WordPress fed me for 6 years. Between 2017 and 2023 I built dozens of websites in WP, installed themes, optimised plugins, fought with updates that broke the site, fought comment spam, configured backups, cleaned hacked sites, trained clients to maintain their site without breaking it. And in 2023 I migrated my own website to Astro. Since then, I haven’t made a new website in WordPress for a client. And this post is to explain why, with real data from my last 8 migrations.
This is not a generic “Astro is better than WordPress” post. It’s a post about “me, Daniel, left WordPress for these specific reasons, with these numbers, in these projects”. If it helps you, great. If not, I recommend a colleague who still does WordPress and does it very well, because WordPress is still useful for specific cases. I’m going to try to be honest on both sides.
My last 8 migrations (real data)
Between January 2024 and June 2026, I migrated 8 client websites from WordPress to Astro (or a similar stack: Next, SvelteKit, Hugo in 2 cases). This is what happened in each one, anonymising the client’s name but with the real data.
Migration 1: law firm website (Madrid). WP + Elementor + 18 plugins, made in 2019. Load before: 3.8 s on mobile (PSI 62). Load after: 0.9 s (PSI 98). Organic traffic 3 months later: +42%. Form conversions: +28%. Cost of migration: €1,400. What cost the most: redoing the 8 practice area pages with their content (the client’s texts were good, I took them out of WP as they were, but the HTML structure had to be redone).
Migration 2: artisan ceramics online shop (Seville). WP + WooCommerce + 32 plugins, made in 2020. Load before: 5.2 s (PSI 48). Load after: 1.1 s (PSI 96). Organic traffic 6 months later: +25%. Conversion rate: +18%. What cost the most: migrating 800 products from WP to the new CMS (I did it with a Python script in 2 days, it wasn’t as much work as I thought). What we lost: the option for the client to edit products visually. We had to teach him to edit a JSON, and although it cost him at first, now he does it without problems. Was the change worth it? Yes, because the website went from “doesn’t appear on Google” to “top 5 for artisan ceramics keywords in Spain”.
Migration 3: sports nutritionist website (Barcelona). WP + Divi + 14 plugins, made in 2021. Load before: 4.1 s (PSI 58). Load after: 0.7 s (PSI 99). Organic traffic 4 months later: +60%. What cost the most: nothing, it was a clean migration because the website had few pages (8). What I learned: the simpler the original website, the faster and more profitable the migration.
Migration 4: architecture studio website (Bilbao). WP + custom theme + 9 plugins, made in 2018. Load before: 2.9 s (PSI 71). Load after: 0.8 s (PSI 97). Organic traffic 3 months later: +18%. What cost the most: redoing the project portfolio, which had 60 projects with photo galleries. The JSON of each project was complex (project data, photos, plans, team). We did it in 1 week with a script.
Migration 5: online English academy website (Madrid). WP + LearnDash + 22 plugins, made in 2020. I did NOT migrate this one. The academy uses LearnDash for their courses, and migrating LearnDash to another system would have been reinventing the wheel. I optimised the WP to the max (light theme, minimum plugin, decent hosting, aggressive cache) and the website went from 4.5 s to 1.9 s. It’s WordPress, yes, but it works. The decision not to migrate was the right one, and the client understood it when I explained the cost/benefit.
Migration 6: medical practice website (Valencia). WP + custom theme + 11 plugins, made in 2017. Load before: 4.5 s (PSI 51). Load after: 1.0 s (PSI 96). Organic traffic 4 months later: +35%. What cost the most: nothing, the client gave me all the content and we took 8 business days. What we lost: the sidebar blogroll the original WP had with links to other medical sites. The client didn’t miss it (it was a 2017 vestige that nobody used).
Migration 7: winery website (Rioja). WP + WooCommerce + 19 plugins, made in 2019. Load before: 5.5 s (PSI 42). Load after: 1.2 s (PSI 95). Organic traffic 6 months later: +30%. Conversion rate: +22%. What cost the most: integration with their stock management system (which was tied to WooCommerce). I had to make a custom API that connected the new CMS with their stock system. That was 2 weeks of extra work.
Migration 8: civil service academy website (Madrid). WP + LearnDash + 28 plugins, made in 2018. I did NOT migrate this one, same as the English academy. LearnDash is too complex to replace. I optimised the WP, changed the hosting, reduced plugins, and the website went from 6.1 s to 2.2 s. It’s WordPress, but it’s the best for that case. The owner told me “I don’t care that it’s WordPress, I care that it works”.
What you gain migrating (the averaged data)
If I average the data from the 6 migrations I did do (discarding the 2 I decided not to migrate), the numbers are:
- Load speed: from 4.4 s average to 1.0 s. -77%.
- PageSpeed Insights mobile: from 55 to 97. +42 points.
- Organic traffic at 3-6 months: +35% average.
- Conversion rate (forms, calls, sales): +20% average.
These are not cherry-picked numbers from a single project. They are the averages of 6 real projects, in different sectors, with different clients. And the trend is consistent: the websites I migrate to static frameworks perform better on all the KPIs that matter to me.
Why? Because WordPress executes PHP on each request, loads jQuery, loads the WooCommerce library if you have it, loads your theme, loads your plugins, loads your shortcodes, loads your widgets. And all of that adds up, especially on mobile with 4G. A framework like Astro generates static HTML at build and serves it directly from a CDN. Your page is an HTML file with minified CSS. No database, no PHP, no execution time. The browser paints it in milliseconds.
What you LOSE migrating (and why WordPress is still valid in some cases)
But not everything is better in Astro. There are things WordPress does well and that a static framework does worse (or requires more work). I tell them to you honestly, because if I don’t tell you, you won’t believe the other part.
The first is visual content editing. In WordPress, with Gutenberg or Elementor, your client can edit a page visually: move blocks, change colours, add sections, without touching code. In Astro (or Next, or SvelteKit), if you want visual editing you need a headless CMS (Sanity, Contentful, Storyblok) that costs €0-200/month depending on the plan, or a custom CMS based on JSON that requires more client training. The question I ask myself before recommending migration is: “is your client going to want to edit the website themselves every month, or just change texts and photos 1-2 times a year?”. If the answer is the former, WordPress is still the better option. If it’s the latter, a custom CMS or a headless CMS is a better 2-3 year investment.
The second is the plugin ecosystem. In WordPress you have 60,000 plugins for almost everything: forms (Contact Form 7, WPForms), SEO (Yoast, RankMath), ecommerce (WooCommerce), membership (MemberPress), LMS (LearnDash, LifterLMS), bookings (Amelia, BookingPress), etc. In Astro you have to build or integrate each thing by hand. This makes the project more expensive (at least in time) if the client needs specific functionalities. For the law firm, it was 1 more week of work. For the ceramics shop, it was 3 days. For the civil service academy, it was a “we don’t migrate” directly because LearnDash has no reasonable equivalent in Astro.
The third is the client’s learning curve. WordPress has 20 years of tutorials on YouTube, forums, local agencies that know it. If your client has a problem and you’re not available, there are 1,000 people in Madrid who can help them with WordPress. With Astro, the people who can help you are many fewer, and more expensive. This is a real risk that should not be ignored, especially with small clients who don’t have budget for continuous professional maintenance.
My final recommendation (the one I apply to my clients)
After 8 migrations and 3 years without making a new website in WordPress, this is the rule I apply when a client asks me for a website:
If the client wants a corporate website, a landing or a portfolio (1-15 pages, with or without blog, without ecommerce or LMS): Astro (or Next if they need more interactivity). It’s the best in 2026 by speed, SEO, maintenance and total cost.
If the client wants a small online shop (less than 500 products, without real-time stock, without complex integrations): Astro + headless commerce (Shopify Lite, Snipcart) or Astro + custom CMS with payment gateway. It works, and performs better than WooCommerce.
If the client wants a large online shop (more than 500 products, real-time stock, ERP integrations): Shopify (not WooCommerce) or stay with WordPress + WooCommerce but optimised to the max. Migrating to Astro in this case doesn’t compensate, the cost/benefit doesn’t work out.
If the client wants an LMS, a membership or an online course: WordPress + LearnDash or Moodle. There is no serious equivalent in Astro, and forcing the migration would be reinventing the wheel and charging the client 3-4 times more.
If the client wants a blog with 5+ non-technical writers: WordPress with good hosting and a modern theme. A headless CMS with Astro is possible, but the operational complexity is high and the ROI is not clear.
If the client wants a directory, a services platform or a marketplace: custom stack with Next.js + database. Neither WordPress nor pure Astro.
What I haven’t told you
There are two things I haven’t told you and that you should know if you’re thinking about migrating.
The first: migrating from WordPress to Astro costs more at the beginning than staying in WordPress. A decent migration costs between €1,500 and €4,000 for a small website, and €5,000-10,000 for a medium shop. If your current website works, doesn’t rank badly, and you don’t have security problems, migrating doesn’t compensate. Stay in WP and optimise it. You save the money in maintenance, not in migration.
The second: post-migration SEO takes between 3 and 6 months to recover and start going up. There’s a “valley” of 4-8 weeks where traffic can drop 10-20% while Google re-indexes and understands the change. If you depend 100% on SEO for your revenue, plan the migration at a time of lower seasonality and not in your best month.
If after reading all this you want me to look at your specific case, write to me at landinowebs@gmail.com or via WhatsApp and I’ll give you an honest diagnosis in 30 minutes. If it’s worth migrating, I’ll tell you. If it’s worth staying in WP and optimising, I’ll also tell you. My reputation depends on that more than on closing projects.