Change management / transformation
Kotter's 8-Step Change Model
John Kotter's sequence for leading large-scale change: build urgency, form a guiding coalition, create and communicate a vision, empower action, generate short-term wins, consolidate, and anchor the change in the culture.
His core claim: transformations fail not because of bad strategy, but because steps get skipped.
- Problem
- Change management / transformation
- Altitude
- Enterprise
- Effort to run
- Moderate
- Evidence base
- Established
Theory & origin
Kotter published the model in Leading Change (1996), based on a study of over 100 transformation efforts, most of which had failed. His diagnosis was sequence errors: declaring victory too early, communicating the vision once instead of a hundred times, or launching a change with no real urgency behind it. The eight steps are deliberately ordered, and each one builds the condition the next one assumes. Critics note that it is top-down and linear, where real change tends to loop, and Kotter himself later reframed it as concurrent "accelerators." But as a way to diagnose why a change is stalling, the original sequence is still the sharpest tool available.
Key components
The parts at a glance. Click any term for the full definition, a field example, and the common failure, in the model below.
Explore the model
How a consultant runs it
- 01 Diagnose before you prescribe: score the change against all eight steps and find the earliest one that got skipped. That is usually where the stall actually lives.
- 02 Test urgency honestly by counting what leaders actually do differently, not what the kickoff deck said.
- 03 Audit the coalition for real power. A steering committee made of delegates is step 2, failed.
- 04 Engineer a visible win inside 90 days and spend it publicly. Wins are the fuel for the long middle of a change effort.
- 05 Do not close the program until the behavior survives without it. Anchoring, step 8, is where most "successful" changes quietly die.
When to use
- 01 Diagnosing why a large transformation is stalling, by locating the skipped step
- 02 Planning a change with real behavioral stakes: an ERP rollout, an operating model shift, a merger integration
- 03 Coaching a sponsor who thinks a kickoff and a comms plan add up to change management
When not to use
- 01 Small, reversible changes, where the ceremony outweighs the actual risk
- 02 As a strictly linear checklist. Real change loops between steps. Use it as a diagnostic, not a project plan.
- 03 Bottom-up or emergent change, where forcing a top-down sequence smothers what is already working
Worked example
An ERP rollout is 9 months in and stalling: training is complete, but adoption sits at 30%. Scoring the eight steps finds that step 1 and step 5 both failed.
Nobody below director level believes the old system is actually going away, and warehouse incentives still reward throughput measured in the legacy tool. The fix is not more training.
The CFO announces a legacy shutdown date (urgency), incentives get re-based on the new system's numbers (empowerment), and one distribution center goes fully live as the engineered win. Adoption hits 85% within two quarters.
Common pitfalls
- 01 Declaring victory at go-live, Kotter's original failure case, and watching the change roll back within a year
- 02 Vision by committee: a paragraph of abstractions nobody can repeat, let alone act on
- 03 Skipping urgency because the executive team feels it, when the shop floor does not
- 04 Running all eight steps as a communications campaign while systems and incentives still reward the old behavior
Sample deliverable
One real engagement, start to finish. Watch the numbers travel from raw input, onto the chart, into the finished artifact.
Input
- Urgency (felt below director)1.8 / 5
- Coalition power3.6 / 5
- Vision clarity3.1 / 5
- Empowerment (incentives)1.6 / 5
- Short-term wins2.2 / 5
Process
Interviews score each step from 1 to 5, and the earliest weak step is where the change is stalling
Change-readiness scan: core-banking migration
- Stallsteps 1 and 5, not training
- Fixshutdown date plus re-based incentives
- Engineered winone DC live in 90 days