Ask ten change practitioners which model they use and you will hear Kotter, ADKAR, Lewin, Bridges and a few others, often with some conviction. Each model has something useful to offer. None of them covers everything, and the people who get the most from them tend to use several, each for the job it does best.
This guide compares the change management models we find most useful: what each is for, where it falls short, and what the research says about them. It is part of our wider guide to change leadership.
Which change management models are worth knowing?
| Model | What it's good for | Where it falls short |
|---|---|---|
| Lewin's three stages: unfreeze, change, refreeze | A simple way to explain that change needs preparation and consolidation as well as action | The popular version oversimplifies Lewin, and can read as too linear for organisations that never stop changing |
| Kotter's eight steps | A clear sequence for large organisational change, with a strong focus on urgency, coalition and early wins | Light on practical detail, and can be applied in a top-down way |
| ADKAR | Tracking where individuals are stuck in adopting a change | Can turn into a checklist if used without judgement |
| Bridges' transition model | Understanding the psychological journey people go through, starting with an ending | Describes the experience but doesn't give you an organisational plan |
| The change curve | Recognising the emotions people move through during a change | Often treated as a fixed sequence, which it isn't |
| Beckhard and Harris change formula | A quick test of whether the drivers of change outweigh the resistance | A diagnostic, not a process |
| McKinsey 7S | Checking that strategy, structure, systems, shared values, style, staff and skills line up | Diagnoses misalignment but doesn't sequence the fix |
| Balogun and Hope Hailey's change kaleidoscope | Designing a change approach that fits the organisation's context | Takes time and experience to apply well |
| Adaptive leadership | Problems with no known answer, where people have to learn their way forward | Less prescriptive, so it asks more of the leader's judgement |
What does the research say about change models?
Less than their popularity suggests. Stouten, Rousseau and De Cremer (2018) reviewed seven widely used models, including Lewin, Kotter and ADKAR, and found that they more often cite expert opinion as their foundation than scientific evidence.
Their review also found that many of the steps these models recommend do have research support. They distilled ten evidence-based steps, among them assessing the problem or opportunity, building a guiding coalition, forming and communicating a clear vision, empowering others to act, developing people's knowledge and ability, using short-term wins, monitoring the change over time, and embedding it in culture and practice (Stouten, Rousseau and De Cremer 2018).
That is a useful way to read any model. The steps that appear across most of them, and that the research supports, are the ones to take seriously. The particular sequence and labels matter less.
Lewin's three-stage model
Kurt Lewin's model is the oldest in common use: unfreeze the current way of doing things, make the change, then refreeze so the new way holds (Lewin 1947). Its central insight is that change needs preparation before it starts and consolidation after it is made.
The popular version is a simplification. Cummings, Bridgman and Brown (2016) traced the unfreeze, change, refreeze model back to a brief passage in one of Lewin's last articles, and argue that the simplified version has obscured his richer work on group dynamics and action research. Burnes (2004) makes a related case that critics who dismiss Lewin as too simple and linear have misread him.
Use it for: explaining to a leadership team that a change needs groundwork and follow-through.
Watch out for: treating "refreeze" literally in an organisation that will face its next change within months.
Kotter's eight steps
John Kotter's model grew out of his analysis of why transformations fail (Kotter 1995) and was set out in full in Leading Change (Kotter 1996). The eight steps are: create a sense of urgency, build a guiding coalition, form a strategic vision, communicate the vision, empower broad-based action, generate short-term wins, consolidate gains and produce more change, and anchor the new approaches in the culture.
It remains the most widely recognised change model. A review of the evidence found research support for most of the individual steps, but no formal studies testing the model as a whole, and concluded that its popularity rests on its practical, easy-to-use format more than on scientific consensus (Appelbaum et al. 2012).
Use it for: large organisational changes that need a clear sequence and strong leadership backing.
Watch out for: running it as a top-down programme. The steps describe what leaders do, so it can underplay what the people affected are going through.
ADKAR
Prosci's ADKAR model focuses on the individual. It sets out five outcomes each person needs to reach for a change to stick: awareness of the need for change, desire to take part, knowledge of how to change, ability to do it, and reinforcement to keep it going (Hiatt 2006).
Its strength is diagnosis. If adoption stalls, ADKAR helps you work out where: people may understand the change but not want it, or want it but lack the skills.
Use it for: tracking adoption across groups, and finding where individuals are stuck.
Watch out for: using it as a form to fill in. The value is in the conversations it prompts, and much less in the scores.
Bridges' transition model
William Bridges distinguished between change, which is the external event, and transition, which is the internal process people go through to come to terms with it. Transition has three phases: an ending, where people let go of the old way; a neutral zone, where the old has gone and the new isn't yet working; and a new beginning (Bridges 1991).
Its most useful insight is that every transition starts with an ending. Leaders who skip straight to the new beginning, without acknowledging what people are losing, tend to meet resistance they don't understand.
Use it for: planning the human side of a change, especially the period when the old way has gone and the new way isn't working yet.
Watch out for: expecting it to give you a project plan. It describes the experience. It doesn't sequence the work.
The change curve
The change curve maps how people's feelings shift over a change, from early optimism through a low point to acceptance. It was adapted from Elisabeth Kübler-Ross's work on grief (Kübler-Ross 1969). Our guide to resistance to change covers it in detail.
Use it for: recognising what people are feeling, and adjusting how you lead at each stage.
Watch out for: treating it as a timetable. People move at different speeds, some go backwards, and some skip stages altogether.
The Beckhard and Harris change formula
Beckhard and Harris (1987) set out a simple test: change happens when dissatisfaction with the current state, a clear vision of the future and credible first steps together outweigh the resistance to change. The three drivers multiply, so if any one is close to zero, the change stalls however strong the others are.
Use it for: a quick check, at any stage, of whether a change has enough momentum, and which driver needs work. Our guide to measuring change shows how to turn it into three questions for any group.
Watch out for: expecting more than a diagnosis. It tells you what is missing, and leaves the fix to you.
McKinsey 7S
The 7S framework looks at seven elements of an organisation that need to fit together: strategy, structure, systems, shared values, style, staff and skills (Waterman, Peters and Phillips 1980). A change to one, such as a new structure, usually needs matching changes to the others.
Use it for: checking, before and during a change, that you haven't changed the structure or systems while leaving incentives, skills or leadership style pointing the old way.
Watch out for: stopping at the diagnosis. 7S shows where things don't line up, and it doesn't tell you what order to fix them in.
The change kaleidoscope
Balogun and Hope Hailey (2008) argue that there is no universal formula for change, and that the right approach depends on context: how much time there is, the scope of the change, how much needs to be preserved, how diverse the organisation is, how much capability and capacity it has, how ready people are, and how much power the leaders have to make the change. Their change kaleidoscope helps leaders read those factors before choosing how to run the change.
Use it for: deciding how to approach a change, before choosing a model to run it.
Watch out for: the skill it demands. Reading context well takes experience.
Adaptive leadership
Adaptive leadership, developed by Ronald Heifetz (Heifetz 1994), separates technical problems, which experts can solve, from adaptive challenges, which require the people involved to change their own priorities, beliefs or behaviour. It offers a way of leading through problems that have no known answer, with fewer fixed steps than the other models. Our guide to adaptive leadership covers it in depth.
Use it for: changes where nobody yet knows the answer, and the people affected have to learn their way forward.
Watch out for: its lack of a checklist. It relies heavily on the leader's judgement.
How do you choose a change model?
Start with the change itself. A few questions help:
- Is the answer known? If it is, a sequencing model such as Kotter works well. If it isn't, adaptive leadership fits better.
- Where is the risk? If the risk is organisational alignment, use 7S. If it is individual adoption, use ADKAR. If it is emotional, use Bridges or the change curve.
- How much time and capacity is there? The change kaleidoscope helps here, and our guide to change fatigue covers the cost of ignoring it.
- Do you need to check momentum? Beckhard and Harris gives a quick answer at any stage.
In practice, most significant changes benefit from three things: one model to sequence the organisational effort, one to understand what individuals are going through, and one to diagnose where things stand.
Models are tools. The common mistake is picking one and forcing every change through it. Our own change framework, ICAP, draws on several of these models, including Kotter, the 7S framework, Balogun and Hope Hailey, and Beckhard and Harris, and adapts them to the change in front of us.
If you are planning a significant change and want help choosing an approach that fits, talk to us about our change management work. Our guide to change leadership covers the wider picture.