Gatsby to Astro
migration

Your components and your styles come across as they are, and we go through the old site and the new one page by page, at every breakpoint, until they match. If you do want design changes while we're in there, that's a separate conversation and a separate quote.

Astro runs React, so anything that needs to be interactive stays a React component and keeps working. What changes is that those parts load on their own, instead of the whole page shipping the React runtime in order to show static text.

Contentful, WordPress, Prismic, Strapi, Sanity, Storyblok, DatoCMS, or just Markdown files in the repo. Your content doesn't move and your editors carry on as they were. The only thing we replace is the Gatsby source plugin that sits in between.

Every page keeps its address. We port the redirects, the canonical tags, the sitemap and the structured data, and check them all before switching anything over, so there shouldn't be anything for Google to relearn.

Astro builds to static files the same way Gatsby does, so wherever you deploy today you can carry on deploying there, on the same plan.

The Astro app gets built alongside the Gatsby one, in the same repo and on its own branch, so your live site stays untouched until you decide to switch. And if something surprises you afterwards, going back is one deploy.

It depends almost entirely on what's in your repo. A straightforward marketing site that stuck to Gatsby's normal patterns is usually one to two weeks from start to handover, which is elapsed time rather than hours of work. One with a lot of custom plugins, unusual data sourcing or heavy client-side code takes longer, and there's no way for us to know which of those yours is until we've read it. That's why the quote is free, and why it's fixed once we've given it.

It depends on the site and the real number comes out of the audit, but we can give you the shape of it. A single page site we'd expect to be somewhere around €200 to €300. A marketing site of ten to twenty pages, with nothing unusual in how it sources or renders its content, we'd expect between €500 and €750. Bigger or stranger than that and we honestly don't know until we've read it. Those are estimates based on the ports we've done rather than a price list, and the audit is what turns them into a number that doesn't move afterwards.

There shouldn't be any reason for them to. Your URLs stay where they are, your content and metadata are identical (or better), and the pages still ship as static HTML like they did before. The thing that does change is how much JavaScript loads on top of that, and that moves in your favour.

Sanity is our go-to and the one we know best, but the port doesn't depend on it. If Gatsby could source from it, Astro can fetch from it, usually with less machinery in between. Contentful, WordPress, Prismic, Strapi, Storyblok, DatoCMS, or Markdown in the repo all work fine. Your content stays where it is.

It depends on what the site is, rather than on which of them is better. Astro is built for sites where the content is the product: marketing pages, blogs, documentation. It ships almost no JavaScript and hosts anywhere as static files. Next.js is built for sites that behave like applications, with accounts, dashboards, and data that changes while you're looking at it. It's very good at that, and it carries the weight to match.

Your Gatsby site is already the first kind: React components that build to static pages. Astro is that same shape with less machinery in the way, which is why the port is mostly mechanical. If yours is closer to an app than a website, we'll tell you at the audit, before you've paid for anything.

Cloudflare bought the Astro team in January 2026, which is close enough to Netlify buying Gatsby in 2023 that the question deserves an answer. What happened to Gatsby is that most of the core team had left within a few months and the releases stopped not long after. The Astro team went to Cloudflare and is still shipping, and Cloudflare has said publicly that Astro stays open source.

We can't see three years ahead any better than anyone else, but there's less to be stuck with either way. Astro has no GraphQL data layer and no plugin ecosystem holding your build together, so your components and your content aren't tied to it the way they were tied to Gatsby, and moving off it later would be a much smaller job than the one on this page.

Any Gatsby version from v2 onwards, and any size, there's no minimum. What moves the price isn't the page count, it's how much your site does that Astro has no direct equivalent for. The usual one is a data layer that joins and reshapes your content before the components ever see it, since that has to be rewritten as ordinary code. Custom source plugins are another. So is interactivity that assumes the whole page is a single React tree, because in Astro the interactive parts are separate islands and anything shared between them has to be rewired. None of it stops a port, and you'll know which of it applies to you before you commit to anything.

They come across. Serverless functions belong to your host rather than to Gatsby, so usually it's little more than a path change. Forms get ported and tested end to end during the parity check, submissions included.

No, nothing they see changes.

You get 90 days of free support after the cutover, the same as on the sites we build from scratch. If the port got something wrong, we fix it and it doesn't cost you. Things that were already broken on the Gatsby site, and anything new you want adding, are a separate quote.

Lisbon, on Western European time. Our clients run from Australia to California and it hasn't caused problems, mostly because a port doesn't take much of your time: we read the repo, we ask a handful of questions, and the next thing you need to look at is the finished site. When we do need a call we'll find an hour that works for both of us.