AI & Strategy: Why implementation remains human

As Jeff Hiatt’s ADKAR model recognises, organisational change happens one person at a time through five sequential building blocks: Awareness (understanding why change is necessary), Desire (personal motivation to participate), Knowledge (understanding how to change), Ability (skills to implement change), and Reinforcement (sustaining change over time). Each element requires human understanding, collaboration, and ongoing support that AI cannot provide alone now.

The challenge isn’t computational—it’s fundamentally adaptive. Consider what implementation actually requires:

The Translation Challenge: Corporate strategy typically defines what an organisation aims to achieve but provides limited guidance on how teams should contribute. Different levels of an organisation face different bottlenecks and constraints—the corporate bottleneck might be market position, a manufacturing division might face capacity constraints, whilst a regional sales team might struggle with customer relationships. Each level needs its own strategy framework that takes corporate goals as aspirations whilst identifying their specific bottlenecks and creating clear decision rules for their context.

Individual Change at Scale: Each person must develop the ADKAR elements for lasting change. AI cannot create the emotional commitment essential for building desire or the cultural understanding needed for effective reinforcement.

Cultural Navigation and Resistance as Insight: Implementation success depends on understanding organisational culture, competing values, hidden assumptions, and unspoken concerns—dynamics that resist algorithmic interpretation and require human insight to navigate effectively. Rather than viewing resistance as an obstacle to overcome, effective implementation recognises resistance as valuable feedback from people who care deeply about outcomes.

David Rock’s SCARF model provides valuable insight into why people resist change, often exposing issues that were not fully appreciated or understood. Resistance frequently stems from perceived threats to Status (feeling demoted or less valued), Certainty (inability to predict what’s coming), Autonomy (loss of control over their work), Relatedness (disconnection from their team or organisation), or Fairness (sense of unfair treatment). When we view resistance through this lens, what initially appears as obstruction often reveals legitimate concerns about implementation approaches that, when addressed, create far more robust and sustainable solutions. This cultural intelligence requires human empathy and collaborative problem-solving that AI cannot replicate.

The problem isn’t that strategies lack analytical rigour; it’s that they fail to engage the human systems required for execution. People don’t resist change because they lack data— more likely they appear to resist change because they lack motivation⁶ from Autonomy (a sense of independent ownership of their part of the strategy implementation), Mastery (confidence in their abilities to execute), and Purpose (a sense of understanding of why change is needed) reinforcing the power of the ADKAR approach.

This is precisely where traditional “do it for you” consulting approaches fail, whether delivered by human consultants or AI systems. As we observe: “They analyse, interview, and deliver an impressive strategy deck. Everyone nods during the final presentation. Six months later? The recommendations gather dust, and nothing has changed.”

The disconnect is clear: corporate strategy often defines what an organisation aims to achieve but provides limited guidance on how teams should contribute. This implementation gap isn’t caused by poor analysis, but by the missing translation layer between high-level direction and day-to-day decisions. AI may enhance corporate strategy development, but it cannot create the nested strategy frameworks that teams need to translate corporate objectives into practical decision rules adapted to their specific contexts.