Nov 14, 2025
For many organizations, AI is still treated as a software upgrade: choose a vendor, buy a platform, and hope it scales. It can work, but it often gets stuck as well. Not because the technologies are inadequate, but because the organization around them isn’t fully adapted to how work changes.
The Structural Difference
AI doesn’t just fit into an existing organization. It reshapes how work flows through a company. It changes who makes decisions, how information moves, and how teams collaborate. It blurs the boundaries between roles. If the structures remain the same, friction arises immediately. You can set up a model. You cannot set up organizational change in the same way.
And that is where most AI strategies get stuck: they optimize the technology but ignore the system it lives in.
The real power of AI emerges when you create harmony between your processes and the tool. That is when the real change happens.
The Real Barriers Are Not Technical
Research underscores this. A recent study from McKinsey, "The State of AI 2025", shows that about 88% of organizations now say they use AI in at least one business function. But only 39% attribute any level of overall EBIT impact to AI.
They find that the single largest factor for profitability impact is the redesign of workflows, meaning how the organization works, not just which tools are used.
Redesigning How Work Is Done
When AI becomes integrated, it affects:
Workflow Structure: which tasks people do, which tasks the AI supports
Decision-Making Responsibility: who uses the results, who refines them, who owns the outcomes
Data and Process Infrastructure: clean sources, well-mapped processes, defined owners
Governance and Measurement: how you track value, how you manage change and risk
McKinsey shows that these types of holistic approaches say more about the return on AI than the investment in models, platforms, and tools themselves.
When the foundation is solid, even modest early AI implementations deliver meaningful results. Conversely, when the foundation is lacking, one does not achieve the full ROI.
The Way Forward
Here is a practical checklist of things to consider:
Map Your Key Processes Before: For communication, this might mean mapping how a press release goes from draft → review → approval so that the AI supports the right steps.
Clarify Roles and Responsibilities: who uses the result, who oversees, who owns the data. Specifically in PR work, this avoids confusion about who approves a pitch or finalizes messaging when the AI drafts it.
Clean and Govern Your Data: Be clear about which data the AI should work with and rely on (one source of truth) and make it accessible and quality controlled. For content teams, this might mean consistently using the same documents as a basis, so the AI works with outdated brand messages or incorrect boilerplates.
Measure and Govern: Define what success looks like and who measures it. In communication, it might mean tracking how much time AI saves in creating press releases or how it helps maintain consistency in messaging.
Align Incentives and Culture: ensure that teams understand how their work changes with AI. For communicators, this means a shift from writing everything manually to monitoring, refining, and validating AI-generated text.
💡 Conclusion: AI is not just about the model or the platform. It’s about how the organization is designed. When the foundation is in place, even relatively small AI steps become springboards for greater transformation.
If you want to know more about how you can leverage AI in communication, please contact us.