Executives rarely disagree that culture matters. The National Bureau of Economic Research published a survey of 1,348 North American firms in 2017. In it, 92 percent of executives said improving their culture would increase their firm's value, and only 16 percent thought their culture was already where it should be.
Change management, meanwhile, has better evidence behind it than most management disciplines. Prosci's research across more than 2,600 change practitioners found that initiatives with excellent change management met or exceeded objectives 88 percent of the time. Only 13 percent did so under poor change management, a gap Prosci summarizes as roughly seven times the likelihood.
The trouble starts when the two get treated as the same purchase. A company decides its culture is broken and buys a change management methodology to fix it. Or it needs disciplined adoption of a new operating model and gets sold a culture program with a values workshop attached. These are related disciplines with different units of work, different time horizons and different owners, and choosing the wrong one costs a year.
What change management actually is
Change management is the discipline of moving a defined population from a current state to a defined future state, on a schedule, with adoption as the measure of success. It assumes you already know what the end state looks like. Its job is to get people to use it.
That definition matters because it sets the boundary. If you cannot describe the future state concretely enough to train someone on it, you do not yet have a change management problem. Three frameworks dominate the field, and they answer different questions.
Kotter's eight steps
John Kotter's model, set out in Leading Change, runs in eight moves. The first four build the case: create a sense of urgency, build a guiding coalition, form a strategic vision, and enlist a volunteer army. The last four deliver it: enable action by removing barriers, generate short term wins, sustain acceleration, and institute change. It is an organizational model, and its strength is political. It is about building and holding a coalition large enough to survive resistance.
Prosci's ADKAR
ADKAR stands for Awareness, Desire, Knowledge, Ability and Reinforcement. It is an individual model, not an organizational one, and that is the point most summaries miss. It describes the sequence one person goes through, and the organizational result is the sum of those individual transitions.
Its practical value is diagnostic. When adoption stalls, ADKAR tells you which element is missing, and the remedy differs sharply. A desire problem is not solved by more training, and a knowledge problem is not solved by another town hall.
Lewin's three steps
Unfreeze, change, refreeze is the oldest shorthand in the field and is routinely credited to Kurt Lewin. Worth knowing: that attribution is contested. Cummings, Bridgman and Brown argued in Human Relations in 2016 that Lewin never set out such a model, and that it was assembled after his death. Bernard Burnes disputed that reading in 2020.
The three step shorthand is still useful as a reminder that the last phase exists. Most failed change stops at step two.
What culture change actually is
Culture is the behavior an organization rewards, tolerates and punishes. It is not the values on the wall. It is what happens when a senior leader misses a number. It is what happens when one behaves badly and still hits the number. It is who gets promoted when two candidates are otherwise equal.
That makes culture change a different kind of work. There is no go live date and no adoption curve. You are changing the consequences attached to behavior. That means promotion criteria, meeting design, what gets escalated, and who sits in which forum. It also means which behaviors a chief executive tolerates in the people closest to them.
Culture work is mostly systems work plus a small number of visible personnel decisions. The communications are the least of it. Anyone selling culture change as a campaign is selling something else.
Side by side
| Dimension | Change management | Culture change | Cost of picking wrong |
|---|---|---|---|
| Unit of work | A defined transition with a known end state | The consequences attached to everyday behavior | You build a training plan for a problem that has no curriculum |
| Time horizon | Weeks to about eighteen months | Two to five years, with early signals in months | You declare victory at the point the real work starts |
| Who owns it | A program lead with a sponsor | The chief executive and the executive team, personally | Culture work is delegated and quietly dies |
| How it is measured | Adoption, usage, proficiency, benefit capture | Behavior, decision patterns, who gets promoted and who leaves | You measure sentiment and mistake it for change |
| What good looks like | People use the new way without being chased | The behavior persists when no program exists | Behavior reverts the week the program office closes |
| Typical trigger | A system, structure, process or integration | A repeated failure the organization keeps explaining away | The same incident recurs with new names |
Where the two get confused in practice
Three confusions account for most of it. The first is a large operational program labelled a culture transformation because that language raises its profile with the board. It is a process change with a communications plan, and calling it culture inoculates the company against doing the actual culture work later.
The second is the reverse: a genuine culture problem handed to a program office, which does what program offices do. It produces a roadmap, a stakeholder matrix, a set of workshops and a monthly status report. All of it is competent. None of it changes what happens when someone senior behaves badly.
The third is sequencing. An organization attempts culture change while leaving the reward system, the promotion criteria and the meeting rhythm exactly as they were. Employees read the systems, not the messaging, and conclude correctly that nothing has changed.
When change management is the right answer
Pick change management when you can describe the end state in operational detail and the behavior you need is mostly procedural. The test is whether the obstacle is capability or sequencing rather than incentive. New ERP. Post acquisition integration of two finance functions. A shift to a new clinical protocol. A restructure where the reporting lines are already decided.
In these cases, culture talk is a distraction that delays a solvable problem. The work is sponsorship, communication, training, and removing the barriers that make the old way easier than the new one.
When culture is the work
Pick culture work when the same failure keeps recurring under different project names. Or when people can accurately describe the desired behavior and still do not exhibit it, because the current behavior is what gets rewarded.
Safety incidents that keep happening after three safety programs. A stated commitment to speed alongside a decision process requiring eleven signatures. A leadership team that says it wants challenge and promotes only the agreeable. No amount of change management touches any of these, because the future state is not unknown to anyone. It is simply not rewarded.
This is also where the measurement question turns concrete, because the usual instruments were built for engagement rather than behavior. Getting culture measurement right is what tells you whether anything actually moved.
Four ways this choice goes wrong
- Using culture change to avoid a personnel decision. When one executive's behavior is the culture problem, a company wide program is an expensive way of not having a fifteen minute conversation. Everyone below them can see it.
- Treating the survey as the intervention. Measuring culture twice a year and publishing the results changes nothing by itself, and a second survey showing no movement damages credibility further.
- Running both at once with one budget and one deadline. The change program has a date, so it wins every resource conflict. The culture work becomes the thing that slips.
- Making the standards explicit without making them enforceable. A behavior standard that never affects a promotion, a bonus or an exit is a preference. Employees work out which it is within a quarter.
The step almost everyone skips: naming the behavior
Before either discipline can start, somebody has to write down the specific, observable behavior that will be different, at a named level of the organization, in a named situation. Not a value. Not an outcome. A behavior a reasonable observer could confirm or deny from the back of a room.
It gets skipped because it is the uncomfortable part. Writing it down makes the standard enforceable, which means someone senior will eventually breach it and have to be dealt with. The programs that avoid this step are choosing comfort now in exchange for the whole investment producing nothing.
This is also the point at which a leadership competency model stops being an HR artifact and starts doing real work. That requires it to be written in behavior rather than adjectives, and short enough that people remember it.
If the honest answer is that you do not yet know which problem you have, that diagnosis is the first piece of work rather than a reason to delay. Our organizational development practice starts by reading the systems that actually govern behavior, which usually settles the question quickly.
Frequently Asked Questions (FAQs)
Is culture change part of change management?
No, though the two overlap. Change management is a method for delivering a defined transition; culture change alters the consequences attached to everyday behavior and has no fixed end state. Treating culture as one workstream inside a change program is a common and expensive category error, because it gives culture a deadline it cannot meet.
How long does culture change take compared with change management?
A change management effort typically runs from a few weeks to about eighteen months, bounded by the project it supports. Culture change runs two to five years before it is durable, although the first credible signals appear within a few months in the form of different decisions and different promotions. Anyone promising culture change in a quarter is describing a communications campaign.
Can you run change management without addressing culture?
Often, yes. A system implementation, a restructure or a process redesign can succeed on sponsorship and training alone when the existing culture does not actively punish the new behavior. The exception is when the change requires people to do something the reward system penalizes, at which point culture becomes the binding constraint.
Which framework is better, Kotter or ADKAR?
They operate at different levels and are commonly used together. Kotter is organizational and political, strongest on building the coalition and the case. ADKAR is individual and diagnostic, strongest at identifying why a specific group has stalled. Choosing between them is usually a sign that nobody has asked which question they are trying to answer.
Who should own culture change?
The chief executive, with the executive team accountable for the behavior in their own areas. It can be supported by HR and by outside help, but it cannot be delegated to either. The moment culture change has a program manager and no visible owner at the top, employees read it as optional and they are usually right.