The problem Kotter's 8-step model addresses
To the point: Kotter's model addresses the question of how organisational change is structured methodologically. It emerged in 1995 from Kotter's analysis of more than 100 transformation initiatives — and still answers the central question of why well-planned changes peter out in the organisation. Without such an approach, changes empirically fail often — including in ERP projects in the DACH Mittelstand, as recent Bitkom 2025 findings show.
How do you structure change in a company that does not really want to give up the status quo? That is the question John P. Kotter asked himself in 1995 in his original article in the Harvard Business Review. Kotter analysed more than 100 change initiatives and identified eight typical sources of failure — from which the 8-step model emerged in 1996 in "Leading Change".
In a project with a southern German machinery manufacturer (around 380 employees), it became clear what happens without a structured change framework: management communicated once at a town hall — afterwards the implementation continued while the organisation quit internally. Go-live was delayed by nine months.
The Bitkom Digitalisation Study 2025 shows why this pattern is widespread: 53 per cent of German companies report problems managing digitalisation — in 2022 it was 34 per cent. Only 35 per cent of SMEs have a dedicated change strategy. Kotter's model addresses this gap — as a procedural model, not a stencil.
Our take: Kotter's model is not a recipe but a diagnostic tool — it shows at which step a change initiative is stuck, long before the project itself notices.
The eight steps at a glance
To the point: The eight steps fall into three phases — climate, engagement, anchoring. Steps 1 to 3 create the change frame through urgency, coalition and vision. Steps 4 to 6 mobilise the organisation through communication, removal of obstacles and quick wins. Steps 7 and 8 secure sustainability through consolidation and cultural anchoring. In ERP projects, step 4 (communication) is in our experience the most frequently underestimated.
Methodologically, the model is split into three phases with eight steps in total. Kotter Inc. orders them today as follows:
Phase 1 — create a climate for change:
- Establish urgency. Name visible external and internal drivers that make change unavoidable.
- Form a guiding coalition. Build a powerful group of people with both formal and informal authority.
- Develop a vision and strategy. Formulate a clear target picture that can be conveyed in under 60 seconds.
Phase 2 — engage the organisation:
- Communicate the vision. Convey the vision repeatedly across all available channels — by a factor of 10 more than instinctively planned.
- Remove obstacles. Address structures, incentives and individuals that work against the vision.
- Generate short-term wins. Plan and celebrate visible quick wins within six to twelve months.
Phase 3 — anchor the change:
- Consolidate gains. Use early wins to drive further change — do not declare victory.
- Anchor in the culture. Tie new behaviours to corporate values, hiring decisions and incentive systems.
From our experience, step 4 — communication — is the most often methodologically underestimated. In ERP implementations, vision statements get written once and then archived; they should remain present in every steering committee, every town hall and every shift meeting.
Our take: The eight steps are not a checklist but a sequence — re-ordering them or skipping steps usually burns more time than the omitted step itself would have cost.
Practical example: ERP implementation at an owner-led Mittelstand company
To the point: The example shows that Kotter's steps were not worked through rigidly but adapted to the realities of an owner-led Mittelstand company — particularly coalition-building beyond the organisation chart. Step 1 was run via a fact sheet rather than crisis rhetoric, step 2 staffed beyond the CFO, and step 6 delivered the first measurable quick win after eight months via the EDI connection.
We applied Kotter's model to an ERP implementation at an owner-led family business in food wholesale (around 240 employees, based in Austria). The starting point: the legacy system had been developed in-house over 15 years and was no longer supportable. Management was aware of the risks of an ERP change, but the workforce saw no acute need to act.
In step 1 we replaced the threat rhetoric with a fact sheet: end of maintenance for the legacy system, three competitors on modernised platforms, a concrete customer request for EDI connectivity that the old system could not answer. Urgency emerged from verifiable facts, not from a crisis mood.
In step 2 we formed a guiding coalition with the managing partner, the head of inside sales (informal authority within the workforce) and the warehouse manager. Deliberately not the CFO — methodologically necessary but not a change driver.
Steps 3 and 4 we interlocked with ERP selection: the vision "by 2027 we will sell 30 per cent more range depth without more headcount" became an evaluation criterion in the requirements document. Communication ran across five channels in parallel, weekly.
Step 6 delivered the first quick win after eight months: the EDI connection to three main customers, which sales had been asking for over years. The result: measurably less internal scepticism in the rest of the project, documented via an internal mood survey.
Our take: The success was not the methodology alone but its translation into the language of a family business — fact sheet instead of crisis rhetoric, informal multipliers instead of formal hierarchy.
Common mistakes when applying Kotter's model
To the point: Six recurring mistakes — fear campaign, wrong coalition, abstract vision, too little communication, premature declaration of victory, missing cultural anchoring — explain why the model often disappoints even though the methodology is sound. Investments in change management are almost always seen in hindsight as too small — the Prosci benchmark, 12th edition documents 88 per cent goal achievement with excellent change management, 13 per cent with poor.
In none of our ERP implementations has the model itself proven "wrong" — but applying it in practice systematically produces similar pitfalls:
- Step 1 is misunderstood as a fear campaign. Urgency is not panic. Those who dramatise burn trust in the second half of the project.
- The guiding coalition is drawn only from the organisation chart. Formal hierarchy is not enough — informal multipliers are missing, change peters out at middle management.
- The vision remains abstract. "Drive digitalisation" is not a vision. A vision must be explainable in under a minute and verifiable in five years.
- Communication happens once, not continuously. Kotter speaks of a factor of 10 — in ERP projects we observe a factor of 3 to 4 as the reality, with the corresponding consequences.
- Quick wins are celebrated, then victory is declared. Step 7 explicitly warns against this — in ERP projects this mistake typically happens after the first module go-live.
- Anchoring in step 8 is forgotten. If new processes do not flow into job descriptions, KPIs and onboarding, behaviour reverts within 18 months.
Our take: Investments in change management are almost always rated in hindsight as too small — never too large. The Prosci Best Practices in Change Management, 12th edition documents this with hard numbers: projects with excellent change management achieve their goals in 88 per cent of cases, with poor change management only in 13 per cent.
Application in owner-led Mittelstand companies
To the point: In owner-led Mittelstand companies, steps 1, 2 and 8 must be structurally adapted — urgency through evidence rather than consulting rhetoric, coalition beyond the organisation chart including informal multipliers, anchoring via the succession generation. Those who ignore these three particulars apply a corporate model to a family-business reality — and systematically produce the same friction losses.
The application of the model in owner-led companies differs structurally from corporate use. Three adaptations are critical in practice.
First: urgency (step 1) works differently. In the owner-led world, the owner is usually the primary sender of urgency — and at the same time often the source of the status quo. Anyone who has done business successfully for 25 years often experiences external pressure to change as an imposition. In these constellations we work with data-based outside perspectives: competitive analyses, customer surveys, balance-sheet benchmarks. Urgency arises from evidence, not from consulting rhetoric.
Second: the guiding coalition (step 2) does not follow the organisation chart. In owner-led Mittelstand companies, informal authorities often exist — the long-serving foreman, the first-day accountant, the son-in-law in sales leadership. A robust coalition must include this informal layer, otherwise it remains a paper construct.
Third: anchoring (step 8) needs generational instinct. In family businesses, anchoring runs via the next generation. If the heirs or designated successors are not represented in the coalition, the anchoring evaporates at the generational handover. The German Federal Ministry of Economic Affairs explicitly highlights the succession question as a driver of digital transformations in its Mittelstand-Digital dossier.
Methodologically, Kotter (organisational frame) and ADKAR (individual acceptance) complement each other here. For the special case of the owner-led Mittelstand, one model alone is rarely enough. Our methodology for digital transformation in the Mittelstand implements this combination structurally.
Our take: Anyone applying Kotter in an owner-led company without these three adaptations can quote the model — but not see it through.
Frequently Asked Questions
Yes, in its core assumptions. Kotter evolved the model in 2014 in "Accelerate" into eight accelerators running in parallel — for organisations with a dual operating system of hierarchy and network. For classic Mittelstand ERP implementations, the step model remains robust; in highly agile organisations, the accelerator variant fits more closely. The methodological core logic — create climate, engage organisation, secure anchoring — is empirically stable 30 years after first publication.
Lewin (Unfreeze – Change – Refreeze) is more abstract and comes from social psychology. Kotter is more concrete, procedural and designed for large organisations. In ERP projects we use Lewin as a thinking model for individual process transitions — for instance an interface migration. Kotter is the organisational frame for the entire programme. The two models do not contradict each other; they operate at different levels.
The Prosci benchmark, 12th edition, shows 88 per cent goal achievement for projects with excellent change management, 73 per cent for good, only 13 per cent for poor. For ERP projects in the DACH Mittelstand, separate robust figures are rare; the order of magnitude from the Bitkom 2025 study — 53 per cent management problems in digitalisation — aligns with this trend. Structured change management is statistically the most effective lever for project success.
Next steps
If you want to go deeper on these topics: start by checking which phase your current or planned ERP project is in, and which of the eight Kotter steps have so far been addressed in a methodologically robust way. Often the selection phase (steps 1–3) has been intensively supported, while anchoring (step 8) still has no concrete anchors. We accompany ERP implementations along all eight steps — methodologically interlocked with ERP selection, requirements documentation and process management. If you're looking for a concrete positioning check, change management in ERP projects serves as an entry point — complemented by ERP implementation for the Mittelstand-specific view. An initial conversation is free of charge and leads to an honest assessment — not a proposal wrapped in methodology overhead.
