Squarespace to a headless CMS
when growth means duplicating pages.
We move Squarespace sites onto DatoCMS or Storyblok with a real codebase underneath. Your design comes with you and your URLs stay where they are. What changes is underneath the pages: content your team reuses instead of duplicates, and custom code that someone actually stands behind.
Trusted by
Trusted by
On Squarespace, the page is the unit.
Squarespace builds a site out of pages, and that is the whole design. It is why the first version goes up in a weekend and looks good doing it. It stops being free the day the same block has been pasted into fifteen pages and a second language means duplicating all of them. Squarespace's multilingual guide says as much: duplicate the pages from your first language, then update the text. That is fine at ten pages. At a hundred and forty it is a maintenance job with a marketing team attached. This is one path through our wider headless migration practice. Teams leaving Wix and Webflow run into their own version of this. The ceiling sits in a different place on each of them.
A second language means a second set of pages. Squarespace's own guide suggests duplicating the pages from your first language and updating the text.
Language subdomains are not available. Squarespace states you cannot have different domains or subdomains for different languages.
Built-in text stays in one language whatever you do. Squarespace says this also affects metadata across the site, like dates on blog pages.
Code injection is the escape hatch and Squarespace says plainly that custom code falls outside the scope of their support, and can cause display issues with future platform updates.
The same block gets pasted into fifteen pages, because the page is where content lives and there is nothing underneath it to reuse.
The export takes one blog page and leaves the rest. Everything else has to be read off the live site.
The Reality
The export is real, and it is partial. Squarespace documents exactly one blog page coming out with its posts, while album, cover, index, info, calendar, portfolio and store pages do not. Getting the rest out is the actual project.
What You Take With You
One blog page exports. We crawl the rest.
A Squarespace migration is mostly a reconstruction job, and knowing which half is which is what keeps it predictable. The export covers a slice, the crawl covers everything else, and both are scoped before anyone writes code. What your visitors see stays the same throughout.
Your blog
comes out through the export, which carries one blog page with all of its posts. A second blog page has to come across through the crawl, and most growing sites have one.
The pages Squarespace will not export
are named in their own documentation: album, cover, index, info, calendar, portfolio and store pages. Those get reconstructed from the live site, which is where the audit starts.
Your visual design
read off the live site and rebuilt in code. Style settings and custom CSS both sit on Squarespace's no-export list, so the rendered page is the only specification of your design that exists.
All your URLs
every page slug, every blog post, every gallery item. Squarespace generates a good number of these for you, so the redirect map gets built from the crawl instead of from a list someone typed.
Your media
sized today by whatever the template asked for. We pull it, resize it against the layout that actually ships, and serve WebP and AVIF.
What Changes
Duplicated pages become one component.
The page-shaped habits go away with the platform that created them. Duplication, injected code, and hand-maintained language copies are answers to how Squarespace stores a site, and they stop being necessary once the content lives somewhere with a structure of its own.
A second language
stops being a second set of pages. Locales live in the CMS against the same entries, and each one gets a real URL you choose.
Repeated blocks become components
that take props, live in source control, and ship everywhere they are used the moment one of them changes.
Code injection goes away
and custom behaviour becomes code in the repository, reviewed and owned by the people who wrote it.
Hosting and CDN
you choose. Bunny, Cloudflare and Vercel are where most clients land, and a static build takes away the page-speed ceiling that arrived with the template.
Lock-in ends
because the content sits in a CMS with an API. Changing agency, host, or CMS later becomes a decision you make on your own timeline.

After The Move
What your team opens after the move.
The destination is a stack: content in DatoCMS or Storyblok, pages in Astro or Next.js, and hosting you choose. On Squarespace the page holds the content. After the move, the content lives in entries and each page is built from them, and that is most of what changes for the person editing. What converts into the CMS during the move is on the headless migration page.
A page is an entry made of blocks
We build each page in Storyblok as a story whose body is a blocks field, and in DatoCMS as a record whose body is a Modular Content field. In both, the person editing adds, removes and reorders blocks to build the page, and each block is a few fields to fill in.
Each field offers its own set of blocks
A landing page can offer a hero, a text section and a call to action, while a blog post offers text, images and quotes. DatoCMS sets this in the field's Validations tab, and Storyblok per blocks field. Where each block sits and how it reflows on a phone comes from its component in the code.
Repeated content becomes one entry
Content that repeats, like a promotion or an opening-hours note, is stored once and linked from every page that shows it. Storyblok does this with a references field that links to other stories, DatoCMS with link fields that point to other records. Whoever edits the site changes it in one place.
Edits wait in a preview
DatoCMS shows the site next to the editing panel through Visual Editing and its Web Previews plugin, where clicking an element opens the record and field behind it. Storyblok's Visual Editor does the same with a draft rendering of the site next to the form, and a click in the preview opens the block. In Storyblok, and in DatoCMS on models with its draft and published system switched on, a change to a published page stays out of sight for visitors until someone publishes it again.
A rebuild puts changes live
On a static build, a published change reaches visitors once the site rebuilds. Storyblok sends a webhook when a story is published, and we point it at the build. In DatoCMS, publishing a record and rebuilding the site are separate steps: the build runs on the deploy service through build triggers, and editors start it with the Publish changes button in the top bar.
A new kind of block is developer work
In Storyblok the development team creates block types, and in DatoCMS developers decide which blocks writers can use. A new block needs its schema in the CMS and a component in the front end. The default Editor role covers content, assets and tags in Storyblok, and in DatoCMS it works with records and has no access to models. Building new components is part of the monthly tier.
How It Works
Rebuilding a Squarespace site as content
Crawl first, export second
We crawl the live site before we touch the export, because the crawl is what tells us how much of the site the export will miss. Gallery and portfolio pages, duplicated language pages, and anything sitting in a code injection block get listed individually. Then we design the content model, including which page types collapse into one entry type. You approve the model before any code is written.
Rebuild in code, migrate the content
We rebuild the design in Astro, faithful to the Squarespace version. The blocks that appeared on fifteen pages become one component used fifteen times. Content is imported into DatoCMS or Storyblok through their API, with each locale modelled as a locale. We test every page against the live Squarespace site before cutover.
Cutover and 30 days of watching
We launch on a low-traffic window, swap DNS, and monitor Search Console, Analytics, and Core Web Vitals for 30 days. Squarespace stays live as a fallback during the watch window. After 30 days you decide whether to keep us on with a monthly tier or take it from here.
A note on custom code
Squarespace is honest about code injection, and that is worth reading carefully.
Their documentation says custom code modifications fall outside the scope of their support, that they cannot guarantee functionality or full compatibility, and that custom code can cause display issues with future platform updates. Code injection also runs on the Core, Plus, Advanced and some legacy billing plans. That is a fair description of an escape hatch. It is also the reason the injection block on most sites has grown into something nobody wants to open.
Start with an auditPricing
Fixed price, scoped from the crawl
Squarespace migrations are priced on how much the export leaves behind. A 30-page brand site with one blog and no second language is one thing. A 140-page site running two duplicated language trees, several portfolio pages and years of injected code is another. The ranges below are where typical projects land. Once the crawl is done, you get a fixed quote.
Full crawl and audit of every page type the export will miss
Content model that replaces duplicated pages with reusable entries
Redirect map for every page, post and gallery item
Faithful rebuild in Astro
Frontend in Astro, CMS in DatoCMS or Storyblok
Locales modelled properly, with a real URL per language
30 days of post-launch monitoring with Squarespace as fallback
Momentum, € 2.490: new pages, sections and components in your CMS, and small changes live the same working day
Ownership, € 4.990: everything in Momentum, plus new integrations built between your CMS, CRM and analytics
All prices are net, excluding applicable VAT.
Who This Is For
Who this is for
This works well if you
Run a second language as a duplicated set of pages and maintain both by hand
Keep pasting the same block into new pages because there is nothing to reuse
Have a code injection block that nobody on the team wants to touch
Care about organic traffic and need every URL preserved through the move
Want the same content to feed more than the website, like an app or a partner feed
Want to own your hosting rather than depend on a single SaaS vendor
This is not a fit if you
Run Squarespace Commerce as the core of the business and need a commerce-first replatform. Different scope. Talk to us first
Run a marketing site that changes a few times a year. Squarespace is doing its job and a migration is overkill
Want a designer-first workflow with no developer involvement. That is what Squarespace is good at
Are not sure yet what is wrong. Start with a Headless Audit
Need a new build from scratch with no legacy to preserve. See Headless Website