HBM 21: Behaviour Change Wheel

A transit agency has completed a thorough COM-B analysis and identified three barriers to increasing bus ridership: residents lack knowledge about routes and schedules (psychological capability), bus stops feel unsafe at night (physical opportunity), and driving is deeply habitual for most commuters (automatic motivation). The diagnosis is clear. But what comes next? How does a planning team move from knowing what is wrong to deciding what to do about it?

This is where many behavior change efforts stall. The barriers are identified, but the selection of interventions is unsystematic — driven by budget availability, political preference, or whatever worked somewhere else. The Behaviour Change Wheel (BCW) was designed to solve this problem by providing a structured, evidence-based process for moving from diagnosis to intervention design to policy implementation.

Why Systematic Intervention Design Matters

In transportation, safety, and sustainability practice, interventions are often chosen before barriers are diagnosed. A city defaults to education campaigns because they are inexpensive. A construction firm defaults to enforcement because it is familiar. A transit agency defaults to fare reductions because they are politically visible. These choices may or may not match the actual barriers.

The consequences of mismatch are significant:

• Resources are wasted on interventions that cannot work because they target the wrong mechanism
• Stakeholders lose confidence in behavior change approaches after repeated failures
• Real barriers remain unaddressed while superficial ones receive attention
• Opportunities for effective, multi-component interventions are missed

The Behaviour Change Wheel addresses this by creating a systematic pathway from diagnosis (COM-B) through intervention function selection to policy category choice. It does not guarantee success, but it ensures that intervention choices are justified by the diagnosed barriers rather than by convenience or assumption.

The BCW Framework Explained

The Behaviour Change Wheel was developed by Susan Michie, Lou Atkins, and Robert West through a synthesis of 19 existing frameworks of behavior change. It is structured as three concentric layers:

The Three Layers

Core Constructs

The Nine Intervention Functions

Each intervention function addresses specific COM-B components. The BCW provides a matrix specifying which functions are appropriate for which barriers:

The Seven Policy Categories

Policy categories are the delivery mechanisms — the institutional and governmental actions that implement intervention functions at scale:

The APEASE Criteria

Not every intervention function that matches a diagnosed barrier is worth implementing. The BCW includes a set of criteria for evaluating candidate interventions:

Causal Logic

The BCW operates through a three-stage causal chain:

STAGE 1: UNDERSTAND THE BEHAVIOR
  ├── Define the target behavior precisely
  ├── Identify the target population
  └── Conduct COM-B analysis
       ├── Capability barriers? → Which sub-components?
       ├── Opportunity barriers? → Which sub-components?
       └── Motivation barriers? → Which sub-components?
STAGE 2: IDENTIFY INTERVENTION OPTIONS
├── Use the COM-B → Intervention Function matrix
│    (e.g., Psychological Capability deficit → Education, Training, Enablement)
├── Generate candidate intervention functions
└── Evaluate candidates using APEASE criteria
STAGE 3: IDENTIFY IMPLEMENTATION OPTIONS
├── Use the Intervention Function → Policy Category matrix
│    (e.g., Environmental Restructuring → Environmental/social planning,
│     Service provision)
├── Generate candidate policy categories
└── Evaluate candidates using APEASE criteria

The logic is that diagnosis constrains intervention choice. If the diagnosed barrier is physical opportunity, then education and persuasion are unlikely to be effective. The appropriate intervention functions are environmental restructuring, restriction, enablement, or possibly incentivisation. This constraint prevents the common mistake of defaulting to the easiest or cheapest intervention regardless of the barrier.

Data Needed

BCW application requires data at each stage:

• COM-B assessment data: From interviews, surveys, focus groups, and observational studies (as described in the COM-B post)
• Intervention mapping exercises: Structured workshops where the design team maps diagnosed barriers to candidate intervention functions using the BCW matrices
• Stakeholder input: Perspectives from policymakers, practitioners, community members, and target populations — essential for APEASE evaluation
• Evidence reviews: Literature on the effectiveness of candidate interventions in similar contexts
• Implementation feasibility data: Budget constraints, institutional capacity, political context, regulatory environment

Methods

• Systematic intervention design: Following the BCW’s three-stage process (understand, identify interventions, identify implementation)
• APEASE evaluation: Structured assessment of each candidate intervention against the six criteria, typically done in stakeholder workshops
• Behavioral specification: Precisely defining the target behavior in terms of who needs to do what differently, in what context, and when
• Mapping matrices: Using published COM-B → Intervention Function and Intervention Function → Policy Category matrices to generate candidates
• Mixed-methods research: Combining qualitative COM-B diagnosis with quantitative evaluation of intervention options

Transportation Example: Increasing Transit Ridership

A regional transit authority wants to increase bus ridership by 25% in a suburban corridor. The team applies the BCW systematically.

Stage 1: COM-B Diagnosis

Interviews and surveys with 200 corridor residents reveal three primary barriers:

Stage 2: Selecting Intervention Functions

Using the BCW matrix, the team identifies candidate functions for each barrier:

BARRIER → CANDIDATE INTERVENTION FUNCTIONS
Psychological Capability deficit:
✓ Education — provide route information, schedule guides
✓ Training — guided first-ride programs, app tutorials
✓ Enablement — simplified fare system, journey planning tools
Physical Opportunity deficit:
✓ Environmental Restructuring — bus shelters, lighting, real-time displays
✓ Enablement — bike-share for first/last mile, on-demand feeder shuttles
✓ Restriction — parking restrictions near transit corridor (push factor)
Automatic Motivation (driving habit):
✓ Environmental Restructuring — make transit more visible, prominent stops
✓ Persuasion — testimonials from new riders, social proof messaging
✓ Modelling — visible role models (employer champions, community leaders)
✓ Incentivisation — free trial passes, employer-matched transit benefits

Stage 3: APEASE Evaluation and Policy Selection

The team evaluates each candidate intervention function against the APEASE criteria:

The Multi-Component Design

The final intervention design is multi-component because the diagnosis identified barriers across multiple COM-B components:

Strengths

Limitations

Best Use Case

The BCW is the right framework when a team has diagnosed behavioral barriers (using COM-B or a similar framework) and needs to systematically select and design multi-component interventions. It is most valuable when:

• The behavior is complex and involves multiple barrier types
• The team wants to avoid defaulting to a single familiar intervention
• Stakeholder buy-in requires a transparent, justifiable selection process
• The intervention must be delivered through policy channels
• There is a need for equity analysis and side-effect evaluation
• The team includes members from different disciplines who need a common process

The BCW is less useful when the behavior is simple, the barrier is obvious, or the intervention is already determined. It is a design framework, not an evaluation framework — it does not tell you whether the intervention worked after implementation.

Key Takeaway

Key References

Exercises and Discussion Questions

The next post introduces the Theoretical Domains Framework — a more granular tool that acts as a magnifying glass for COM-B, breaking its broad categories into 14 specific behavioral domains for detailed barrier analysis.