Change management in a rank hierarchy
Lewin, Kotter and ADKAR applied to a real culture-change intervention for junior sailors: run from the middle of the chain of command, not the top, and measured rather than asserted.
In my experience, junior personnel are the group least likely to feel heard in a rank-based organisation. A workforce that does not feel able to speak up is an expensive workforce to keep.
In March 2025, while acting as training coordinator for a department of around 60 personnel, I designed and ran a one-day intervention, Empowerment Day, to test whether a structured, junior-led forum could shift that dynamic at unit level. This piece sets out what I built, which recognised change models I used to design and evaluate it, what moved, and what did not.
A mid-ranking individual with no authority to change policy used established change-management theory to design an intervention and measured its effect. The results show where the theory held and where the hierarchy pulled back.
The Diagnostic
Why the gap existsEmpowerment is structurally hard in a rank-based organisation, as a three-level view of culture shows (Schein, 1985). The visible artefacts, such as uniforms, insignia and formal ceremony, work as constant signals that rank equates with authority, regardless of technical expertise. The espoused values (teamwork, excellence) stand at odds with decision-making that stays tightly held at senior level. That mismatch breeds a defensive routine (Argyris and Schön, 1978): people stop believing the stated values apply to them.
Power bases sharpen the point (French and Raven, 1959). Junior personnel often hold expertise: they know the equipment, the systems, the daily workarounds, but positional authority still outranks it, and that information has to travel upward slowly through a hierarchy built for control. I used a "drip-tray" diagnostic (borrowed from innovation-network practice) to track recurring low-level complaints as early-warning signals: repeated concerns about being excluded from decisions, feeling overlooked, and leadership appearing disconnected from the operational reality on the floor. Individually, each complaint looked minor. Cumulatively, they pointed at a psychological-safety deficit.
Three Models, Applied Together
Lewin · Kotter · ADKARNo single change model covers the strategic, behavioural and individual levels of a change like this at once. I applied three in combination instead: Lewin's Unfreeze-Change-Refreeze for the behavioural transition, Kotter's Eight-Step model for building organisational momentum, and Hiatt's ADKAR for the individual experience of the people being asked to change. Each model is strong where the others are weak, and each has a documented blind spot the others cover. Running all three side by side lets this evaluation show where the intervention worked and where it did not.
Lewin: Unfreeze → Change → Refreeze
Unfreezing removed the visible signals of rank: epaulettes off, senior leaders in a listening-only role, to lower the psychological barrier to honest speech. The change phase worked inside the room. Refreezing is where it struggled. Personnel returned to a hierarchy that had not changed, and without formal reinforcement, behaviour reverts once the novelty fades (Lewin, 1951).
Kotter: Eight Steps
Steps one to three landed well: urgency was backed by retention data, a coalition formed across ranks, and a clear vision was written into the training memorandum. Short-term wins were concrete: 42 ideas generated, 25 assessed viable, 12 implemented within 60 days. Steps six to eight, consolidating gains, producing more change, anchoring the new approach, were only weakly met, because nothing formally tied inclusive leadership to how people got assessed or promoted.
ADKAR: The Individual Lens
Awareness was high (84% of junior personnel recognised the underlying problem). Desire was split: 60% approached the day with openness, 25% with scepticism about whether leadership meant it, 15% treated it as performative. Knowledge and ability both rose measurably during the day itself. Reinforcement was the weakest link, exactly as ADKAR predicts it will be in a hierarchy with no built-in mechanism to sustain new behaviour once the event ends.
What Moved
Measured resultsThe point of running three models in parallel was to avoid treating a well-received event as evidence of durable change. The pre- and post-event survey data draws that line clearly.
Two events have now run: the March 2025 pilot and a second in December 2025, with senior endorsement to continue on a termly basis. Kotter warns against mistaking that kind of short-term win for institutionalised change, because a recurring event on a calendar does not by itself change how the organisation operates (Kotter, 1996). Whether Empowerment Day becomes a change in how the organisation operates is the question the next section addresses.
Where the Model Predicts the Limit
Herzberg's hygiene factors were still unresolved months later: the December feedback minutes recorded the same complaints about parking, accommodation, and equipment reliability that had been raised before, with one item deferred three separate times in 90 days. No amount of empowerment messaging fixes a hygiene factor that management has not addressed, and the theory predicts that outcome (Herzberg, 1959): recognition and voice can lift motivation only so far against unresolved friction.
The deeper limitation is structural. Kotter's institutionalisation steps require decision-making processes, appraisal criteria, and resourcing to change, not just an event calendar. A termly Empowerment Day shows goodwill. Inclusive leadership built into how the organisation promotes and resources people is a different thing, and the model says that gap is where reverted behaviour comes from.
What This Means in Practice
Four recommendations- Formalise junior input into decision-making: standing forums feeding into the actual planning cycle, with a visible route for ideas to be accepted, implemented, or explained away
- Put inclusive leadership into appraisal criteria: protecting junior voice and acting on feedback should be a visible factor in who gets selected for the next role
- Anchor the initiative to existing strategy: linking it to published organisational culture and people strategy documents gives it institutional legitimacy and a claim on resources
- Train the change agents themselves: running a session like this well requires the same change-management literacy used to design it, plus the emotional skill to handle the status friction that inclusive leadership creates for people used to command-and-control
None of this is unique to a military hierarchy. Any organisation with a steep authority gradient, such as a partnership structure, a regulated industry with strict sign-off chains, or a large operational workforce reporting through several management layers, hits the same wall: a well-designed, well-received intervention that does not survive contact with the incentive structure underneath it. The fix is making the event's outputs visible in exactly the systems, such as appraisal, promotion, and resourcing, that decide what the organisation rewards.