CMS guideOpen-source headless CMS

Payload CMS: who it suits, and how to migrate to it or away from it

Payload is an open-source, TypeScript-first CMS configured entirely in code. It can run inside a Next.js app and gives developers direct control over content, access rules and the admin panel.

Who uses Payload

Development teams and agencies building with Next.js and React choose Payload when they want to own the stack. It appeals to developers who prefer configuration in code over a hosted dashboard.

How Payload works

Payload is configured in code. You define collections, fields and access rules in TypeScript, and Payload provides the admin panel and APIs from that config. It can run inside a Next.js app and works with a database you choose.

A Strapi article, written into Payload

Strapi Payload
A Strapi article, written into Payload
StrapibecomesPayload
titletitle
content (blocks or Markdown)content, as Lexical rich text
coverfeaturedImage, uploaded
category and tagscategories and tags, linked
authorauthor, linked
publishedAtpublishedAt
slugslug, and a row in redirects.csv
Components and dynamic zonesgroups, arrays and blocks of the same name, with their uploads
Relationsrelationships to the same records, once they are written
Where each part of a Strapi article goes in a typical Payload posts collection. The Map step shows the real fields of both projects, and you confirm each one.

Payload migration routes

Reloque moves Payload content in both directions, and between two Payload sites. Reloque reads your collections, uploads and relationships with a user's API key, from Payload 1, 2 or 3, with blocks, arrays and groups read as structured content, not flattened. A localized site is read in every locale, and its globals (site settings, header, footer) go to the new site's globals or single types of the same name. A header's or footer's links become menus on WordPress, and the redirects plugin's redirects move too, each to its page's new address. Payload has no front end, so you can give the public site's address when connecting, and links to it in the content are pointed at the migrated records. Each route is mapped, previewed on the destination and checked by reading every record back.

Payload to WordPress

Posts follow the site's permalink setting, and with the Redirection plugin installed each old address redirects to its new one, along with the old site's own redirects. Media lands in the Media Library with its alt text, featured images, categories and tags are linked, with categories under their parents, and authors are matched to existing users. Every category and tag comes over, including those no post uses, and so does every term of your own taxonomies. Members can move too, as subscribers, without their passwords. The site's title, tagline, logo and timezone are set where the new site has none yet. Menus arrive as WordPress menus, at the theme location they had, with each link pointed at the moved page. Comments arrive with their replies, under each commenter's own name and email, at their own dates, approved or waiting for a moderator as they were. SEO data goes to Yoast SEO or Rank Math, whichever the new site runs, its social images copied to the Media Library. The WordPress guide covers the other side. Payload to WordPress, step by step.

Payload to Contentful

Records go into the space's own content types, matched by name, or into Blog post and Page types Reloque adds when the space has none and your role allows it. Bodies become Rich Text, with images uploaded as assets and links between records as links between entries. Categories, tags and authors become entries of the types the space links to, or Contentful tags. Each language goes into the locale of the same language, in the fields that are localized. SEO data goes into the type's SEO fields or an SEO entry, menus into menu and menu item entries and redirects into redirect entries, with types added for them where the space has none. Published records are published. The Contentful guide covers the other side. Payload to Contentful, step by step.

Payload to Sanity

Records go into the dataset's own document types, matched by name, writing the fields their documents have. A dataset with no posts or pages gets documents shaped as Sanity's blog starter (post, page, category, tag and author), which the Studio shows once its schema has those types. Bodies become Portable Text, with images uploaded as assets and links between records as references. Categories, tags and authors become documents of the types the dataset refers to. Each language becomes a document of its own, tied to the others as document internationalisation does. SEO data goes into an SEO object, menus into navigation documents and redirects into redirect documents. Drafts stay drafts, and published records are published. The Sanity guide covers the other side. Payload to Sanity, step by step.

Payload to Webflow

Records go into the site's own collections, matched by name, or into the collections of Webflow's blog template, which Reloque adds when the site has none: Blog Posts, Pages, Categories, Tags and Authors. Bodies stay rich text, with images uploaded as assets and links between records pointing at the new items. Categories, tags and authors become items of their collections, and SEO data goes into the collection's SEO fields. Published records are published, and drafts stay drafts. The Webflow guide covers the other side. Payload to Webflow, step by step.

Payload to Strapi

Bodies are written as Strapi blocks or Markdown, whichever each field is, images go to the Media Library, and cover images, categories, tags and authors are linked, categories under their parents where they have a parent relation. Strapi 5 sets its own publish date, so the original goes into a date field of your own. A site in several languages moves into each locale of a translated type. Members can move too, as Users & Permissions users in the role of their group, without their passwords. With a Redirect type of your own (from and to fields), each old address redirects to its new one; with strapi-plugin-redirects, the handover package has a file to import in its admin page. Menus from another CMS fill a single type's list of links, a main menu the header's and a footer the footer's. The Strapi guide covers the other side. Payload to Strapi, step by step.

Payload to Ghost

Bodies are sent as HTML, which Ghost turns into its own editor format. Categories become tags, posts keep their slugs at /<slug>/, and the redirects are added to Ghost's own when you connect with a staff access token (Ghost 5 and later); otherwise the handover package includes a redirects.yaml ready to upload. Comments arrive on their posts at their own dates, each by the member with the commenter's email: you choose on the Map step whether to add a free member for each commenter who is not one, signed up to no newsletter and sent no email. The Ghost guide covers the other side. Payload to Ghost, step by step.

Payload to Drupal

Each post and page becomes a node at its old address, as an article, a basic page or the content type you choose, with its fields in each field type's format. Categories and tags go into their vocabularies, nested where they were, images into the Media Library, and authors are matched to the site's users. SEO data from another CMS goes into Metatag's field, with sharing images copied. Menus are added to Main navigation, Footer or a menu of the same name, as JSON:API cannot create menus. Comments arrive with their replies, each by the site's user with the commenter's email or as a visitor. Members become users with no password, who set their own with Drupal's password reset page, and get a role only where it cannot edit the site. JSON:API cannot create Paragraphs, so blocks from another CMS are added to the body. A site in several languages comes in its default language, or each language as a node of its own, as JSON:API cannot add a translation; you choose on the Map step. With the Redirect module installed, each page's old address redirects to its new one, and the old site's own redirects come too. Drupal's JSON:API cannot change site settings, so the run report lists the site name, slogan and time zone to set in Drupal's admin. The Drupal guide covers the other side. Payload to Drupal, step by step.

Payload to Umbraco

Each post and page becomes a document, under a document type Reloque creates with the shared fields when yours does not exist. Umbraco sets its own publish date, so the original goes into a Publish date field. A site in several languages moves in every language, and a language the new site lacks is added. Dictionary items, the labels your templates show, move in every language too, and each document keeps its domains and hostnames. SEO data from another CMS fills the document type's SEO properties, such as seoTitle, metaDescription, noIndex and ogImage, with sharing images copied, and what the type has no property for is named in the run report. The Umbraco guide covers the other side. Payload to Umbraco, step by step.

Payload to Storyblok

Records go into the space's own content types, matched by name, or into Article and Page types Reloque adds, with categories and authors, when the space has none and your role allows it. Posts go into the folder their type's stories are in, or a Blog folder, and pages to the root. Bodies become rich text, with images uploaded as assets and links between records as links between stories. Categories and authors become stories in folders of their own, and tags Storyblok's own tags. Each language goes into the space's language of the same code, in the fields that are translatable. SEO data goes into the type's SEO fields, menus into a config story's navigation and redirects into redirect stories. Drafts stay drafts, and published records are published. The Storyblok guide covers the other side. Payload to Storyblok, step by step.

Payload to Payload

Moving between Payload projects or versions, such as Payload 2 on MongoDB into Payload 3 on Postgres, carries records, uploads and relationships across, with rich text rewritten for the destination's editor. Payload to Payload, step by step.

Moving to Payload

Moving to Payload means writing collections for your content model. Because the config is code, plan the mapping from your current fields and relationships before you import anything.

WordPress to Payload

Reloque reads posts, pages, your own post types, media, categories, tags, your own taxonomies, authors, menus, comments, SEO data, the Redirection plugin's redirects and custom fields through the REST API, an export file or the bridge file. The REST API shows only the fields a plugin registers, such as ACF's; the export file and the bridge read every one. SEO data comes from Yoast SEO or Rank Math: the SEO title and description you set, the focus keyword, the canonical address, noindex and nofollow, and the social title, description and image, but not what the plugin makes from its templates. More on WordPress to Payload.

Contentful to Payload

Reloque reads the space through its Content Management API, in every status and every locale. Contentful has no built-in posts or pages, so content types are told apart by name and fields: a type for posts, one for pages, your own types as types of their own, authors, and categories and tags, with Contentful's own tags too. Assets come as media. Rich Text keeps its headings, lists, links, formatting and images, and links between entries point at the moved records. SEO fields, or an SEO entry an entry links to, reach the other CMS's SEO fields, and menu and redirect entries come as menus and redirects. More on Contentful to Payload.

Sanity to Payload

Reloque reads the dataset through Sanity's API, drafts included. Sanity keeps its schema in the Studio's code, so document types and their fields are read from the documents themselves, and told apart by name and fields: a type for posts, one for pages, your own types as types of their own, authors, and categories and tags. Images and files come as media. Portable Text keeps its headings, lists, links, formatting and images, and references between documents point at the moved records. Translations made with document internationalisation come as each record's languages, an SEO object reaches the other CMS's SEO fields, and menu and redirect documents come as menus and redirects. More on Sanity to Payload.

Webflow to Payload

Reloque reads the site's collections with their fields, and every item, drafts included. Posts, your own collections, authors and categories are told apart by name and fields, references come with what they point at, and categories keep their parents. Images and files come as media, rich text keeps its headings, lists, links, formatting and images, and links between items point at the moved records. SEO fields reach the other CMS's SEO fields, and static pages come as pages with their text, images and SEO settings. More on Webflow to Payload.

Strapi to Payload

Reloque reads your content types, media and relations with an API token, or as a Users & Permissions user on Strapi 3. Rich text in blocks or Markdown is read as formatting, not as raw text. A translated type is read in every locale, each published or a draft as it is. Single types, such as the site's settings, go to the new site's single types or globals of the same name, and their lists of links become menus on WordPress. Redirects come too, from a Redirect type of your own or strapi-plugin-redirects on Strapi 4. More on Strapi to Payload.

Ghost to Payload

Reloque reads posts, pages, tags, authors, images and each one's SEO and sharing fields through the Admin API, from Ghost 3 to 6, or from an export file. The navigation and secondary navigation come too, from Ghost 5 and later or an export file, with their links pointed at the moved pages, and so do comments, with their replies, from Ghost 5 and later. Connected with a staff access token, the site's redirects come too. More on Ghost to Payload.

Drupal to Payload

Reloque reads every content type's nodes through JSON:API, or the bridge file on Drupal 6 to 8.6: articles as posts, basic pages as pages, your own types as types of their own, with their fields, Paragraphs as blocks, Media and files, every vocabulary's terms and each node's author. SEO data comes from Metatag: the tags an editor set for a node, not Metatag's defaults, so it reaches the other CMS's SEO fields. Members come too when you choose them: users who only visit the site, never your editors, with their roles and fields but never their passwords. Menus come too, with each link pointed at the moved page, and comments with their replies, visitors' names and emails, and whether each was approved. Every language comes too: each node with its translations, in their own titles, aliases and fields. So do the Redirect module's redirects, each pointed at its page's new address, and the site's name, slogan, language, logo and icon. Links between nodes point at their aliases. More on Drupal to Payload.

Umbraco to Payload

Reloque reads documents, media and each editor's values through the Management API (15 and later), the Delivery API (12 to 14) or the bridge template (7 to 13). SEO properties, such as seoTitle, metaDescription and noIndex, are read as the page's SEO data too, so they reach the other CMS's SEO fields, and Umbraco's own redirects from renamed and moved pages come with them (15 and later, or the bridge). More on Umbraco to Payload.

Storyblok to Payload

Reloque reads the space's content types from its components, and every story with its content, drafts and every language included. Posts, pages, your own content types, authors and categories are told apart by name and fields, and Storyblok's own tags come too. Components inside stories move as blocks, assets come as media, and rich text keeps its headings, lists, links, formatting and images, with links between stories pointing at the moved records. SEO fields reach the other CMS's SEO fields, a config story's navigation comes as menus, and redirect stories as redirects. More on Storyblok to Payload.

Planned routes

These connectors are planned. Each guide says what to weigh and plan in the meantime.

Ways to connect Payload

Reloque only ever reads from the source, and connects with what your Payload version allows.

API key

Payload 1, 2 and 3Source and destination

The API key of a user whose collection has API keys enabled (useAPIKey), with access to the collections to move.

Checked on Payload 3.92 (Postgres), 2.32 and 1.15 (MongoDB). The Connect form finds the version at your address and offers the right way in, with a setup guide to download.

What moves from Payload

Reloque scans your Payload site first, maps each field to the destination and previews the move on it before the live run.

Payload logoContent inventoryRead by the connector
  • Collections
  • Uploads
  • Relationships to categories, tags and authors
  • Rich text, as HTML or Lexical
  • Drafts
  • Publish dates
The scan lists what your site holds before anything moves, with anything it cannot carry named.

What Payload does well

Code-first configuration

Collections, fields and access rules are defined in TypeScript.

Fits Next.js

It can live in the same project as your front end.

You own your data

Run it on your own infrastructure with the database you choose.

Access control built in

Field-level and document-level rules are part of the config.

What to weigh before you move

No CMS fits every team. These are the points that come up most when people plan to leave Payload, or to make it their next home.

  • Developer-ownedSetup and changes need developers.
  • You run itHosting, upgrades and backups are your responsibility unless you use the cloud offering.
  • A younger ecosystemThere are fewer ready-made plugins than on long-established platforms.

What does not move yet

Better to know now than after the switch. The scan names anything it finds that cannot move.

  • Payload has no front end, so redirects assume pages at /<slug> and posts at /blog/<slug> until you change them to your routes.
  • Original publish dates are kept when the collection has a publishedAt field.
  • With GraphQL turned off, fields are learned from your existing documents, so a collection with none yet takes plain values and leaves out what Payload refuses.
  • Payload's locales are set in its config only, so add each language the old site has to localization, and mark the fields to translate localized, before you migrate.
  • Payload has no comments of its own, so comments from another CMS are listed in the handover package.
  • SEO data goes into an SEO group, your own or the SEO plugin's meta group, field by field; what the group has no field for is named in the run report.

Planning a move from Payload

Payload's content model is defined in code, which makes the starting point easy to read.

  • Review your collection and global configs
  • List relationships and fields that depend on custom code
  • Check uploads and where media is stored
  • Note drafts and versions, and decide whether history needs to move

Payload questions

What is Payload?

Payload is an open-source, TypeScript-first CMS configured entirely in code. It can run inside a Next.js app and gives developers direct control over content, access rules and the admin panel.

Who uses Payload?

Development teams and agencies building with Next.js and React choose Payload when they want to own the stack. It appeals to developers who prefer configuration in code over a hosted dashboard.

What content can be migrated from Payload?

Your collections and their records, uploads, relationships to categories, tags and authors, rich text, drafts and publish dates. Reloque scans the project first and shows what it found before anything moves.

What should I prepare before migrating from Payload?

Payload's content model is defined in code, which makes the starting point easy to read. Four checks help most: (1) review your collection and global configs; (2) list relationships and fields that depend on custom code; (3) check uploads and where media is stored; (4) note drafts and versions, and decide whether history needs to move.

Can I migrate from WordPress to Payload?

Yes. WordPress posts, pages, media, categories, tags and authors are written into your Payload collections. Bodies become Lexical rich text, images are uploaded to your upload collection, and each record keeps its old address so redirects can be generated.

Does Reloque work with Payload 2 and Payload 3?

Yes. Payload 1, 2 and 3 are all supported as sources and destinations, on MongoDB and Postgres, and mixed versions work: Payload 2 content can be moved into a Payload 3 project.

Can Reloque migrate Payload today?

Yes. Payload 1, 2 and 3 connect with a user's API key, as a source and a destination: to and from WordPress, Strapi, Ghost, Umbraco, Drupal and another Payload project. Start with a free assessment: Reloque scans the project and shows what it found before anything moves.

Free assessment

Moving Payload content?

Reserve a free assessment and we look at your Payload site with you: every field mapped and the move previewed before anything is written.

Get your free assessment
Early-access members get a free migration assessment.
  1. Connect both sitesReloque reads the source and checks it may write to the destination.
  2. Check the mappingEvery field is matched, and the uncertain ones are yours to decide.
  3. Test a sampleA few records are written to the destination, read back and removed.
  4. Move the restEvery record is written and checked, with redirects for old addresses.

More CMS guides

All guides