Every few years the content management system gets reinvented, and every time the pitch sounds the same: this one finally separates content from the mess of building and shipping it. The difference between the generations is not the marketing. It is what the system actually emits, and who has to decide before anything goes live.

The first generation bundled everything into one system. WordPress, Drupal, AEM and Sitecore coupled content, templates and delivery together, which made them fast to start with and hard to change later. A marketer could publish a page without asking an engineer, as long as the page fit a template someone had already built.

The cost arrived at redesign time. Because content, presentation and delivery were tangled together, every redesign became an excavation. Teams dug through themes, plugins and hard-coded markup to find where a heading actually lived, then rebuilt the whole thing to change how it looked. The monolith solved the publishing problem of its era, and left the redesign problem for everyone after it.

Headless platforms such as Contentful, Sanity and Storyblok decoupled content from presentation. Content lived in a structured store with an API, and any front end could pull from it. That genuinely solved multi-channel delivery: the same article could feed a website, an app and a screen in a shop without being rewritten for each.

The trade was quieter. Rendering, SEO and forms stopped being things the CMS handled and became things engineering teams owned. The marketer gained clean, reusable content and lost the ability to see the finished page without a build step in between. Headless moved the boundary of the problem rather than removing it.

The third generation changes who operates the system. An agentic CMS is one an AI agent can operate end to end under governance: composing, editing and migrating content, with humans deciding what ships. The agent does the repetitive work of drafting, restructuring and moving content between systems; the human keeps the decision about what is good enough to publish.

This is not a headless platform with a chatbot bolted on. The distinction is structural. In an agentic CMS the model works against validated data and governed actions, not raw code or markup, and every change carries its evidence with it. The system is built so an agent can be trusted to operate it, and so a person can review what it did.

Most vendors now use the word agentic somewhere on their site. The label is cheap; the architecture is not. Three tests cut through the positioning.

What does the model emit? If it emits code or markup, it is a generation one or two system with a writing assistant attached. If it emits validated data against a governed schema, it is operating the CMS itself.

Can output ship without a human decision? If a change can go live the moment the model finishes, the system has no governance boundary. An agentic CMS makes the human approval step a first-class part of the workflow, not an afterthought.

Does evidence travel with the content? If the model can tell you where a fact came from and why a change was made, the content carries its provenance. If it cannot, the system is generating text, not managing content.