Three layers of the shift

To understand the change, it helps to separate three layers that are often mixed together:

• CMS (the tool): The publishing layer itself, including interfaces, workflows, and content management.
• Content platforms (the architecture): How content is modelled, stored, and distributed across multiple channels through APIs and structured data.
• Content operations (the way of working): How the organisation governs, produces, quality-assures, and reuses content over time.

It is where these three layers intersect that the real complexity in modern CMS initiatives emerges.

From web publishing to distributed content

Historically, CMS platforms have primarily been built for web publishing. The architecture has been based on the website as the primary channel, structured around content pages and relatively clear publishing workflows with defined editorial processes.

This worked well in a world where the website was the hub of digital presence.

Today, usage looks different. Content is no longer limited to a website, but needs to function across apps, customer portals, internal systems, and e-commerce platforms. It also needs to support new types of consumption beyond human users, such as AI-driven interfaces and recommendation engines.

This means content can no longer be treated as pages, but as structured information that must be reusable across multiple contexts.

This is where composable architecture and headless-based models become logical consequences rather than technological trends.

Instead of asking “How do we publish a page?”, organisations need to start asking: “How do we structure, distribute, and govern content across our entire digital ecosystem?”

As content is used in more channels, the way it is created and managed also changes. Where there were previously separate departments and responsibilities for different types of content, the same information now needs to be reused in multiple ways.

Product information, for example, may need to work across everything from websites and advertising to internal sales tools, AI assistants, and support environments. This increasingly turns content into infrastructure rather than publishing.

When content becomes infrastructure, new requirements emerge:

• shared content models

• metadata and taxonomies

• governance

• versioning

• API structures

• channel-agnostic distribution

• quality assurance

• ownership and accountability

This is why many CMS platforms today are positioning themselves far more broadly than before. They no longer sell publishing alone. They sell control over content.

The organisational shift

The most significant shift is not happening in the technology, but in organisations. CMS platforms have traditionally been tools for marketing and communications. Today, they affect far more parts of the business.

Content is no longer used only to inform, but also to drive customer experience, sales enablement, customer journeys, support, and self-service. This means content can no longer be managed or owned within a single function.

Instead, it becomes a cross-functional concern where marketing, product, communications, IT, and data need to collaborate far more closely than before.

This creates new challenges:

• who owns the content model

• who defines rules for structure and metadata

• who is responsible for quality across multiple channels

• how central governance is balanced with local flexibility

This is where many organisations underestimate the complexity of modern CMS initiatives. What starts as a website project often becomes, in practice, a business transformation project.

AI is changing the requirements

Many organisations see AI as a layer on top of existing content, a tool for writing faster, creating more variations, or improving production efficiency.

In practice, something different is happening. AI is changing the requirements for how content needs to be structured in order to work across digital systems.

When content is used in AI-driven interfaces, search experiences, and automated workflows, it becomes clear that:

• unstructured content is difficult to interpret and combine

• page-based content is difficult to reuse effectively

• lack of metadata reduces the quality of generated answers and recommendations

This affects both how content needs to be modelled and how it must be governed over time.

At the same time, volume increases. AI makes it possible to produce more content, faster, and in more variations than before. This creates a new kind of challenge: not only how content is produced, but how quality, relevance, and consistency are maintained as volume scales.

The focus shifts from production to governance. And in doing so, AI becomes not just a technical question, but in practice a question of content modelling and governance.

Future CMS selection

As content takes on a more central role in the organisation, what a CMS actually needs to be is also changing.

It is no longer about publishing features, editorial experience, or content structure within a single channel. These aspects still matter, but they are no longer decisive.

Instead, platforms need to be evaluated based on how well they support a broader ecosystem of content work.

This also means that many traditional evaluation models need to be complemented. It is no longer enough to compare features in a requirements checklist. You also need to understand the operational model the platform is designed to support.

Ultimately, it is not the tool itself that determines the outcome, but how well it enables the organisation’s ability to work with content, data, and AI in a structured way over time.

This is not a shift about CMS becoming “more advanced”. It is about content itself becoming one of the organisation’s most critical operational assets.

And in that world, it is no longer enough to simply publish content. It also needs to be governed, reused, and made to work within a much broader context than before.