Change leadership: a practitioner's guide

Change is a given. Leading it is the real work. This guide sets out what change leadership is, why change so often stalls, and what leaders can do about it. It draws on the research and on experience leading people where the stakes were real.

What is change leadership?

Change leadership is the work leaders do to move people through change: setting direction, building commitment and holding the course until new ways of working become normal. It sits above the plans and tools of change management, and it can't be handed off to a project team.

John Kotter drew the line between management and leadership more than three decades ago. Management copes with complexity: plans, budgets, structures and control. Leadership copes with change: setting direction, aligning people and motivating them to move (Kotter 1990). Change leadership applies the second to the hardest moments an organisation faces.

It is also in short supply. DDI's assessment data on more than 100,000 leaders found that only 8% of executives are strong in change leadership (DDI, reported by ATD 2026). In DDI's Global Leadership Forecast, just 18% of leaders said they felt very prepared to manage change (DDI 2025).

Our view is simple. At the core of any organisation are humans, not hardware. Strategy, systems and structure all matter, but people decide whether a change takes hold.

How is change leadership different from change management?

Change management is the structured process and toolkit for moving an organisation from one state to another. Change leadership is the direction, influence and example that make people willing to make the move. Organisations need both, yet most invest heavily in the first and assume the second will follow.

Beer and Nohria (2000) found the same tension in their study of corporate change. Programmes driven only by hard economic levers (their Theory E) and those driven only by building culture and capability (Theory O) both tended to fall short. The organisations that did best combined the two. In practice the plan and the leadership have to turn up together.

Why do most change efforts stall?

Change usually stalls when leaders treat an adaptive challenge as a technical one. Technical problems have known fixes that an expert can apply. Adaptive challenges ask people to change beliefs, habits and loyalties, and only those people can do that work.

The distinction comes from Ronald Heifetz at Harvard's Kennedy School (Heifetz 1994; Heifetz, Grashow and Linsky 2009). A new system, a new structure or a new policy is technical work: it can be designed, scheduled and delivered. Getting people to work differently once it lands is adaptive work. Leaders under pressure reach for the technical fix because it fits on a Gantt chart. In our experience, change is about ten percent a technical problem and ninety percent a human one.

A word on statistics. You will often read that 70% of change initiatives fail. When Mark Hughes traced that figure back to its sources, he found "no valid and reliable empirical evidence" to support it (Hughes 2011). The better evidence makes a more useful point: the quality of the people work changes the odds. In Prosci's research with 2,668 practitioners, 88% of projects with excellent change management met or exceeded their objectives, compared with 13% where change management was poor (Prosci 2023).

Consider a mid-sized client that ran its entire order book on paper. Orders lived in folders, and now and then one slipped out and disappeared. The first anyone knew of it was when a customer rang to chase the order. Everyone wanted new software. Then the rollout began, and the same people who had asked for it started to push back.

Further reading: Adaptive leadership in practice.

Why do people resist change?

Most resistance is a rational response to an expected loss: of status, competence, relationships, certainty or identity. People weigh losses more heavily than equal gains, so a change sold only on its benefits will meet resistance. Leaders who name the losses openly can address them.

Back to the order book. Most of that team had built their working lives on paper and weren't confident on a computer. The new system asked capable, experienced people to become beginners again, and that is a real loss: of competence, of comfort, of the standing that comes from knowing your job cold. Training covered the buttons. The harder work was making it safe to learn.

We worked with the client's leaders to regulate the pressure person by person. Everyone got one-on-one time. Those taking longer had some of their workload lifted, so the strain of learning never tipped into overload. The research points the same way. People learn new ways of working when it feels safe to make mistakes in front of others (Edmondson 1999). They commit when they see their leaders backing the change (Armenakis and Harris 2009). And they need enough pressure to change, but not so much that they shut down, which Heifetz calls the "productive zone of disequilibrium" (Heifetz, Grashow and Linsky 2009). In the end, lost orders dropped to zero. Response times improved, customers were happier, and so was the office. We tell the full story in The technology company that ran on paper.

Three ideas from the research explain what happened:

• Losses loom larger than gains. Kahneman and Tversky (1979) showed that people feel a loss more sharply than an equivalent gain. A team can want the outcome of a change and still resist the cost of getting there.
• Every change starts with an ending. Bridges (1991) separated the change itself, which is situational, from the transition people go through, which is psychological. Transition begins with letting go of the old way, and leaders who skip that step meet resistance they didn't plan for.
• Resistance carries information. Ford, Ford and D'Amelio (2008) argued that resistance is often a response to how a change is led, including broken promises and poor communication, and that it can point leaders to real flaws in the plan. People also differ in how readily they accept change (Oreg 2003), which is why one approach rarely fits a whole team.

The familiar change curve, with its dip from shock to acceptance, was adapted from Elisabeth Kübler-Ross's work on grief (Kübler-Ross 1969). The origin is a reminder of what people are working through.

Further reading: People don't resist change. They resist loss..

What is change fatigue, and how do leaders prevent it?

Change fatigue is the exhaustion and passive resignation that build when people face more change than they can absorb. It shows up as disengagement, cynicism and turnover. Leaders prevent it by counting the total change load on each team, spacing initiatives out and stopping work that is no longer worth the effort.

The costs are measurable. Gartner reports that change fatigue can reduce employees' intent to stay by as much as 42% and their performance by as much as 27%. In a July 2023 Gartner survey of 180 HR leaders, 82% said managers were not equipped to lead change and 77% said their employees were fatigued from all the change (Gartner 2023). Researchers have built a validated measure of change fatigue and found that it wears people down emotionally, which in turn weakens their commitment and increases their intention to leave (Bernerth, Walker and Harris 2011).

Fatigue is cumulative. Each initiative may be reasonable on its own, but people experience them all at once. Leaders can act on that:

• Count the load. List every change landing on each team in the next six to twelve months, including the small ones.
• Sequence. Space the heaviest changes so teams recover between them.
• Stop something. Retire initiatives that have run their course before adding new ones.
• Protect capacity. Lift workload for the people carrying the most change, as the order book team did.
• Watch for signals. Cynicism, silence in meetings and rising sick leave often arrive before resignations.

Check capacity before committing to another change. When change arrives faster than people can absorb it, the strain doesn't stay at work. It follows people home.

Further reading: Change fatigue: what leaders can do about it, Thriving in change workshop.

What do change leaders do before, during and after a change?

Before a change, leaders build a credible case and align the leadership team on one story. During it, they communicate early and often, involve the people closest to the work and treat resistance as information. After it, they reinforce the new behaviours until they become routine.

We've framed each stage with a saying from the military environments our partners come from. They are blunt, and they carry over well to business. The same sequence sits behind our own change framework, ICAP: do the diagnostic work before committing, adjust as the change meets reality, and keep reinforcing until the new way of working holds.

Before: time spent in reconnaissance is seldom wasted

Most change problems are visible before launch to anyone who goes looking. The trouble is that capable, trusted leaders tend to overestimate readiness and underestimate resistance, and never more than when the change is their own. Gather evidence before you commit: what people think, what the data shows, what has gone wrong before.

Then build a case people can believe. Armenakis and Harris (2009) summarise readiness as resting on five beliefs: that change is needed, that this change is the right one, that the organisation can do it, that leaders support it, and that it will benefit the people being asked to change. A case that answers all five gives people a reason to move.

Make sure the leadership team tells one story. When four senior leaders give four different reasons for the same change, staff hear none of them.

A common pattern shows why diagnosis comes first. A business with slow, error-prone processes decides the answer is AI. It buys the tools, plugs them into the workflow, and finds the same problems arriving faster. AI amplifies whatever process it sits on. The organisations that get value map their processes first, fix what is broken, and then add AI where it clearly helps. Michael Hammer made the same argument about computers in 1990: don't automate, obliterate (Hammer 1990).

During: no plan survives first contact

Communicate more than feels necessary. Kotter (1995) lists under-communicating the vision "by a factor of ten" among the most common errors leaders make. Involve the people closest to the work, because they will see problems first. Treat resistance as information and regulate the pressure so people stay engaged without being overwhelmed.

Expect the plan to change. Fix what is failing quickly, reinforce what is working, and keep your head up for opportunities you didn't plan for. Hold to the end state while the path to it shifts.

After: be brilliant at the basics

One of the classic late failures is declaring victory too soon (Kotter 1995). A new system goes live and attention moves on before the new behaviours have settled. Lewin described this final stage as freezing the new level, now usually called refreezing: making the new way stable (Lewin 1947; Burnes 2004). In practice it means the ordinary disciplines, repeated: leaders using the new way themselves, recognising people who do, and removing the old workarounds. Over time, those habits become the culture.

Further reading: Preparing for change, Planning to change, Leading through change.

What skills does a change leader need?

Change leaders need to diagnose before they act, explain the reasons for a change plainly and often, listen to resistance without taking it personally, set a pace people can sustain, and keep their word. None of these is unusual, and most leaders could list them. The difficulty is doing them consistently under pressure.

The capability gap is wide at every level. In DDI's assessment data, 30% of mid-level leaders were strong in leading change, and only 15% of emerging and frontline leaders were strong in facilitating it (DDI, reported by ATD 2026). Frontline leaders matter here because they have the most conversations about the change with the people doing the work.

The skills we see make the most difference:

• Diagnosing the situation. Working out what is actually going on before deciding what to do. Heifetz describes this as getting off the dance floor and onto the balcony, so you can see the patterns you miss while you're in the middle of them (Heifetz 1994; Heifetz, Grashow and Linsky 2009).
• Explaining the why. Giving the same reasons, in plain language, to every audience, and repeating them long after you're tired of hearing yourself say them.
• Listening to resistance. Treating objections as information about the plan, and asking what people expect to lose.
• Setting the pace. Judging how much pressure a team can take, and easing off or pushing on as needed.
• Deciding with incomplete information. Making calls before everything is known, and being open about what might change.
• Keeping commitments. Doing what you said you would do. Broken promises during a change are one source of resistance that leaders create themselves (Ford, Ford and D'Amelio 2008).
• Developing others. Building the capability of the leaders below you, so the change doesn't depend on a handful of people.

Further reading: Leadership development, Technical leadership package.

What is the role of the executive sponsor?

The executive sponsor owns the reasons for the change and gives it authority. In practice that means explaining the case in person, staying visible throughout, clearing the obstacles that only senior people can clear, and holding peers to account for their part. A sponsor who announces the change and moves on to other things leaves everyone else to guess how much it matters.

Sponsorship is the factor most consistently linked to success in change research. Prosci reports that sponsorship has ranked first among contributors to success in every one of its benchmarking reports since 1998, and that projects with an extremely effective sponsor met or exceeded their objectives more than twice as often as those with a very ineffective one (Prosci n.d.). Kotter (1995) made a similar point from a different angle: one of the most common reasons transformations fail is that the leaders behind them never form a strong enough coalition.

Take a logistics business consolidating three regional depots into one. If the CEO announces the decision at a town hall and hands it to a project manager, middle managers will usually read the silence as a sign that the change can wait. Old habits carry on, and the new site runs well below capacity for longer than anyone planned. If the same CEO explains the reasons in person at each depot, sits in on the early planning sessions, sorts out the supplier contract holding up the move, and asks the leadership team every fortnight what is stuck, the managers get a very different message about where the change sits on the list. The plan and the budget can be identical in both cases.

Sponsors aren't the only leaders who matter. Middle managers translate the change for their teams, and they carry a heavy emotional load doing it. Huy (2002) found that middle managers who balanced their own commitment to the change with attention to their people's emotions helped their organisations change while keeping the business running. They need to be briefed before their teams are, and given the time to do that part of the job.

What a good sponsor does:

• Explains the reasons for the change personally, more than once.
• Stays visible for the life of the change, well beyond the launch.
• Removes obstacles that sit above the project team's authority.
• Makes sure the leadership team gives one consistent account.
• Asks regularly what is stuck, and acts on the answer.

Further reading: The leader's role in change, Executive coaching, Board development.

How do you lead people through AI adoption?

AI adoption is a change like any other, with one difference: many people worry it will replace them. Leaders get better results when they are honest about what the technology will and won't change, involve staff in redesigning their own work, and fix the processes before automating them. Treating AI purely as a technology rollout tends to produce expensive tools that few people use well.

The gap between ambition and readiness is large. Kyndryl's 2026 survey of 1,100 senior leaders across eight countries found that only 23% of organisations say their workforce is fully ready for AI, and 79% of leaders expect the speed of AI to outpace their workforce, governance and operating models (Kyndryl 2026). An Adecco Group study of 2,000 C-suite executives across 13 countries found that only 36% say their talent strategy clearly demonstrates that AI will create opportunities for employees, and only 39% are involving employees directly in redesigning jobs (Adecco Group 2026). In a 2026 benchmark survey of senior data and AI leaders, 93% named culture and change management as the main barriers to AI adoption, far ahead of the technology itself (Lab Manager 2026).

A few things help:

• Say what you know and what you don't. People fill silence with the worst case. If roles will change, say so, and say how the business will support people through it.
• Involve people in redesigning their own work. The people doing a job know where AI would help and where it would make things worse. They are also more likely to use a tool they helped shape.
• Make the approved tools easier to use than the unapproved ones. When the official tools are slow or locked down, people use whatever is on their phone, and the organisation loses sight of where its data is going.
• Fix the process first. As covered above, AI speeds up whatever process it sits on, good or bad.
• Build skills alongside the rollout. Training works best when people can use it on real work straight away.

Further reading: Leading people through AI adoption, The technology company that ran on paper.

How do you know whether change is working?

You know change is working when people's behaviour has shifted and the shift is holding, as well as when the milestones are met. Track leading indicators, such as understanding, confidence and early use of the new way, alongside lagging indicators, such as error rates, cycle times and retention. Leading indicators give you time to act, while lagging indicators tell you whether the change paid off.

The change formula, developed by David Gleicher and popularised by Beckhard and Harris (1987), offers a quick test that holds up well. Change happens when dissatisfaction with the current state, a clear vision of the future, and credible first steps together outweigh the cost and resistance of changing. The three drivers multiply, so if any one is close to zero, the change stalls however strong the others are. Leaders can ask three questions of any change at any point:

1. Do people feel the cost of staying where we are?
2. Can they describe where we're going?
3. Can they see practical first steps they believe in?

A weak answer to any of the three tells you where to put your effort next.

Short anonymous pulse surveys, a few conversations with people at the front line, and a look at whether the old workarounds are still in use will tell you more than a status report coloured green.

Further reading: How to tell whether change is holding.

Which change models are worth knowing?

No single change model covers everything. The most useful approach is to know a handful well and use each for what it is good at: one to sequence the organisational effort, one to understand what individuals go through, and one to diagnose where things stand. The table below covers the models we find most useful.

Treat models as tools, and avoid forcing every change through the same one.

Further reading: Change management models compared.

Why is culture the change beneath the change?

Culture is the set of habits and assumptions that decide how work actually gets done. Most significant changes ask people to behave differently, and that means shifting some part of the culture, whether or not anyone calls it that. Culture changes when leaders consistently reward the new behaviour and stop tolerating the old, especially under pressure.

Edgar Schein described culture in three layers: the visible artefacts, such as offices, processes and dress; the values people say they hold; and the underlying assumptions that guide behaviour without being discussed (Schein 2010). Most change effort goes into the first layer, but whether a change lasts depends on the third.

Values get tested in hard decisions. If a stated value gives way the first time it costs revenue or comfort, people learn what the real values are. We wrote about one example in The gun hire, where a high-performing hire with questionable conduct did lasting damage to a growing company's culture.

Psychological safety matters too. People will only try new ways of working, raise problems early and admit mistakes if they believe they won't be punished for it (Edmondson 1999). Leaders build that belief through how they respond the first few times someone speaks up.

Further reading: Culture Health Check, Purpose, vision and values whitepaper.

What do high-stakes environments teach about leading change?

Environments where mistakes are costly tend to produce simple, practical rules about leading people. Three that carry over well to business change: put people before equipment, do the thinking before you commit, and bring in the people you depend on early. None of them is new, but they are skipped often enough to be worth repeating.

Humans over hardware

In high-performing teams, the people come first and the equipment second. Systems can be bought and skills can be taught, but judgement, trust and the willingness to keep going are much harder to build. Most change programmes put their effort the other way round, because technology and structure are easier to plan and measure. In our experience, change is about ten percent a technical problem and ninety percent a human one, and the effort should follow that split.

Proper planning prevents poor performance

The value of planning is the thinking it forces, more than the plan it produces. Time spent working through what could go wrong, before anyone is committed, is what allows a team to adapt quickly when things change. Leaders who announce a change first and plan it afterwards are running this rule backwards, and the problems that follow are usually predictable.

No comms, no bombs

A small team rarely holds everything it needs. It depends on others for support, and that support works best when those people were involved early and know what is coming. Change works the same way. The project team holds very little of what the change will need: budgets, systems, policies, and the team leaders who will make it work day to day. Bring finance, HR, IT and frontline leaders in at the start, so they are ready when you need them.

Getting help with a change

Every change is different, and the hardest part is often seeing your own situation clearly. Our change management work uses ICAP, our six-step framework built on the research in this guide, and our leadership programmes build the capability your leaders need to carry change themselves.