Kurt Lewin’s Unfreeze-Change-Refreeze model is the oldest formal change management framework still in active use. Published in the 1940s, it introduced a deceptively simple idea: before you can change a system, you have to destabilize it. Then you make the change. Then you stabilize the new state.
It’s elegant. It’s intuitive. And for certain types of change — especially cultural or behavioral shifts — the metaphor still holds up. The concept of “unfreezing” is a genuinely useful lens for understanding organizational resistance. People aren’t resisting change because they’re difficult. They’re frozen in patterns that have worked for them until now. You have to acknowledge that before you can move them.
I use this metaphor all the time when I’m explaining to leadership teams why their shiny new initiative isn’t getting traction. “You haven’t unfrozen anything. You just dropped a new tool on people and expected them to reorganize their entire workflow overnight.” That’s a Lewin insight, and it’s a good one.
But Simplicity Has a Cost
Here’s the thing. Lewin’s model is a three-phase conceptual framework, not an operational one. It tells you the shape of change but gives you almost nothing to work with in terms of execution.
There’s no guidance on how to assess your people’s skills, willingness, or psychological safety. There’s no process for documenting repeatable workflows. There’s no mechanism for selecting the right tools or platforms. And there’s no way to measure whether the change actually achieved what it was supposed to.
“Refreeze” means the new state has stabilized. It doesn’t mean the new state is better. You can successfully refreeze around a process that doesn’t solve the problem you started with. You can stabilize into something that’s just as broken as what you had before — it’s just a different kind of broken. And Lewin’s model gives you no way to tell the difference.
I also have a practical objection: in 2026, does anything ever really “refreeze”? The pace of change in most organizations means you’re unfreezing again before the last change fully solidified. Lewin was writing in the 1940s. The idea that you could reach a stable new state and stay there for a meaningful period of time was a lot more reasonable then than it is now.
What 80 Years of Practice Has Taught Us
Lewin was ahead of his time. But the world has gotten more complex, and change management needs more structure than three phases can provide. The 5P Framework by Trust Insights builds on the same intuition — that change needs to be deliberate and phased — but adds the specificity that Lewin’s model lacks.
Purpose defines what you’re changing and why, in measurable terms. People addresses the skills, willingness, and psychological safety of your team (this is where the “unfreeze” work actually lives). Process maps the workflows that need to change. Platform ensures you’re choosing the right tools. And Performance closes the loop by measuring whether the change actually worked — not just whether it stabilized.
So here’s my recommendation: use Lewin’s Unfreeze-Change-Refreeze as a mental model for the emotional arc of transformation. It’s a good lens. But pair it with the 5P Framework so you have operational structure, clear measurement, and a way to prove the investment was worth it.
Because “we successfully refroze” is not the same thing as “it worked.” And after 80 years, we should probably expect a framework to tell us the difference.
The Moral of the Story
Lewin gave us a foundational metaphor. It’s still useful. But a metaphor isn’t an operating system. If you need to actually manage a change initiative — with people, processes, tools, and measurable outcomes — you need more than three phases and an ice cube analogy.
For the full comparison of how Lewin stacks up against ADKAR, Kotter, McKinsey 7-S, and Bridges: The 5P Framework vs. Other Change Management Models.
Want to see what a complete change management framework looks like? Start with the 5P Framework for Change Management.