Skip to content
Beyond Prompt AI Studio
Change Management for AI Adoption

Two Levels, Two Models: Kotter and ADKAR

Anyone looking for "the" change management model for AI adoption mostly finds two names: Kotter and ADKAR. They're often presented as competitors – but they actually answer fundamentally different questions, and the interesting part is how they complement each other.

Four model realities – worth remembering

Try it yourself: the five ADKAR stages

Stage 1

Awareness

Build awareness of the need for the change.

Kotter: the organisational perspective

John Kotter's 8-step model is top-down and organisational: create a sense of urgency, build a guiding coalition, develop a vision, communicate that vision, remove obstacles, generate short-term wins, use that momentum to drive further change, and anchor the new practice in the culture. Kotter answers the question: how does an entire organisation move from A to B?

ADKAR: the individual perspective

Prosci's ADKAR model picks up exactly where Kotter leaves off: at the individual person. ADKAR stands for five sequential stages that each person must go through individually for a change to actually take hold: Awareness (of the need for change), Desire (the personal wish to participate), Knowledge (of how the change works), Ability (the actual capability to implement it), and Reinforcement (so the change sticks instead of reverting).

Why skipping a stage causes failure

ADKAR's practical value lies in its sequence: whoever delivers Knowledge before Desire is even present wastes the training – the information evaporates because the motivation to apply it is missing. This exact mistake happens especially often in AI rollouts: a company trains a tool extensively without first clarifying why employees should even want to use it.

Practice section: think both levels together

For an AI rollout that means concretely: Kotter supplies the organisational frame (create urgency, guiding coalition, vision), ADKAR supplies the check at the individual level of whether that frame actually lands with each person. A team check after every Kotter stage that asks "Are we at Awareness? At Desire?" surfaces gaps before they turn into quiet failure – the pattern from the first module of this course.

The key points

  • Kotter's 8-step model is organisational and top-down: urgency, guiding coalition, vision, anchoring in the culture.
  • ADKAR is individual: Awareness, Desire, Knowledge, Ability, Reinforcement – five stages each person goes through individually.
  • The two models don't compete, they answer different questions: organisation versus individual person.
  • Skipping an ADKAR stage (e.g. Knowledge before Desire) lets training evaporate without effect – a common mistake in AI rollouts.
  • In practice, a team check after every Kotter stage that verifies the individual ADKAR level pays off.

Why Most AI Projects Fail on People, Not Technology

Quick check: did it sink in?

1 / 3

Which question does the ADKAR model answer, in contrast to Kotter?

Want to combine Kotter and ADKAR specifically for your next AI rollout?