ADKAR, the Book: Hiatt's Model of Individual Change, and What the Research Says

What the book argues

ADKAR: A Model for Change in Business, Government and Our Community is by Jeffrey M. Hiatt. Prosci published the first edition in 2006 (Open Library, n.d.; Bookshare, n.d.). I could not find a current publisher's page for this title: Prosci's online bookstore lists its other books but not this one (Prosci, n.d.-b).

The model is an acronym. Prosci names its five elements as Awareness, Desire, Knowledge, Ability and Reinforcement, and states the premise that organizational change happens one person at a time (Prosci, n.d.-a). The synopsis in an accessible-books library record describes the book as presenting ADKAR as a simple, results-oriented model that pulls together readiness assessments, sponsorship, communications, coaching, training and resistance management (Bookshare, n.d.).

In my words: the book asks a change leader to stop treating the organization as the unit of change and to ask, person by person, which of five things is missing. For more on how the model works and where it came from, see the encyclopedia entry on ADKAR.

Who it is for

Change practitioners, and managers in organizations that already use Prosci's methods. It is useful for a manager who has one or two people not adopting a change and wants a structured way to ask why. It says less about how a whole system of teams changes, which is what a model like Kotter's is aimed at.

What the research says

I found no controlled study that tests ADKAR as a model in the evidence I use, which matches what the ADKAR encyclopedia page found in the published research. What can be checked is the evidence behind some of its building blocks.

Knowledge and Ability: training helps, more for knowledge than for results. Collins and Holton (2004) pooled 83 studies of managerial leadership training. According to their abstract, effects were large on knowledge, moderate to large on expertise, and smaller on organizational results. For knowledge outcomes, the most effective programs were in studies that measured one group before and after, with no control group. That fits the model's separation of knowing how from being able to do it.

Reinforcement: follow-through. Two trials come close to this, though neither tested follow-through after a change launch. Haunstrup and Jensen (2024) randomly assigned 226 managers in Danish job centers to leadership training plus short, timed reminders, or to no training, and eight months later employees rated the trained managers higher on transformational leadership; the design cannot separate the training from the reminders. And Grill (2025) randomized 59 managers in one municipality to training in giving their staff performance feedback, or to a waiting list (49 managers stayed in the analysis). Employees of trained managers reported more improvement in their managers' feedback, with no significant slowing over the 18 months measured. Both fit the idea that a change needs something that keeps it going after the launch, but neither tested it.

Desire: readiness goes with how change is led. Peng and colleagues (2021) pooled 30 studies. According to their abstract, transformational leadership went with more readiness and openness to change and less resistance, though the links were weaker for some outcomes in studies that followed people over time. These are correlations.

Measuring readiness is harder than it looks. Miake-Lye and colleagues (2020) reviewed survey instruments for organizational readiness in health care. According to their abstract, 897 of 1,370 items (about two thirds) were about the setting inside the organization, such as climate, communication and culture, and most instruments were tailored to a specific change. That is a caution for anyone who uses a person-by-person readiness score as if it measured the whole picture.

Where the evidence is thin

The model's sequence, the claim that the five elements build in order and that the first missing one is the place to act, has not been tested in the evidence I use. Most of the supporting studies test practices that resemble one element, not the model. And the book's synopsis says the model came out of research without saying which, and I found no independent test of it.

How it maps to the Unfrozen Loop

This mapping is my judgment. ADKAR focuses on the individual, while the Unfrozen Loop focuses on the manager and the team around that person. Awareness and Desire overlap with Make sense together, top included, where the team works out what a change means for their own work. Reinforcement is close to Review on a rhythm, then hold the pace, which asks for short, regular reviews that keep running after launch. The Loop adds what ADKAR leaves to the side: the manager's own load, in The middle's own weather.

What I take from it

ADKAR is a good diagnostic question with thin evidence as a model. Its strongest part is the one the research supports best: reinforcement. If you use it, pair it with a rhythm. Check each person's sticking point once, then come back every two weeks for two months to see whether it moved. For a comparison with Kotter, see Kotter vs ADKAR, and for more books, the change management reading list.