Choosing Among ADKAR, Kotter, and Bridges

COMPEL Specialization — AITE-WCT: AI Workforce Transformation Expert Article 18 of 35

A change methodology is a lens, not a procedure. The three methodologies that dominate corporate practice — Prosci ADKAR, Kotter 8-step, and Bridges Transitions — each answer a distinct question about a change programme. They are complements, not competitors, and the first decision the expert makes on any AI workforce transformation is which lens is primary, which are supporting, and where the boundaries lie. The practitioner who adopts one methodology as a matter of vendor certification, or of employer history, will push every problem into its shape and miss the problems that need another lens.

This article sits at the opening of Unit 4 because the entire change-methodology unit depends on this framing decision. Articles 19, 20, and 21 go deep on each methodology in turn; this article teaches the expert to choose among them and to combine them defensibly.

The three methodologies in one paragraph each

Prosci ADKAR frames change as a linear sequence of five states that individuals move through: Awareness (of the need for change), Desire (to participate), Knowledge (of how to change), Ability (to implement the change), and Reinforcement (to sustain the change). The methodology is at its strongest when the change is clearly definable, the population is identifiable, and the question is “how do we move this population through the transition.” It offers a diagnostic (which stage is blocking) and an intervention map (what action unblocks which stage). The methodology has been developed by Prosci since the late 1990s and is now the most-common corporate change certification; the original research base is open-methodology but the training assets are commercial. Source: https://www.prosci.com/methodology/adkar.

Kotter 8-step frames change as an organisational sequence that leadership executes: create urgency, build a guiding coalition, form a strategic vision, enlist volunteers, enable action, generate short-term wins, sustain acceleration, institute change. The methodology is at its strongest when the change is strategic, cross-functional, and requires enterprise-scale coordination rather than individual-level transition. John Kotter’s 1996 Leading Change is the canonical reference; the 2012 Harvard Business Review article Accelerate! is an evolved framing that emphasises dual-operating-system thinking. Source: https://hbr.org/2012/11/accelerate.

Bridges Transitions frames change as a psychological process that distinguishes the external event (the change) from the internal process (the transition). The transition has three phases: Ending (letting go of what was), the Neutral Zone (the uncomfortable in-between), and the New Beginning (commitment to what will be). The methodology is at its strongest when the change involves loss — role change, redundancy, identity shift — because it treats the psychological work as the central work rather than as a communications sidebar. William Bridges’s Managing Transitions (1991, 4th edition 2017) is the canonical reference. Source: https://wmbridges.com/resources/.

These are not the only three methodologies. McKinsey’s 7S, the Beckhard change-formula, Conner’s Managing at the Speed of Change, and numerous more recent practitioner frameworks are also in the field. The three below dominate because they are complementary, collectively cover the scope of change work, and have each generated practitioner certification markets that reinforce availability.

The choice framework

The expert’s choice framework has six criteria. Working through them produces a defensible selection.

1. What is the unit of analysis?

• ADKAR’s unit is the individual. Good for: literacy roll-out, process change for a defined population, adoption campaigns. Weak for: strategic repositioning, enterprise-wide structural change.
• Kotter’s unit is the organisation. Good for: strategic transformation with enterprise scope. Weak for: individual-level adoption questions.
• Bridges’s unit is the person in transition. Good for: role change, redundancy, identity-level change. Weak for: adoption campaigns where the identity of the individual is not the primary issue.

For AI workforce transformation, the unit shifts across the programme. Literacy roll-out is individual (ADKAR). The strategic frame (Article 1) is enterprise (Kotter). Role redesign (Articles 24–29) involves identity (Bridges). A competent methodology choice names the unit shift and assigns lenses accordingly.

2. What is the dominant blocker?

• If the dominant blocker is “people don’t know how”, ADKAR’s Knowledge-and-Ability stages are the primary lens.
• If the dominant blocker is “leadership isn’t aligned”, Kotter’s Coalition step is primary.
• If the dominant blocker is “people aren’t letting go”, Bridges’s Ending phase is primary.

The expert diagnoses the blocker before the methodology. Asking “which change methodology should we use” before asking “what is actually blocking us” is the common failure.

3. Over what time horizon?

• ADKAR works at 3–12 month horizons (a defined campaign).
• Kotter works at 18–36 month horizons (strategic transformation).
• Bridges works at the pace the transition itself sets, which can be weeks to years depending on the depth of the identity work.

AI workforce transformation runs at 2–5 years total; the methodology mix must accommodate the full horizon. A pure-ADKAR framing will outrun its horizon and produce fatigue. A pure-Kotter framing will underserve the individual-level work. A pure-Bridges framing will skip the strategic enablement.

4. How visible is the sponsor?

• Kotter requires a visible sponsor coalition. If the executive sponsor cannot commit to visible, sustained, public commitment, Kotter will not work.
• ADKAR can run with a less-visible sponsor as long as the line-management layer is active.
• Bridges is largely pastoral and can run even under sponsor-absent conditions, though sponsor legitimacy helps.

5. What is the loss content?

• Bridges is built for loss. Role change, redundancy, identity shift — Bridges is the primary lens.
• ADKAR can address loss through the Desire stage but does not centre it.
• Kotter largely ignores the psychological work; its Leading Change chapter on overcoming resistance is the weakest part of the original canon.

AI transformations frequently involve loss. The expert who chooses ADKAR or Kotter alone, in a programme with heavy loss content, will see the loss surface as resistance in Article 23’s framing and will be disarmed in responding to it.

6. What does the sponsor expect to buy?

• Sponsors familiar with Prosci will expect ADKAR. The expert must meet that expectation with a clear explanation of where ADKAR is and is not primary.
• Sponsors familiar with Kotter (typical in strategy-led executive teams) will expect Kotter. Same requirement.
• Sponsors unfamiliar with Bridges are common. The expert’s job is to introduce Bridges where its use is warranted, with a short, plain-language briefing.

The political dimension is not trivial. Methodology choice also signals about how the expert engages. Openness to combination across methodologies is a signal of maturity; dogmatic advocacy of one is usually a signal of recent certification.

The combination patterns that work

The three methodologies combine gracefully in three recurring patterns.

Pattern A — Kotter for the programme, ADKAR within each workstream. The strategic transformation runs under Kotter’s 8-step (coalition, vision, short-term wins, etc.). Each workstream — literacy, talent pipeline, role redesign — runs a defined ADKAR campaign. The two methodologies interlock: Kotter provides the strategic scaffold, ADKAR provides the operational workstream design. This pattern is appropriate for most full AI workforce transformations.

Pattern B — Bridges overlaid on role-change workstreams within a Kotter/ADKAR programme. Where role redesign or redundancy is part of the programme, Bridges is the lens applied to those workstreams specifically. Communications, manager conversations, and timing decisions are all filtered through the Ending / Neutral Zone / New Beginning frame. Learners are not assumed to be at the same place; segmentation is by transition phase.

Pattern C — Bridges alone for a localised identity change. Where the programme is narrow (a specific professional community re-identifying its role, e.g., senior analysts becoming AI-augmented advisors), Bridges alone can be sufficient. ADKAR and Kotter add overhead without adding value.

The pattern the expert avoids: every methodology at once. A programme that tries to run ADKAR, Kotter, and Bridges simultaneously at full strength produces ceremony fatigue and confuses learners about which framework governs their experience.

The dogmatic-methodology failure

The dogmatism failure has a recognisable shape. Its signs include: the sponsor or change-management lead has a single methodology certification and treats the certification as methodology; the plan templates are the certification’s templates rather than the programme’s; the language of the programme is the methodology’s language rather than the organisation’s; the methodology’s framework diagrams are used in executive communications without translation into the organisation’s strategic frame. Every sign is a symptom of a methodology being used as identity rather than as lens.

The remedy is to hold the methodology as a tool and the programme as the customer. The expert’s communications are in the organisation’s voice; the methodology appears in the internal planning artefacts and is translated out in learner-facing materials. The methodology frame of the week is not in the CEO’s briefing; the programme’s substantive progress is.

Case — a hybrid in practice

A European financial-services institution we reviewed ran a 24-month AI workforce transformation using Pattern A with Bridges overlay. Kotter’s 8-step provided the strategic sequence: an early “burning platform” communication from the group CEO (urgency), a guiding coalition that named Head of AI, CHRO, and three business-unit heads (coalition), a strategic vision articulated in the group strategic plan (vision), and a short-term-wins cadence at quarterly intervals. Within the programme, the literacy workstream ran a defined ADKAR campaign (Awareness — group-wide communications; Desire — line-manager conversations; Knowledge — curriculum delivery; Ability — applied practice; Reinforcement — post-training coaching cadence). The role-redesign workstream, affecting about 400 analysts, used Bridges overlaid on the ADKAR campaign: each analyst’s transition was mapped through Ending (letting go of the pre-AI advisory role), Neutral Zone (the uncomfortable 3–6 months of working with AI tools they did not yet trust), and New Beginning (the AI-augmented advisor identity).

The hybrid did not run without friction. Early-stage communications tried to be Kotter and ADKAR at once and became noisy; the programme team redesigned the cadence to separate the strategic communications (quarterly) from the workstream-level communications (monthly) from the transition communications (per-cohort, timed to each cohort’s Bridges phase). The redesign materially improved learner clarity. The lesson carried forward: methodology integration is a design task in itself, not a by-product of choosing multiple lenses.

Two real-world anchors

Prosci’s methodology evolution and its limits

Prosci’s ADKAR methodology, developed by Jeff Hiatt and colleagues since the late 1990s, is documented in Hiatt’s ADKAR: A Model for Change in Business, Government and Our Community (Prosci Learning Center Publications, 2006, updated 2014) and in Prosci’s ongoing research benchmark reports. The methodology has been evolved by Prosci to include change-saturation (§22) frameworks and sponsor-activity models, but the core five-stage sequence has remained stable. Source: https://www.prosci.com/methodology/adkar.

The methodology’s limits — well-acknowledged by serious practitioners — are in organisational change that is not reducible to individual transition. An AI-transformation programme that has to renegotiate the psychological contract between workforce and organisation, redesign the performance-management system, and simultaneously reshape executive narrative is larger than ADKAR covers. A Prosci-certified change lead who treats the methodology as sufficient will need to borrow lenses; a Prosci-certified lead who uses ADKAR as the individual-transition lens within a larger frame is using it well.

Kotter’s 2012 Accelerate as an AI-era update

John Kotter’s 2012 Harvard Business Review article Accelerate! revised the 8-step framework by emphasising simultaneous operation of two systems: the hierarchical organisation that runs steady-state operations, and a networked, volunteer-driven system that runs change. The revision directly addresses a problem the original 1996 model under-treated: change that runs continuously in a complex environment, rather than as a one-time event. Source: https://hbr.org/2012/11/accelerate.

The 2012 revision is particularly apposite for AI transformations, which are continuous in character — AI system portfolios evolve; regulatory landscapes shift; workforce expectations move. A Kotter-based transformation designed with only the 1996 framing risks treating the transformation as a finite event and running out of steam before the underlying continuous change settles. The 2012 dual-system framing remains, fifteen years on, the strongest Kotter framing for sustained organisational change.

Learning outcomes — confirm

A learner completing this article should be able to:

• Distinguish ADKAR, Kotter, and Bridges by unit of analysis, dominant blocker, time horizon, sponsor visibility, loss content, and sponsor expectation.
• Apply the six-criterion choice framework to a described AI workforce programme and produce a defensible methodology selection.
• Name the three combination patterns (Kotter + ADKAR; Kotter/ADKAR with Bridges overlay; Bridges alone) and the contexts each fits.
• Identify the dogmatic-methodology failure and argue for methodology as lens rather than as identity.
• Argue why continuous AI change warrants Kotter’s 2012 dual-system framing over the 1996 event framing.

Cross-references

• EATF-Level-1/M1.6-Art05-Change-Management-for-AI-Transformation.md — Core Stream change-management anchor.
• EATE-Level-3/M3.2-Art05-Enterprise-Change-Architecture.md — enterprise-scale change architecture.
• Articles 19, 20, 21 of this credential — methodology-specific deep dives.
• Article 22 of this credential — change saturation and pacing.
• Article 23 of this credential — resistance analysis.

Diagrams

• Matrix — methodology (ADKAR × Kotter × Bridges) × six choice criteria; cells describe fit.
• BridgeDiagram — programme need on the left (individual adoption / enterprise strategic / identity transition), methodology on the right, with “combination pattern” annotations for mixed needs.

Quality rubric — self-assessment