DatoCMS Glossary

AEO is GEO's close cousin, focused on getting your content picked as the direct answer to a question, whether that lands as a featured snippet, a voice assistant reply, or an AI chatbot response. It rewards content that's concise, well structured, and written the way people actually ask things.

AI crawlers are bots like GPTBot, ClaudeBot, and PerplexityBot that fetch web content to train models or to ground real-time AI answers. Same as search-engine crawlers, you steer them with robots.txt and Content Signals, deciding what they can read, use for answers, or train on.

AMP stands for Accelerated Mobile Pages, a technology that creates stripped-down versions of web pages to load blazingly fast on mobile devices. It's all about speed and optimizing for mobile browsing, ensuring users get information swiftly and smoothly.

API, short for Application Programming Interface, allows different software applications to communicate and interact with each other. It's a set of rules and protocols that enables one piece of software to access and use the services or data provided by another, making it possible for apps, websites, or systems to work together seamlessly. It's the behind-the-scenes magic that helps diverse technologies play nice and collaborate effectively.

API-first means you design your content and services to be consumed through APIs from day one, rather than bolting an API onto an existing monolith. It's a core MACH principle, and the payoff is simple: when everything speaks API, your stack is far easier to connect, extend, and future-proof.

API tokens are like VIP passes granting users or services access to interact with an API. They're unique keys or strings of characters that serve as authentication—think of them as digital IDs. It's a secure way to verify their identity and permissions, allowing them to make requests and receive responses from the API without compromising sensitive information that they shouldn't have access to.

An asset is a file within DatoCMS - this can include images, audio files, videos, and other documents. Dato's Media Area is where you interact with your assets to add, organize, localize, and manage them. DatoCMS also offers an inbuilt image editor and asset metadata manager to make it easier for content editors to work with assets.

Audit logs keep track of every action and event happening within DatoCMS. They capture a chronological record of all actions within a project, helping to track changes, identify issues, and ensure security by providing a trail of actions taken by users or processes.

Blocks are unique to DatoCMS, allowing to define complex and repeatable structures that can be embedded inside records. Blocks can be embedded into Structured Text and Modular Content fields to create highly customized and dynamic layouts.

Build triggers are action starters in the world of continuous integration and deployment. They're events or conditions that kick off the process of building, testing, and deploying software or websites. DatoCMS natively offers integrations to set up deploy triggers to Netlify, Vercel and GitlabCI.

The Content Delivery API playground is where you can test your queries against DatoCMS's GraphQPI API before moving them into your repo. With inbuilt docs and a GraphQL explorer for all your content records, the CDA playground ensures that your queries are interacting with the right content in a sandbox environment before you query your project on production.

A CDN, or Content Delivery Network is a global network of serversto store and deliver web content like images, videos, and web pages from multiple locations worldwide. When you access a website using a CDN, it serves you data from the server closest to your location. This speed boost reduces load times and ensures a smoother browsing experience for users across the globe.

A CMS, or Content Management System, is the control center for your content for websites, eCommerce stores, knowledge bases, and other digital platforms. It's a software that allows creating, editing, organizing, and publishing digital content.

DatoCMS's collaboration tools let manage teamwork and ensure that no data is lost when switching between users. With features like Content Stages, Presence Indicators, Record Locking, Workflows, and more, DatoCMS encourages teams to work together without losing time and effort on manual back-and-forth on creating content.

"Composable" software is a growing approach to companies building their ideal tech stacks using modular APIs and microservies rather than buying all-in-one suites. Related to the MACH (Microservices, API-first, Cloud-native, and Headless) approach, going composable allows companies to work with software that's flexible, scalable, and efficient to their particular business needs.

Content as a Service delivers your content through APIs so it can be reused across any channel, from your website to an app to a kiosk to a smartwatch, all from one source. Instead of trapping content inside one platform's templates, CaaS treats it as structured data you can send anywhere.

The Content Delivery API, or CDA, is used to retrieve content from your DatoCMS projects and deliver it to your web or mobile projects. Our CDA is written in GraphQL, and all content is served via a robust CDN with multiple datacenters around the world to ensure minimal latency.

Content governance is the set of rules, roles, and workflows that keep content accurate, consistent, and on-brand as teams and volume grow. It covers who can create and publish what, how content gets reviewed, and how standards are enforced: the guardrails that stop a big content operation descending into chaos.

Content Link is a feature supported on deployments to Vercel, using Vercel's site previews. Content on your frontend previews corresponding to text, structured text, or the `alt` field of any asset fields within a DatoCMS project get a `Open in DatoCMS` link, allowing content editors to quickly made changes to content without needing to know where to navigate to.

The Content Management API, or CMA, is used to manage the content of your DatoCMS projects. This includes creating, updating, deleting, and fetching content of your projects, featuring 40+ resources and 150+ endpoints.

Content modeling is the process of designing the skeleton for your website's content. It involves defining the various content types, their attributes, relationships, and how they'll be organized.

Content Signals are directives you add to robots.txt to tell AI crawlers how your content can be used: for search, as input to AI answers, or for training models. They give you a machine-readable way to set your terms, so you can welcome the citations while drawing a hard line exactly where you want one.

A content type (sometimes called a model) is the blueprint for a particular kind of content, like a blog post, a product, or an author, including all its fields and rules. Define it once and every record you create follows the same predictable shape.

Content versioning keeps a full history of changes to your records, so you can see what changed, who changed it, and roll back whenever you need to. It's a safety net for content teams: experiment freely, because a previous version is always one click away.

A content workflow is the defined path content travels from draft to published, through stages like review and approval, usually with roles controlling who can do what. It keeps bigger teams organized and makes sure nothing goes live before it's actually ready.

Core Web Vitals are Google's set of user-experience metrics (loading, interactivity, and visual stability) that measure how a page actually feels to real visitors. They're also a ranking signal, which makes performance a direct SEO concern rather than a nice-to-have.

A DAM system stores, organizes, and delivers your digital assets like images, video, and documents, with metadata, search, and access controls. In a modern stack it keeps your media tidy and reusable across every channel, usually with on-the-fly transformations baked in.

Data residency is about where your content is physically stored and processed, meaning which country or region the servers actually live in. It matters for performance, but mostly for compliance, since rules like GDPR can require certain data to stay inside specific borders.

A decoupled CMS splits content management from presentation, but unlike a purely headless CMS it often still ships a built-in front end you can use if you want. You get the flexibility of an API-driven backend with a ready-made delivery layer within arm's reach.

A DXP, or a Digital Experience Platform, is a category of software aiming to tackle the primary goal of perfecting the customer experience, or CX. DXPs can be a full suite, or a composable stack of complementary APIs brought together to create a modular DXP.

Edge rendering runs your page logic on servers physically close to your visitors, out at the edge of the network instead of one central box. You get dynamic, personalized content delivered at nearly the speed of a static file, because the work happens milliseconds away from the user.

Embeddings are numerical representations of text or images that capture their meaning, so machines can measure how similar two pieces of content are. They're the engine behind semantic search and RAG: turn your content into vectors and an AI can find the most relevant record even when the exact words don't match.

Environments are like different playgrounds for testing and deploying changes to the schema within the CMS. The primary environment is for editors to author records. Ideally, developers should not make changes to its schema directly, but fork it, test the changes on sandbox environments, and then merge the changes via migration scripts.

Each content model consists of multiple field types - within DatoCMS the fields available are Single-line strings, Multiple-paragraph text, Modular content, Structured text, Asset galleries, Single assets, Videos, Date and DateTime, Integers, Booleans, Geolocations, Colors, SEO meta tags, Slugs, Links (relations), and JSON.

Validations are a strong tool to enforce content structure and integrity by enforcing certain restrictions on what content can be created. This can be something as simple as ensuing entries are localized, to something more complex like having a rule for content entries to match a specific regex pattern. Each field type in DatoCMS offers varied applicable validations.

A focal point marks the most important part of an image, so automatic crops for different aspect ratios keep the subject in frame. Set it once in your CMS and every thumbnail and hero crop stays sensible, with no more accidentally decapitated headshots.

GEO is how you get your content surfaced and quoted by AI answer engines like ChatGPT, Perplexity, and Google's AI Overviews. Classic SEO chases blue links. GEO is about being the source the model actually cites, which comes down to clean, structured, machine-readable content an LLM can parse and trust.

GraphQL is a modern query language and a runtime for APIs, widely seen as a successor to REST APIs. It's built around the concept of "getting exactly what you asked for", without any under fetching or over fetching of data. GraphQL also makes it easier to aggregate data from multiple sources, allowing you to source from multiple locations within a single query, rather than fiddling with several endpoints.

A headless CMS is a backend-only CMS that provides an API to make content accessible to any platform or digital channel. Unlike a traditional CMS such as WordPress or Drupal, a headless CMS does not dictate where or how content is shown - but rather requires teams to build their own custom frontend (or presentation layers) using their preferred frameworks like React, Angular, and Vue.

Period in which bursts of changes made to the same record by the same user (or API token) will be grouped into a single revision version. Versions created in the last 7 days do not get grouped.

DatoCMS retains all content history changes for a specific amount of time, to refer back to as required when making changes or reverting content to a previous state. The content retention period is defined by which plan you're on.

A hybrid CMS mixes the API-first flexibility of headless with the convenience of built-in presentation tools, so teams can pick per project. Developers get clean APIs for custom frontends, marketers keep their familiar previews and page building, and nobody has to pick a side.

Hydration is the step where JavaScript attaches behavior to server-rendered HTML once it reaches the browser, turning a static page into an interactive one. Modern tricks like partial or lazy hydration do this selectively to keep things fast, rather than booting up everything at once.

Image optimization is the art of shipping the smallest possible image that still looks great, through compression, modern formats, and correct sizing. Since images are usually the heaviest thing on a page, getting this right is often the fastest route to a faster site.

The Images API within DatoCMS is built upon Imgix, providing a best-in-class API for image processing along with a global image CDN. The Images API allows for powerful features out of the box, such as asset localization, image editing, watermarketing, and image optimization.

Internationalization is designing your content and systems up front so they can support multiple languages and regions without a rebuild later. Nail i18n and localization becomes the easy part, because the structure is already there and you're just filling it in.

Islands architecture ships a mostly static HTML page and only hydrates the interactive bits, the islands, with JavaScript. Instead of loading a heavy framework for the whole page, you keep it light and only pay for interactivity where you actually need it.

ISR is the happy middle ground between static and server rendering. Pages are served statically for speed, but quietly regenerated in the background on a schedule or on demand, so you get CDN-fast delivery without rebuilding the whole site every time one record changes.

Coined by Netlify, Jamstack is a modern approach for building websites or web apps. It stands for JavaScript, APIs, and Markup. Instead of relying on traditional server-side rendering, it pre-builds the website and serves it through a CDN as flattened HTML. This approach boosts speed, security, and scalability by separating the front-end presentation from the back-end.

Lazy loading holds off on loading images and other heavy assets until they're about to scroll into view, so the initial page loads faster and lighter. Visitors only download what they actually see, which is kinder to both load times and data plans.

Links, also referred to as Relations or References, are bi-directional links between multiple content records and types. A common use case could be seeing which blog posts are related to a single author, or which landing pages match a certain category.

llms.txt is a simple Markdown file you drop at the root of your site to hand AI models a clean, curated map of your best content. Think robots.txt, but for large language models: instead of making crawlers dig through your nav and markup, you give them a tidy index of what matters and where to find it.

Locale fallback is what kicks in when content isn't available in a visitor's language: the system serves a default locale instead of showing them nothing. It means every reader gets a complete page, even while some translations are still catching up.

Localization is the work of adapting your content for a specific locale, translating copy but also adjusting dates, currencies, imagery, and tone so it feels native to that audience. It's the "making it local" half of going global.

MACH stands for Microservices, API-first, Cloud-native, and Headless—a set of principles shaping modern digital architectures. As a modern approach to composable enterprise software strategy based on smaller solutions that seamlessly integrate with one another, it is a base for a modern enterprise approach to keep their stack pluggable, scalable, and replaceable.

Activating maintenance mode in DatoCMS means making the primary environment read-only. This is particualrly useful when larger changes need to be made to the primary environment (schema migrations from sandbox environments, etc.), in order to ensure no content changes or updates are done to the CMS project until the operation is complete.

MCP is an open standard that lets AI assistants plug into external tools and data through one consistent interface. Instead of every app inventing its own integration, an MCP server exposes its capabilities so agents like Claude or ChatGPT can read, query, and act on your content directly. We run our own remote MCP server, so you can manage DatoCMS straight from your favorite AI client.

The Media Area is where you interact with all your files and assets in DatoCMS. DatoCMS accepts all formats for images, videos, audio, and documents, making it simpler to manage all your content from a single place. Aside from Digital Asset Management (DAM), DatoCMS also allows plugins into the media area such as Unsplash for stock images, or Dall-E for AI image generation.

DatoCMS allows you to programmatically add models and records using migrations, either via writing scripts manually, or by having the DatoCMS CLI automatically generate migrations for you.

WebP and AVIF are next-gen image formats that hit the same visual quality as JPEG or PNG at a fraction of the file size. Serving them automatically, with older formats as a fallback, is one of the easiest wins going for faster pages and happier Core Web Vitals.

Modular content fields are used to define dynamic areas for customized page layouts, rather than restricting content editors to use rigid templates. For example, when building a blog post, the content editor can add blocks such as CTAs, quotes, or videos into the article depending on which blocks are accepted into the article field.

Omnichannel means delivering consistent content and experiences across every channel a customer touches: web, mobile, email, in-store, voice, the lot. A headless, API-driven CMS is what makes it doable, because you model your content once and publish it everywhere without copy-pasting.

Projects on DatoCMS can be tied to an organization. Particularly useful for teams or agencies handling multiple projects or clients, an organization allows for better visibility and shared ownership between DatoCMS projects. Similar to personal accounts, organizations can also upgrade to paid plans and manage billing information.

Per-locale publishing involves releasing or making content available based on specific regions or languages. Content is published or displayed selectively, targeting particular locales or language groups. This approach ensures that the right content reaches the right users in their preferred language or region, without making unnecessary changes to all content unintentionally.

Plugins allow you to easily expand and customize the capabilities of DatoCMS. With popular plugins for services like SEO readability, cloud deployments, computed fields, Shopify products, and more, plugins are React applications communicating with your DatoCMS project to enable custom functionalities and workflows, as well as to interact with external APIs without leaving your project.

A project starter is an out-of-the-box DatoCMS project built with a complete CMS configuration, corresponding frontend, and integrations to popular deployment platforms like Netlify and Vercel. They help users rapidly set up common projects (like blogs or websites) and get familiar with best practices when working with a Headless CMS.

RAG makes AI answers more accurate by pulling relevant, up-to-date content at query time and feeding it to the model as context, instead of relying only on what the model memorized during training. For content teams that's the whole pitch: a well structured CMS becomes the trusted knowledge base an AI grounds its answers in.

The Real-time Updates API (sometimes refered to as a Subscriptons API) allows client to listen for content changes via a stable connection to stream events as they occur. This is particularly useful for real-time content refreshing for use-cases like live sports updates or stock prices, when new content needs to appear without a user having to refresh the page.

A record is any content entry you make within your DatoCMS project based on your content models and schema - think of it as a table row in a database. This can include anything from a blog post, to a landing page or a hosted video.

Record information are system records for each content entry within a DatoCMS project that can be queried via the CDA - this includes metadata such as the Record ID, publishing status, timestamps for creation, and information on publishing.

A reference field links one record to another, connecting a blog post to its author or a product to its category, so you model real relationships instead of duplicating data. It's how structured content stays connected and DRY: define a record once and reuse it wherever you need it.

Responsive images serve the right image size and format for each device and screen, so phones aren't downloading desktop-sized files. Delivered through an images API, they're one of the highest-impact things you can do for page speed and Core Web Vitals.

Rich text is formatted content like headings, bold, links, and embedded media, created in a friendly editor without touching code. The good implementations store it as clean, structured data rather than raw HTML, so the same content renders beautifully on any platform.

Each user and API Token can be assigned specific roles within DatoCMS, with a custom and pre-defined set of permissions they're allowed to perform. These can range anywhere from being able to create content in a specific locale, moving content between a specific stage, or reading content from a specific environment.

Scheduled publishing is a feature in DatoCMS that allows you to plan when content goes live (or unpublished) on your website or platform. Instead of immediately publishing, you can schedule articles, updates, or posts to appear at a specified date and time. It's a handy tool for content creators and marketers, ensuring timely and organized releases without manual intervention.

Schema refers to the structure that defines the organization and format of content within DatoCMS or any other CMS. It's like the framework that outlines how data should be arranged (content records under content models), including the types of data (fields), their relationships, and constraints.

Semantic search actually understands the intent behind a query instead of just matching keywords. Powered by embeddings and vector search, it returns results that mean the same thing as what you asked, so it's far more forgiving of natural, conversational phrasing than old-school keyword search.

Many website use-cases call for single-use templates or content types, such as homepages, about-us pages, or navigations. Single-instance models are useful for content that isn't meant to be duplicated or exist across multiple places. When creating a new model, simply select `Single Instance` within DatoCMS to enable this for content editors.

A single source of truth is one central, authoritative place where each piece of content lives, so every channel pulls from the same well-governed data. It kills the drift and duplication you get when the same info is maintained across five different systems.

DatoCMS offers an in-built approach to implementing Site Search results to your website visitors. With minimal configuration needed, new content is crawled and fetched with every new deploy. This can be an ideal approach for most websites, however for advanced search use-cases, DatoCMS offers plugins and integrations to popular services like Algolia and Cludo.

A slug is the human-readable, URL-friendly ID for a piece of content, like "my-first-post" in yoursite.com/blog/my-first-post. Keep them short, descriptive, and stable, because good slugs are better for readers, better for SEO, and better for anyone trying to make sense of your links.

Static Site Generation pre-builds your pages into plain HTML at build time, so they can be served instantly from a CDN with no per-request server work. Fast, secure, cheap to scale. The catch is that content changes need a rebuild, which is exactly where a headless CMS and build triggers earn their keep.

Single Sign-On, or SSO, is a user authentication process that allows individuals to access multiple applications or systems with just one set of login credentials, using common SSO services like Okta and OneLogin. SSO is currently an enterprise feature on DatoCMS.

Server-Side Rendering builds each page on the server the moment it's requested, so visitors always get fresh, personalized content. Great for pages that change constantly or depend on the user, at the cost of a bit more work than serving a pre-built static file.

Structured text in DatoCMS is a field type that allows content editors to create rich content entries with a powerful editor allowing for `/` slash commands to embed other blocks like images, galleries, and embeds. The content is stored in a semantic JSON format, making it simple to query.

Taxonomy is how you classify and organize content with categories, tags, and hierarchies, so it's easy to browse, filter, and relate. A good taxonomy is what turns a pile of records into a navigable, connected content library.

Tree-like models enable for common use cases where content needs to exist in a hierarchial structure. Common use-cases for this include examples like eCommerce categories, taxonomies, and product navigation, allowing content creators to manage content in a tree-like data structure. DatoCMS lets you create these by enabling `Records can be organized in a tree` for content editors.

Vector search finds content by meaning instead of exact keywords, comparing the embeddings of your query against the embeddings of your content. Ask for "ways to speed up my site" and it can surface an article called "Improving performance" with zero shared words. It's what powers modern AI search and recommendations.

DatoCMS's Video API, powered by Mux, provides powerful features such as adaptive bitrate, global caching, and thumbnails, ensuring videos are loaded and streamed exceptionally quickly with the right format for each device, on any player.

Webhooks deliver real-time notifications or data to other applications or systems whenever a predefined event occurs. Instead of constantly asking for updates, webhooks wait for triggers, such as a new order or a form submission, and then immediately send that data to a predefined URL or endpoint. They're a way for different systems to communicate efficiently, facilitating timely actions based on events.