I Moved My Blog from WordPress to Astro: Why, What Changed, and When You Should Stay

My readability problems weren't WordPress's fault. I moved anyway, and here's the honest list: the reasons, the numbers, the traps, and the cases where I'd tell you not to.

DevelopmentWordPress9 min read

Status, November 17, 2026
Moved on November 11. I'll add the Search Console and Bing results at two and four weeks.

This blog ran on WordPress for years. On November 11, 2026, I moved it to Astro, hosted on Cloudflare. Same domain, same URLs, same posts.

"WordPress vs Astro" is a loud argument right now, on X, on Hacker News, in agency blogs. Most posts on either side are selling something: a migration service, a plugin, a hosting plan. So here's what actually happened on a small, real blog, including the part where the first thing I measured pointed away from moving at all.

Short answer

Move to Astro if your site is mostly content, one or two people publish it, nobody depends on the WordPress editor, and you're tired of updates, plugins and hosting quirks. Stay on WordPress if non-technical people edit the site, you rely on plugins for forms, memberships or a store, or your real problem is a slow theme. A theme can be fixed in a day; a migration takes several.

Why I looked at this in the first place

It started with a feeling: the blog was hard to read. Before changing anything, I measured one post on desktop and mobile:

  • On a 1440×900 screen, the first paragraph started at 970 px, below the fold. A gray header band and an empty "10 MIN READ" strip filled the first screen.
  • A contact form popped up on its own at 20% scroll and covered the text (on mobile, almost the whole screen).
  • Code blocks had no background, and tables broke into stacked cards on mobile.
  • The page loaded 15 scripts: jQuery and jQuery Migrate, emoji, syntax highlighting, sticky sidebar, smooth scroll, stats, anti-spam, the form, analytics.

Here's the honest part: none of that was WordPress's fault. Every item was the theme's CSS or a form setting. My first recommendation to myself was to keep WordPress and fix the theme, about a day of work.

So if you're here because your WordPress site feels slow or messy, measure first. It might be the theme, not the platform.

Why I moved anyway

Readability wasn't the reason to leave. These were:

  1. My host was blocking AI crawlers. A server-level bot list on my shared host returned 403 to ChatGPT, Claude and Perplexity's crawlers, and I couldn't change it. I had a workaround running through Cloudflare (the full story is here), but a workaround isn't a fix. Moving the blog off that server ended the problem for good.
  2. Maintenance. Core updates, plugin updates, a caching plugin, a security posture to think about. For a blog with eleven posts, that's a lot of moving parts.
  3. How I write now. I draft posts with AI tools and publish through a review step. With WordPress, that meant pushing content through the REST API, which once created a duplicate post when a request timed out. With Astro, a post is a Markdown file in a Git repository: every change is visible as a diff and can be reverted.
  4. Cost. A static Astro site on Cloudflare's free plan costs nothing to host.

Why so many people are talking about this in 2026

My reasons were personal, but the conversation around them is bigger. A few things changed in the past year:

  • Plugin security keeps getting worse. Patchstack's State of WordPress Security in 2026 counted 11,334 new vulnerabilities in the WordPress ecosystem in 2025, up 42% from the year before: 91% in plugins, 9% in themes, and only 6 in WordPress core.
  • Governance fatigue. The legal fight between WP Engine and Automattic, which started in 2024, is still in court. For people who build on WordPress for clients, uncertainty about who controls the ecosystem is a real cost, even if their sites never change.
  • Astro got a big backer. Cloudflare acquired the company behind Astro in January 2026. Astro stays open source (MIT) and can be deployed anywhere, but it now has long-term funding. In April, Cloudflare also previewed EmDash, an open-source CMS built on Astro that it calls a "spiritual successor" to WordPress. It's early, with almost no plugin ecosystem yet, but it shows where the momentum is.
  • WordPress is losing share. According to W3Techs, WordPress ran 43.2% of all websites on December 1, 2025, and 38.5% on November 17, 2026. It is still by far the most used CMS, but the drop is steady, month after month.

What people who switched say

I wanted to know whether my reasons were typical, so I read what people wrote about moving from WordPress to Astro, or about Astro itself, between November 2025 and November 2026: about 30 first-hand accounts in communities. Treat this as a sample, not a survey. Most voices are developers running their own sites, and several posts come from agencies or tools that sell migrations.

What they like

  • Speed. The most common reason. People coming from page builders or heavy themes report Lighthouse scores going from the 50s or 70s to 100, because Astro sends no JavaScript unless a page needs it (one example).
  • Nothing to maintain. No plugins, no database, no PHP version, no admin login for bots to probe. A developer who has built WooCommerce stores for eight years explained that this is why his own content site runs on Astro.
  • Content in Git. Posts are Markdown files, layouts are components, and every change is a diff. Developers who had used Gatsby or plain PHP describe it as the static site tool that finally felt simple.
  • Free or near-free hosting, usually on Cloudflare, and easy to move between hosts.
  • It works well with AI coding tools. This is the new reason in 2026. In the Hacker News thread on Cloudflare's acquisition, one commenter described a non-technical assistant updating their company's Astro site with an AI coding agent. I work on this blog the same way.
  • WordPress fatigue. Plugin conflicts that make every update a gamble, and the WP Engine and Automattic dispute, come up as reasons to leave more often than any Astro feature (example).

What they don't like

  • Editors lose the dashboard. The most common objection, mostly from agencies. Clients can't edit MDX, so you either add a headless CMS (a second system to learn) or keep editing the site yourself. Even an agency that moved its own site to Astro says it isn't for sites that non-technical staff edit often.
  • You rebuild what plugins did. Forms, comments, search, structured data, redirects; shops and memberships are the hardest. Migration write-ups list the same slips: dropped comments, no draft preview, shortcodes converted by hand, images still loading from the old host.
  • Upgrade pace. Astro 6 shipped in March 2026 and Astro 7 about four months later. In the Hacker News thread on Astro 7, one reply was a single word: "Exhausting." The maintainers answered that the breaking changes were few and came from the new version of Vite underneath.
  • Cloudflare lock-in. When Cloudflare bought Astro in January, a vocal minority remembered how Gatsby faded after its own acquisition. EmDash added to the worry, because its plugin sandbox depends on Cloudflare's runtime. Ten months later Astro is still MIT-licensed and deploys anywhere, but the question keeps coming up.
  • Wrong tool for apps. Some developers find it great for blogs and awkward for interactive apps; a few found it over-engineered and chose Eleventy.
  • Going back is rare, but it happens. The one first-hand account I found: a developer moved his personal site from Astro back to WordPress.com in November 2025 because he missed the integrated tools he uses every day at work.

The pattern matches my own case. The people happy with Astro mostly edit their sites themselves; the people who stay on WordPress usually have someone else editing. That's the next section.

When you should stay on WordPress

I'd tell a client to stay if any of these are true:

  • Non-technical people edit the site. The WordPress editor is the product. Astro has no admin screen out of the box; you either write Markdown in Git or bolt on a separate CMS.
  • You rely on plugins for features. Forms with logic, memberships, bookings, multilingual, search, WooCommerce. On Astro, each of those becomes a third-party service or custom code.
  • Your problem is speed or design. A lighter theme, a cache, and removing unused plugins usually get a WordPress site most of the way. That's days cheaper than a migration.
  • WordPress is still moving. WordPress 7.0 shipped in May 2026 with a built-in AI client and a refreshed admin dashboard.
  • You need to hire. It's much easier to find someone who can maintain a WordPress site than an Astro one.
  • The site already ranks and you have no time to test. A migration that breaks URLs can cost you traffic you spent years earning.

There's also a middle path: keep WordPress as the editor and use Astro only as the front end ("headless"). You keep the editor, but you also keep maintaining WordPress, so it doesn't solve the maintenance problem.

What changed after the move

Same posts, measured before and after:

WordPressAstro on Cloudflare
First load of a postabout 335 KBabout 76 KB
Script files on a post153 (the contact form loads only when someone opens it)
Hosting cost for the blogpart of a shared hosting plan$0 (Cloudflare free plan)
AI crawlers (ChatGPT, Claude, Perplexity)403 from the host, fixed with a workaround200, no workaround
Updates to applycore, theme, pluginsnone on the server

What got worse

  • No editor. Writing from my phone is now awkward. Possible (GitHub's web editor), but not pleasant.
  • No new comments. The blog had 13 public comments in three years. I kept them as static text under their posts and replaced the comment form with a "send me a question" form.
  • Stats lost continuity. Jetpack Stats is gone; Cloudflare Web Analytics replaced it. My before and after numbers come from two different tools.
  • Rolling back is harder. On WordPress, undoing a change was switching a theme off. Undoing this move means DNS changes and restoring the old server.
  • More of my stack is on one company. Astro is now owned by Cloudflare, and the blog is hosted on Cloudflare. Astro is open source and runs on other hosts, so I can leave, but it's worth knowing.

What to check before you migrate

These are the things that would have hurt if I had missed them:

  1. Keep every URL exactly the same, including the trailing slash. I tested 125 old URLs and images before switching DNS; all of them had to return the same page or a 301.
  2. Keep image paths. Images stayed at /wp-content/uploads/YYYY/MM/..., because other sites and Google Images link to them.
  3. Redirect the pages you're dropping. Tag and author pages had no search traffic, so I didn't rebuild them. Each one got a 301 to the closest category or post.
  4. Keep old section anchors. My Vercel guide's table of contents generated links like #Host_WordPress_for_Free_on_Vercel, and those deep links had about 10,500 impressions in Google over 12 months. Astro would have generated #host-wordpress-for-free-on-vercel and broken all of them. I kept 187 old anchors.
  5. Don't forget the feeds and sitemaps. /feed/ stayed at the same address, and the old Yoast sitemap names (/sitemap_index.xml, /post-sitemap.xml) redirect to the new sitemap.
  6. Rebuild what your SEO plugin did. Titles, meta descriptions, canonical tags, Open Graph cards, and Article and Person structured data. Compare them page by page with the live site.
  7. Test redirect rules on the real host. Cloudflare didn't apply a placeholder rule I wrote for old tag feeds (/tag/:tag/feed/), and they went to the homepage. I found it the day after the move and replaced it with one fixed rule per tag. Test every pattern on the live host, not only locally.
  8. Keep the old site as a backup, not public and not indexed, until you're sure.

The part I can't tell you yet

Google's data for the days after the move hasn't come in yet. The move went live on November 11, and Search Console's latest data was from before that. I'll update this post with the before and after numbers from Search Console and Bing at two and four weeks: indexed pages, impressions, clicks, and whether any old URL slipped through.

If you're deciding whether to move your own site, or you've already decided and want it done without losing your URLs, get in touch. Tell me what's on your site and who edits it, and I'll tell you whether moving is worth it.

Need this sorted for your business?

Thinking about moving your own site? Tell me what is on it and who edits it, and I will tell you whether moving is worth it.

Get in touch

Yunus ErtürkDesigner and developer. I build websites, mobile apps and digital products, and I fix the technical problems that slow small businesses down: hosting and migrations, email deliverability, payments and performance.