Article

x

Two Point O

Your Marketing Team Is About to Chat with Their CMS (Here's Why That Changes Everything)

A profile picture of Bert Swinnen
Bert Swinnen
AI
Customer Experience
Composable

Introduction

Picture this..

You walk into the office, check your content calendar, and realise you need to publish that article on MCP today. Instead of diving into another 30-minute navigation adventure through your CMS, you simply open a chat and type:

"Create that blog post on MCP we discussed, link it to Bert Swinnen as the author, use our standard CTA, and get it live by 2 PM."

Done. Published. Perfect.

This isn't wishful thinking—it's what we're already facilitating for our clients. The Model Context Protocol (MCP) is transforming how marketing teams interact with their content systems, moving us from click-heavy interfaces to natural conversations that actually get work done.

From Clicks to Conversations: A New Reality

Let me tell you what we see happening in marketing teams every day. The modern headless CMS, with complex relationships between articles, authors, SEO metadata, CTAs, and tags, has become incredibly powerful, but managing all these connections through traditional interfaces takes significant time.

Here's why this overhead exists and why it's worth it: headless architecture enables true omni-channel content delivery. When you create your article, it doesn't just live on your website, the same structured content automatically powers your mobile app, feeds your email campaigns, populates social media, appears in sales presentations, and drives your customer portal, all from a single source of truth. Your content becomes platform-agnostic, reusable, and infinitely flexible.

But achieving this omni-channel superpower requires managing intricate content relationships, every piece needs proper metadata, author attribution, CTAs, and strategic tags for discoverability across all channels. When your content flows to multiple touchpoints, getting these relationships right becomes mission-critical, which is exactly why the traditional click-through-menus approach can feel somewhat cumbersome.

Here's how content creation actually works today, illustrated through a real scenario I handled recently, specifically, getting this article online 🙂.

The Old Reality: Navigation Marathon

Publishing one strategic article required jumping between screens like a mad man.

An image of a content model consisting out of an Article, Page, Content-types and meta-data.
A diagram of content model for an article structure

  1. 1

    Create Article → Navigate to content types → Add entry → Copy/paste → Save

  2. 2

    Link Author → Find author field → Search "Bert Swinnen" → Link → Save

  3. 3

    SEO Setup → Switch to SEO section → Create metadata → Link back → Save

  4. 4

    Page Configuration → New page entry → Set slug → Link article → Save

  5. 5

    CTA Addition → Find CTA library → Link consultation CTA → Save

  6. 6

    Tag Application → Add tags individually → Final save

Six different content entries. Dozens of clicks. Multiple publishes. One article.

The New Reality: Conversational Creation

With an MCP-powered approach, that same complex workflow becomes:

Me: "Create a thought leadership piece: Your Marketing Team Is About to Chat with Their CMS.'

- Author: Bert Swinnen.
- Use our default CTA.
- Slug: chat-with-your-cms-with-mcp.
- Tags: MCP, Content Operations, Digital Transformation.

Suggest additional relevant tags?

MCP enabled Agent: "Perfect! Article created with all relationships. I suggest adding 'Marketing Automation' and 'Headless CMS' tags. Everything's ready as draft. Publish now?"

Me: "Yes, publish it."

MCP enabled Agent: "Live and optimised. All SEO metadata auto-generated from content analysis."

Same result. One conversation. Zero navigation headaches.

Understanding this MCP thing

So what makes this conversational magic possible? MCP creates a universal standard that lets AI agents talk directly to your headless CMS through its APIs.

MCP is rapidly becoming an open standard for the AI agentic era.

Demis HassabisCEO of Google DeepMind


In practical terms, MCP gives AI agents three superpowers with your CMS:

  • Understanding Structure: The AI agent understands your content models, field relationships, and required connections.

  • Performing Actions: The AI agent can execute complex operations like updating multiple entries or finding specific content.

  • Retrieving Information: The AI agent can read your content and use it to enrich its responses and decision-making.

Critical note

When you provide an AI agent with CMS access, it sees your entire content repository. That includes proprietary strategies, unpublished campaigns, and confidential client work.

For organisations handling sensitive information, this raises important intellectual property concerns—which is exactly why we often recommend self-hosted AI solutions that keep everything within your controlled environment.

Why This Is a Game-Changer for Marketing

We are implementing conversational CMS solutions for clients across industries, and the transformation goes far beyond time savings. Here's what we aim for:

Marketing teams stop being system administrators
Instead of spending 30% of their time navigating interfaces, they focus on strategy, creativity, and audience engagement. At the time of writing, it is too soon to tell how big the effect will be, but this article was done under 2.5 hours, including the research, editing and consolidation. Not too bad in my book. Just saying.

Campaign velocity increases dramatically
When market opportunities arise or breaking news demands quick response, teams deploy content in minutes instead of hours.

Content quality improves paradoxically
With less cognitive load spent on remembering field requirements and navigation paths, teams invest mental bandwidth in content strategy, audience research, and creative development.

Governance becomes sophisticated yet simple
Complex content strategies remain maintainable because AI understands relationships and can proactively suggest improvements, identify content gaps, and ensure brand guideline compliance across all channels.

Some quick back of a napkin math

The efficiency compounds.

Save 30 minutes per content piece × 20 pieces monthly × 5 team members = 50 hours reclaimed monthly. That's 600 hours or a whopping 75 mandays annually redirected from interface management to strategic work.

The Interface Evolution: From Daily Driver to Specialised Tool

As conversational interaction moves into the limelight, traditional CMS interfaces evolve rather than disappear. We're seeing the UI transform from the go-to interaction method to a specialised tool for specific use cases.

Daily tasks like content creation, simple edits and content discovery, increasingly happen in chat windows, integrated assistants, or directly within collaborative workspaces whilst the traditional UI focuses on what it does best:

  • Complex Configuration: Content model design, workflow setup, and system administration benefit from visual interfaces that clearly display relationships and dependencies.

  • Visual Page Building: While AI assists with suggestions and content placement, sophisticated layout management still requires visual editors.

  • Advanced Governance: Approval workflows, publishing rules, and compliance management work better with traditional UI elements.

For most content authors, the primary interface becomes conversational, with the traditional UI serving as a "power tool" for sophisticated operations requiring visual feedback.

Who's Leading the MCP Revolution?

The MCP adoption across headless CMS platforms is accelerating rapidly, with leaders emerging:

Sanity, currently is leading the pack with their official MCP implementation. Their server connects structured content directly to AI agents, enabling natural language content queries and document management. Knut Melvær, Principal Developer at Sanity, demonstrates the power in his demo video showing Claude automatically finding and updating content across multiple documents in seconds.


Contentful
currently relies on a ~community (by their own) implementation. Developers have already created an MCP solution that enables full CRUD operations, space management, and bulk operations. Given our website gets its content sourced from Contentful, we used the community MCP implementation courtesy of Ivo Toby (AI-ML Engineer at Contentful). Worked like a charm, so wondering how long it will stay community-driven?

Directus, which we like for its capability to be self-hosted, provides official MCP support through their Content MCP Server, supporting multiple AI clients including Claude Desktop, Cursor, amongst others with advanced configuration options and customisable system prompts.

Storyblok offers MCP integration through a detailed implementation guide, enabling teams to connect their content management workflows directly with AI agents.

This listing is obviously not exhaustive. Conceptually every CMS with a well-defined API can be hooked up with agents by means of MCP one way or another. However, CMS's with an existing MCP server are easier to help you get started.

Balancing Innovation with Control

Here's what I tell every client considering MCP implementation: The technology is transformative, but you need to understand the trade-offs.

The Advantages Are Real:

  • Standardisation reduces integration complexity across your entire tech stack

  • Accessibility democratises sophisticated content operations across skill levels

  • Productivity gains that compound monthly as teams adapt to conversational workflows

  • Future-proofing as AI capabilities advance within a stable protocol framework

The Considerations Are Important:

  • Security challenges including prompt injection risks and tool permission vulnerabilities identified by security researchers

  • IP protection concerns when AI agents access your complete content repository

  • Learning curve as teams develop prompt engineering and AI workflow skills

The Future of Content Management

The transformation from interface-driven to conversation-driven content management represents a fundamental shift toward more strategic, efficient operations. We're helping clients navigate this transition not just as a technology upgrade, but as a competitive advantage.

As MCP adoption accelerates and AI capabilities advance, marketing teams can focus entirely on what drives business results: strategy, creativity, and audience engagement. The technical complexity of modern content management gets handled conversationally while the sophisticated, relationship-rich content that enables omni-channel experiences remains unchanged.

For marketing leaders, the question isn't whether this transformation will happen, it's how quickly your organisation embraces conversational content management to gain competitive advantages in speed, efficiency, and strategic focus.

The future of content management is here, and it speaks your language.

Feel inspired? Connect with us, make the click.

Let's chat

Ready to make the click?

Get in touch