Moving off WordPress in 2026: when it pays and how it works
By Kumar Deepansh ·
Moving off WordPress is worth it when your site is mostly content that changes rarely, and you are paying in maintenance, plugin risk or speed for features you do not use. The move itself is four jobs: export the content, rebuild the templates, map every old URL to a new one, and redirect with permanent redirects. If you rely on WordPress plugins for things like memberships or shops, staying, or going headless, is often the better call.
When is it worth leaving WordPress?
It tends to be worth it when most of these are true:
- The site is a brochure, blog or documentation site, and the content changes weekly or less.
- You update plugins mostly out of fear, not because you use their features.
- Nobody on the team needs a visual page builder.
- You are tired of the maintenance bill or the security worry.
On the security point, the numbers are about the ecosystem, not your site. Patchstack counted 11,334 new WordPress vulnerabilities in 2025, with 91% in plugins, 9% in themes and 6 in core. A static site has no PHP plugin layer to exploit, which removes that class of problem. That is my inference from the architecture, not a figure from Patchstack. It does not make a static site automatically secure, since your hosting account and build pipeline still matter.
It is usually not worth it when you run a store, memberships, forums or complex forms that depend on plugins. Astro’s own migration guide is blunt about this: WordPress has an extensive plugin and theme market, Astro’s ecosystem is smaller, and you may need to build features yourself.
What are my options: static, headless or stay?
| Option | Content lives in | Best when |
|---|---|---|
| Static site (Astro) | Markdown files in the project | Editors are comfortable with files or a Git-based workflow |
| Headless WordPress | WordPress dashboard | Editors want to keep the WordPress editor |
| Stay on WordPress | WordPress | You depend on plugins that have no replacement |
Astro’s docs describe both of the first two. For a full move, content moves into Markdown files in the project, and you edit code instead of using a dashboard. For headless, they say you can use WordPress as a headless CMS: you keep writing posts in the dashboard, and Astro builds and presents the rest of the site, deployed separately. In the headless case you still run and patch a WordPress install somewhere, so the security benefit is smaller.
How does the content export actually work?
WordPress has a built-in export under Tools. Its documentation says it creates an XML file in the WXR format, covering posts, pages, custom post types, comments, custom fields, categories, tags, custom taxonomies and users. The documentation I read does not mention media files, so I do not rely on the export for images.
For Astro, I convert that XML to Markdown. Astro’s guide points to a community tool, wordpress-export-to-markdown, and warns that you may need to adjust its output and that large or complicated sites need more conversion work. The tool’s npm page says it:
- saves each post as a Markdown file with frontmatter, and also saves drafts, pages and custom post types
- downloads images and updates the references to them
- needs the export to be set to “All Content”
After conversion I check the results by hand: broken frontmatter, leftover shortcodes and page-builder markup, image paths and internal links. Anything a plugin generated, such as forms, galleries, sliders and related-post widgets, has no direct equivalent and gets rebuilt or dropped. Decide which ones you keep before you start.
How do I keep my SEO safe?
This is where migrations quietly fail, so I treat it as its own project. Google’s site-move documentation gives the core rules:
- Map old URLs to new ones before launch. Google tells you to prepare a mapping of old to new URLs. It also warns against redirecting many old URLs to one unrelated destination.
- Use permanent redirects. Google says 301 and 308 mean a page has permanently moved, and that 301 and other permanent redirects do not cause a loss in PageRank. Google also recommends a server-side redirect whenever possible, and says to use JavaScript redirects only if nothing else is possible.
- Keep the redirects for a long time. Google’s guidance is for as long as possible, generally at least 1 year.
- Expect a delay. For small to medium sites it can take a few weeks for most pages to move, and larger sites take longer. Google suggests timing the move for lower traffic if possible.
- Use Change of Address only for domain changes. Google says you only need that Search Console tool when moving from one domain or subdomain to another. If your URLs stay on the same domain, redirects alone are the job.
The easiest SEO win is to keep URLs identical. If your WordPress permalinks were /blog/post-name/, make Astro produce the same paths and you need fewer redirects. Where URLs must change, I build the redirect map from a crawl of the old site plus the sitemap, then test every old URL after launch.
What does hosting a static site cost?
A static Astro site can run on Cloudflare’s free tier, with limits you should know. As of November 2026, Cloudflare’s Pages limits page lists these for the Free plan:
| Limit | Free plan |
|---|---|
| Builds per month | 500 |
| Concurrent builds | 1 |
| Build timeout | 20 minutes |
| Files per site | 20,000 |
| Largest single file | 25 MiB |
| Custom domains per project | 100 |
_redirects |
2,000 static and 100 dynamic redirects |
_headers rules |
100 |
Check those against your site. A blog with a few thousand images can approach the file limit, and a large redirect map can exceed 2,000 static redirects.
Cloudflare’s current Astro guide deploys a static site to Workers with static assets rather than Pages, and confirms a static Astro site needs no Cloudflare adapter because pages are pre-rendered at build time. Cloudflare’s migration notes say requests for static assets on Workers are free, but also that Workers does not support domains whose nameservers are not managed by Cloudflare, which Pages does. I have not checked Workers’ own limits table for this post, so read it before choosing that route.
How does the cutover go?
My usual order:
- Build the new site on a staging URL and block it from search engines.
- Compare old and new page lists, titles and metadata.
- Load the redirect map and test old URLs in bulk.
- Switch DNS or the domain at the host, as Astro’s guide describes for keeping the same URL.
- Submit the new sitemap and watch Search Console for errors in the following weeks.
- Keep the old WordPress backup, and leave it offline, not public.
FAQ
Will I lose my Google rankings if I move off WordPress?
Not automatically. Google says permanent redirects do not cause a loss in PageRank, but pages still need to be crawled again, and Google says a move can take a few weeks. Rankings can wobble if URLs, titles or content change at the same time.
Can I keep editing in the WordPress dashboard?
Yes, with a headless setup, where Astro’s docs say your content stays managed in WordPress while the site is built separately. You still maintain WordPress itself.
What happens to my plugins?
They have no direct equivalent in Astro. Features like forms, search or galleries are rebuilt with integrations or custom code, or replaced with a third-party service.
How long should I keep the redirects?
Google’s guidance is for as long as possible, generally at least 1 year.
Sources
- Migrating from WordPress (Astro Docs)
- How to move a site (Google Search Central)
- Redirects and Google Search (Google Search Central)
- Tools Export screen (WordPress.org Support documentation)
- wordpress-export-to-markdown, version 3.0.6 (npm, published July 2026)
- Cloudflare Pages limits (Cloudflare Docs, read November 2026)
- Astro on Cloudflare Workers (Cloudflare Docs, read November 2026)
- Migrate from Pages to Workers (Cloudflare Docs, read November 2026)
- State of WordPress Security in 2026 (Patchstack, February 25, 2026)