By Jason Hreha

Glossary

Change Curve

Reviewed by Jason Hreha

Published 4 min read

A company announces a restructure. The next week, HR shares a slide with a curve that drops, bottoms out and climbs again. Its labels run from shock and denial through anger and depression to acceptance. When an analyst objects that the new reporting lines will slow approvals, her manager tells her, kindly, that this is a normal part of the denial stage.

The manager may be right that people react strongly to change. But the analyst made a specific claim about approvals. The curve turned it into a symptom. Where does the curve come from, and what can it show?

What is the change curve?

The change curve is a model of how people respond emotionally to change, drawn as a sequence of stages that dips and then recovers. Later versions also plot morale or performance along the same line, falling during the transition and rising once people accept the change. The best-known version, often called the Kübler-Ross change curve, uses stages adapted from psychiatrist Elisabeth Kübler-Ross’s 1969 book On Death and Dying: denial, anger, bargaining, depression and acceptance. The many variants differ in the number and names of stages but share the shape and the claim of a common emotional path.

From dying patients to employees

Kübler-Ross based On Death and Dying on conversations with terminally ill patients at the University of Chicago’s Billings Hospital. She presented the five stages as chapters about how dying patients came to terms with their diagnosis. Margaret Stroebe and colleagues note that one chapter applied the stages to patients’ families. In 1974 she wrote that the stages applied to grief, change and loss more broadly, and that most of her patients showed two or three stages at once, not always in the same order.

Logan Holdsworth and Todd Bridgman trace what happened next in a 2025 intellectual history. In the 1970s, writers carried the idea to other life events. Psychiatrist Walter Menninger drew Peace Corps volunteers’ changing morale as a curve, and Adams, Hayes and Hopson proposed a seven-phase cycle for all life transitions as a hypothesis they did not test.

From the 1980s the curve moved into organizations. In the 1990s, writers began linking emotions to performance, drawing partly on Katzenbach and Smith’s 1993 team performance curve. In 2002 Elrod and Tippett reviewed 15 models, almost all of which passed through a “death valley” of change. The authors call the result bricolage: a model assembled, mainly by consultants, from whatever pieces served the purpose at hand.

The name itself is a registered trademark. According to the Elisabeth Kübler-Ross Foundation, the Kübler-Ross family’s limited partnership holds it and licenses its use in training materials.

What the evidence shows

Two questions need separating. Do people facing death or bereavement move through these stages? And do employees facing organizational change follow a curve like this?

On the first, research is mixed.

In a 2007 study in JAMA, Maciejewski and colleagues followed 233 bereaved people in Connecticut from 1 to 24 months after a death from natural causes. They measured five indicators: disbelief, yearning, anger, depression and acceptance. Some results ran against stage theory. Disbelief was not the dominant early reaction, and acceptance was the most frequently endorsed item throughout. But after rescaling, the five indicators peaked in the order stage theory predicts.

In a 2017 review in Omega, Stroebe, Schut and Boerner cite published letters that questioned that study’s design. They conclude that little evidence supports a sequential progression through stages, and that expecting stages can harm bereaved people who do not follow them. Among the alternatives they recommend is the dual process model, in which, as Stroebe and Schut described it in 1999, grieving people oscillate between confronting the loss and dealing with the other changes it brings.

On the second question, Holdsworth and Bridgman found little high-quality research validating the change curve in organizations. Elrod and Tippett reported an experimental check of the models in one organization introducing teams. Holdsworth and Bridgman argue that Elrod and Tippett’s broader conclusion, that the dip is the crux of the human change process, rested more on the number of similar models than on rigorous empirical research. Many similar diagrams are not the same as evidence that employees in general follow one path.

When the curve helps and when it misleads

The curve can help when it tells people that strong and mixed reactions to change are common. An employee who feels angry about a reorganization may be reassured that the feeling is not unusual.

It misleads when it is used to predict or interpret what a person will feel. Return to the analyst. Suppose her manager reads her objection as a stage and waits for acceptance, and the approval delays she predicted then arrive on schedule. A manager working without the curve would have asked a simpler question: is she right?

Holdsworth and Bridgman point out that the curve can make resistance unfalsifiable. Any objection can be read as denial, so the model can never be shown wrong.

They also describe a real case at their own university. During a downsizing that cut 230 full-time-equivalent roles, a consultant shared a change curve video with staff. Some staff found it patronizing to be told they would eventually accept a change they considered unnecessary and harmful, and the curve, though well intended, generated cynicism.

The authors trace this blind spot to the curve’s origin. Kübler-Ross’s patients faced death, which cannot be avoided, so coming to accept it brings comfort. Most planned organizational changes, they note, involve choices among options, including keeping the status quo.

Change that condition and the fit improves somewhat. If a site closure has already been decided and announced, there is no open choice left to argue, and the curve’s assumption of eventual acceptance sits closer to its origin. Even then, the grief research is a reason not to assign stages to individuals or to predict their order.

Using the idea carefully

Holdsworth and Bridgman suggest returning to what was useful in Kübler-Ross’s original work: listening closely to people and taking their reactions seriously, without telling them what stage they are in. For the analyst, that means treating her objection as two things at once: a feeling worth acknowledging and a claim worth checking.

The guide to change management models compares the change curve with process models such as Kotter’s 8 steps, Lewin’s change model and ADKAR.

References