STRUCTURED CONTENT MODELING
The schema is your content infrastructure, not a byproduct of your content
Define it once. The same schema serves every brand, channel, and surface.
A platform for every role to do their best work
Technical teams
Shared schema across every brand, channel, and surface, no duplicate builds
Content relationships modeled at the schema level, not hardcoded in the frontend
Schema changes validated before they reach production
Content teams
3x faster time-to-market: new brands and markets inherit the schema, no rebuild
One shared schema, zero duplicate infrastructure
Launch campaigns and content variants without an engineering ticket
How Hygraph models content as a system, not a page
Define exactly what your content is
Every content type is a precise, typed definition, not a loose collection of fields.
Typed fields for everything
String, rich text, number, boolean, date, color, geolocation, enumerations, and assets, so every field means exactly what it says.
Up to 500 content models per project
Enough room for the most complex catalogs, sites, or product lines without hitting a ceiling.
Connect it to everything else
Content types connect to each other directly, so a change in one place is reflected everywhere it's referenced.
Bi-directional relations
Any content type links to any other, in both directions, without a separate lookup table to maintain.
GraphQL union types
Polymorphic relations let one field point to multiple, different content types.
Reusable components
- Build a layout once and reuse it across as many pages or experiences as needed.
Tailor experiences
A taxonomy classifies content once. Variants adapt it by market, audience, or channel, without duplicating models.
Change it without risk
Schema changes move through a controlled path, tested and staged before anything reaches production.
Environments
Experiment with schema changes in isolation before they touch production.
Content stages
Draft, review, published, with custom stages supported, so changes move through the right checkpoints.
Management SDK
Apply schema changes programmatically, with type safety, through the same CI/CD pipeline as everything else you ship.
Ranked #1 in customer satisfaction on G2
Explore more Hygraph capabilities
Let's talk about your project
Book an appointment with our sales team and learn how Hygraph can accelerate and scale your content operations.
What you can expect:
- A discussion to understand your use case and intended outcomes
- A vision of how Hygraph could support your team
- Next steps to deep dive on your technical, operational, and strategic questions
FAQS
You’ve got questions. We’ve got the answers.
Hygraph is a headless CMS built for structured content modeling: define content types with typed fields, connect them through bi-directional and polymorphic relationships, and reuse components across channels. Content lives as a graph of entities, not a stack of pages, and delivers anywhere through a GraphQL-native API.
A page-based CMS ties content to a specific layout, so reusing it elsewhere means duplicating or rebuilding it. Content modeling defines entities and relationships independent of any page: a product, a feature, or a review exists once and can be delivered to a website, app, or AI system without copying it.
AI systems need structured context to work reliably, not just text. Modeling content as entities, relationships, and metadata gives AI assistants and agents something queryable to reason over, instead of unstructured pages. That's structure augmenting human-authored content, not replacing the people who create and govern it.
Schema-as-Code is an out-of-the-box feature that lets you generate schema snapshots and store them in Git, like application code.
A content model is the template for a content type, such as Page, Author, or Blog Post, defining its fields and structure. A component is a reusable group of fields, such as an SEO block, that can be inserted into models. It reuses structure but not content, and components don't count as content entries.
Hygraph's schema system includes Taxonomies for category management, Enumerations for controlled vocabularies, Remote Sources for pulling in external data, Components as reusable content blocks, and Models as content types. Cross-references, variants, and localized fields are supported across schemas.
Enums are flat, fixed lists, such as Parents, Public, or Private, best for simple, filterable tags. Taxonomies are hierarchical trees, such as Continent, Country, City, Neighborhood, and Street, which fit when nested categories or rollups are needed.
You reference models rather than embed them; embeds are reserved for rich text inline blocks. For page widgets or sections, you create separate models and reference them, which is functionally similar to embedding but done through references.