Part 2 | AI Won’t Fix A Broken Content Operating Model

Part 2 of the CoMakers series, Designing the In-House Agency of the Future

The conversation around AI in creative organizations often starts with tools.

Which platform should we use?

What can we automate?

How much faster can we make content?

How many more versions can we produce?

Those are reasonable questions.

But they aren't the first questions organizations should be asking.

Before deciding what AI can produce, leaders need to understand where AI belongs in the way work actually gets done.

Because AI isn't simply becoming another creative tool.

It is becoming part of the operating infrastructure around creative and content.

Start with the operating problem, not the technology

Technology decisions often begin with capabilities.

A new platform promises faster versioning, automated workflows, content generation, localization, personalization, or greater productivity.

The organization then tries to determine where to insert that capability.

I'd reverse the sequence.

Start with the operating problem.

Where is work slowing down?

Where are teams spending time on repetitive tasks?

Where is demand exceeding capacity?

Where are handoffs creating friction?

Where is content being recreated unnecessarily?

Where are teams struggling to find or reuse existing assets?

Where are decisions taking too long?

Then ask whether AI is the right intervention.

That distinction matters because not every operating problem is an AI problem.

Sometimes the issue is unclear ownership.

Sometimes it's poor prioritization.

Sometimes it's fragmented data.

Sometimes it's a workflow people don't follow.

Sometimes it's simply a decision that nobody has been given authority to make.

Adding AI doesn't automatically solve any of those things.

Faster fragmentation is still fragmentation

AI has the potential to dramatically increase content velocity.

That's exciting.

It also creates an important question:

What happens when dramatically more content enters a system that already struggles to manage today's volume?

If intake is unclear, AI can create more inputs.

If governance is weak, AI can create more risk.

If approvals are slow, AI can create more work waiting for approval.

If asset management is fragmented, AI can create even more assets to manage.

If teams don't know which content actually creates value, AI can simply help produce more of everything.

That isn't transformation.

It's acceleration without orchestration.

AI needs an intentional place in the workflow

The more useful question is not:

Where can we use AI?

It's:

Where should AI enter the operating model?

That requires thinking beyond generation.

AI may have a role in:

Planning.

Brief development.

Ideation.

Research.

Asset retrieval.

Versioning and adaptation.

Localization.

Production workflows.

Quality control.

Measurement.

Optimization.

Administrative work.

But each use case comes with different implications.

Who owns it?

What data can be used?

What requires human review?

What needs legal approval?

Which tools are approved?

Where does the output go?

How is quality evaluated?

What happens when something goes wrong?

Who is accountable?

Those are governance and operating-model decisions.

Adoption matters as much as capability

Organizations often underestimate the human side of transformation.

A workflow can be beautifully designed and still fail if nobody uses it.

An AI platform can be technically impressive and still create little business value if teams don't understand where it belongs.

That means implementation cannot end with deployment.

Teams need clarity around expectations, roles, training, approved use cases, and decision rights.

They also need to understand why the new way of working is better than the old one.

Otherwise people do what people have always done when systems don't work for them:

They create workarounds.

AI readiness is really operating-model readiness

The organizations best positioned to benefit from AI may not be the ones experimenting with the most tools.

They may be the ones with the clearest understanding of their own system.

They know how work enters.

They understand demand.

They know where decisions happen.

They have clear governance.

They understand their content and data.

They know where internal teams and external partners fit.

And they can identify precisely where AI can improve the system.

That creates a very different starting point for AI adoption.

AI should not be layered onto the operating model as another tool. It should be designed into the operating model as a capability.

Next in the series

Part 3: More Content Isn't Always a Capacity Problem

When teams are overwhelmed, the instinct is often to add resources. But many apparent capacity problems begin somewhere else in the system.

Want to go deeper?

This four-part series introduces several of the ideas explored in the CoMakers Group flagship report:

Designing the In-House Agency of the Future

The full report goes deeper into the operating system behind modern in-house organizations, including the strategic shifts shaping the model, stages of organizational maturity, emerging gaps, leadership implications, predictions for the next five years, and a practical path forward.

The full report is available upon request.