the Astro sites, and smaller where we did not. Our own opinion, and it is an opinion rather than a measurement, is that leaving WordPress was about 100 times better than staying for these sites.
What is happening with WordPress in 2026?
WordPress is in the news for its governance as much as its software. WordPress 7.0 shipped on 20 May 2026 with an AI client in core. In the dispute between WP Engine and Automattic, a federal judge kept WP Engine's antitrust claims alive in an order on 24 September 2026. In September, Automattic's board tried to put Matt Mullenweg on leave as CEO and he returned within days. It is worth watching if you are betting a site on it.
How can you tell a WordPress site was built badly?
A WordPress site was built badly when it is slow, carries many plugins nobody can explain, and cannot be edited or updated safely by the person who owns it. The design can look fine and the build can still be poor.
- Count the active plugins and check when each was last updated.
- Time the main content on mobile data; several seconds is a warning sign.
- Try editing one sentence yourself, and note whether anyone is responsible for updates.
Our 20 minute checklist walks through each check. If you want us to look at your site, the free website audit is where to start, and our web design page shows how we build.
Common questions
- Is WordPress dead in 2026?
- No. W3Techs measured WordPress on 39.5% of all websites on 5 November 2026. Its share has fallen every month for a year, from 43.2% in November 2025, so it is shrinking slowly rather than dying. Part of that fall coincides with a rise in sites where no CMS is detected.
- Is WordPress dead because of AI?
- No. WordPress 7.0, released on 20 May 2026, added an AI client to core. What AI changes is the cost of a custom build, which makes leaving cheaper for small sites that nobody wants to maintain.
- Is WordPress safe to use?
- The core software is, with 6 low-priority core vulnerabilities recorded in 2025. The risk is plugins: Patchstack counted 11,334 new vulnerabilities in the WordPress ecosystem in 2025, 91% of them in plugins, and 46% were not fixed by the vendor when they were published. It is safe enough when someone applies updates weekly and keeps the plugin list short.
- Is WordPress free, or is it paid now?
- The WordPress.org software is free. Hosting, premium plugins and themes, and the time to maintain the site are paid. WordPress.com is a separate hosted product with its own plans.
- Which is faster, Wix or WordPress?
- On published field data Wix passes Google's Core Web Vitals on 81% of mobile sites against 48% for WordPress (HTTP Archive, September 2026). The trade is control and ownership: a hosted builder is quick to start and harder to move away from.
- Is Next.js or TanStack Start faster than WordPress?
- Not automatically. In the same HTTP Archive data 35.7% of Next.js sites and 19.3% of TanStack Start sites pass on mobile, both below WordPress. Static-first builds such as Astro pass on 71.1%. Speed comes from how much JavaScript a site ships, not from the framework badge.
- Should I use WordPress or build from scratch?
- If nobody will own the updates and the site is a service or brochure site, a custom static build is less upkeep. If you rely on a plugin that only exists for WordPress, or you have an in-house team that knows it, stay.
- Can I leave WordPress without losing my Google rankings?
- Yes, if the redirect map is done properly. Rankings are lost in migrations because old URLs are left to 404, not because the platform changed. The work is inventorying every indexed URL, deciding where each one goes, and carrying titles and descriptions across.
- How can you tell if a WordPress site was built badly?
- Look at what it is carrying rather than how it looks. Count the active plugins and check when each was last updated, time the main content on mobile data, and try to edit one sentence yourself. A site nobody can safely edit, running plugins their authors abandoned, is badly built however good the design is.
Want a website that turns readers into customers?
A custom site, hosting and upkeep for one monthly fee, with your leads and Google results in a dashboard you can check any time.