How to migrate from Squarespace to Payload CMS
Move Squarespace content to Payload CMS with an export plan. Map supported pages, media and URLs, then verify your published retail guidance.
- Published
- Last updated
To migrate from Squarespace to Payload CMS, export supported editorial content, transform it into your Payload collections and rebuild the frontend separately. The export supplies part of the content. Your team must acquire media files and recreate unsupported features before switching the public site.
This procedure fits retail publishers moving advice and policy content into a structured content service. Treat checkout and customer accounts as separate projects with their own acceptance checks.
Define the content move and its limits
Squarespace’s site-content export instructions describe a WordPress-format XML file. Supported content includes layout pages, one blog with its posts, text blocks and image blocks. Some other blocks export text with little structure. Gallery-page support is limited to version 7.0.
Store pages, product blocks, drafts and additional blogs are outside that export. So are style settings and custom CSS. Other excluded elements include audio or video blocks and page-specific headers, footers or sidebars. The XML is not a complete store backup.
Payload suits a team that wants reusable content fields across several delivery surfaces. For example, a retailer can separate a delivery policy’s exceptions from its main explanation and reference that policy from several guides. The team must build and maintain those relationships and their public rendering.
Retaining Squarespace is simpler when its editor meets your needs and you have one site to maintain. Moving to a headless CMS adds responsibility for the frontend and its publishing connections. A new content model needs an editorial benefit that justifies that work.
Export and map the supported records
Inventory the public site before exporting, then compare the downloaded file with that inventory. Give every page and post a source identifier. Preserve its original URL, title, publication state and any relationships you need to retain.
Record image origins and alternative text. List product references, download links and links between policy pages. Keep a separate register of content you must copy manually, including unpublished editorial work.
Keep the original site intact, retain the downloaded XML unchanged and save owned asset files separately. Capture unsupported content and the settings needed to reconstruct delivery. These copies complement the export.
Download the XML
On a published, public Squarespace site with a paid subscription, open Import & export content. Choose Export, then the WordPress icon, even though Payload is the destination. Select the primary blog if prompted and download the completed XML. A trial site cannot export through this route.
Keep the source site available while acquiring image files. An image URL in the XML does not prove that you have a portable local copy.
Map a fictional retail publisher
The following fictional transfer separates extracted values from destination work. Its collection names are examples you configure in Payload.
| Source item | Available input | Payload destination | Retained information | Rebuilding or repair |
|---|---|---|---|---|
| Delivery-policy page | Accepted XML record | policies collection | Source identifier, URL, title and approved body | Convert body formatting and model exceptions explicitly |
| Buying-guide post | Accepted blog record | guides collection | Title, body and useful source metadata | Replace inline product links with maintained product references |
| Guide image | Image reference plus acquired file | media upload collection | File, caption and recorded alternative text | Rewrite the guide’s image reference to the new upload ID |
| Policy link | Recorded source URL | Relationship or mapped public URL | Intended policy destination | Resolve through an old-to-new URL map |
| Pending policy revision | Manually retained editorial copy | Unpublished destination revision | Draft wording and approval responsibility | Re-enter and review before publication |
| Storefront styling | Design and settings inventory | New frontend | Approved visual requirements | Rebuild layouts and interactive behaviour |
Parse the actual XML and report which records your transformation accepts. Keep rejected records in a repair list with their source identifiers. A page in the inventory without a usable export record needs manual acquisition before it can enter the import manifest.
Payload’s rich-text field stores structured JSON. Convert exported markup into the format required by your chosen editor, or use an explicitly designed HTML field and renderer. Pasting XML into a rich-text field does not create usable content.
Retain the original markup privately so an editor can compare a conversion with its source. Record deliberate losses, such as an unsupported embed, before approving the replacement.
Import content and rebuild delivery
Configure the destination model before creating records, then import dependencies before the records that reference them. Use a small sample to prove the conversion and relationship mapping before processing the accepted manifest.
- Define the policies, guides and media collections. Add a unique source identifier field, publication rules and the required relationship fields.
- Acquire original media files and create upload records. Save each returned Payload ID against its source asset reference.
- Transform accepted editorial records into collection-shaped data. Repair unsupported formatting and replace source image URLs with the mapped upload IDs.
- Create records through Payload’s Local API with the collection and transformed data. The API also supports uploads from a local file path.
- Map internal references after their targets exist. Use a second update pass for links between records created in the same batch.
- Read each record back and reconcile errors against the import manifest. Look up its source identifier before a repeated creation attempt to prevent duplicates.
If you use Payload’s Import Export plugin, convert the content into supported CSV or JSON first. Configure the destination collection before importing. The plugin’s import drawer is an alternative entry point for transformed records, not an XML-to-Payload website converter.
Local API calls bypass access control by default. Set overrideAccess to false and pass the intended user where the migration must follow configured permissions. Review collection hooks before a bulk import to control downstream updates and notifications.
Rebuild what customers and editors use
Build templates for each destination content type and connect them to the intended published version. Recreate preview access so an editor can inspect a pending revision without exposing it publicly. Set up editor roles and approval responsibilities independently of the imported body text.
Reconnect commerce through its separately approved integration. Validate product identifiers and destination URLs wherever a guide recommends an item. Text import does not establish that payment processing, orders or member access work on the new frontend.
Verify the retailer’s published guidance
Compare the destination with the accepted manifest and then with the source inventory. The first comparison catches import failures. The second catches pages that never entered the manifest.
For a fictional manifest of 18 pages and 35 posts, expect 53 mapped editorial records. Any extra records need an identified purpose. Each omitted source page needs an explicit disposition, such as manual recreation or approved retirement.
Use the following checks before accepting delivery:
- Open representative long pages and every unusual block conversion. Headings and links must retain their meaning.
- Request each uploaded image from its new storage location. It must remain available independently of the source site.
- Follow every changed URL and check its intended final page. Redirects must avoid loops and unrelated destinations.
- Compare product references with the current catalogue. Each must identify the intended product or have an approved replacement.
- Check public and preview versions separately. Pending editorial wording must remain outside public delivery and shopping-answer retrieval.
Trace a delivery-policy edit through retrieval
Use a policy change with a clear expected result. Suppose the fictional retailer changes its approved dispatch guidance from one business day to two business days. Keep the pending draft’s wording different so the test can detect accidental draft exposure.
Confirm the approved value in Payload and on the public policy page. Refresh any built pages or caches through the frontend’s configured publication process. Follow the content into the feed or documents used by your shopping-answer service.
Ask a fresh dispatch question and inspect the retrieved source. Expect the approved two-business-day guidance. Repeat with an exception, such as a made-to-order item, to confirm that the conversion preserved the condition.
Remove the superseded Squarespace page and old document uploads from active retrieval sources. Search those sources for the previous dispatch statement. A correct new page does not prevent a duplicate old document from answering customers.
Withdraw the test policy and confirm that retrieval stops using it. Restore its approved state after the check. Keep approval and withdrawal controls in the migration acceptance record.
Switch sources with a recovery plan
Assign the content freeze, routing changes and integration updates to named owners. Keep the registrar and hosting access available to the person responsible for delivery. Give editors one agreed publishing system during the change window.
Capture edits made after the initial export, then import and validate them before switching traffic. Apply the redirect map with the new frontend. Switch shopping-answer retrieval to the accepted destination source in the same window.
Keep Squarespace and its assets available throughout acceptance. Retain the previous delivery and retrieval configuration. If the new site fails acceptance, route delivery back to the retained Squarespace site and restore the previous approved answer source.
Reconcile any content edited in Payload during that period before resuming source editing. The recovery route restores the retained service. It does not automatically convert new structured records back into Squarespace.
Squarespace’s XML export cannot be imported into another Squarespace site as a recovery method. Preserve the original site instead of cancelling it when the first import succeeds.
For export omissions, contact Squarespace support with the page type and sanitised export details. For conversion or relationship failures, give the Payload maintainer the source identifier, destination collection and sanitised error. Close those failures before retiring the retained source and its recovery access.
Give your store its own shopping agent
Customers ask in their own words. Your agent finds, compares and checks out from your catalogue, right on your site.