COM-B and the Behavior Change Wheel: S-Tier Behavioral Designer’s Guide

Every behavior-change project that fails starts the same way: the team picks an intervention before diagnosing what is actually broken. Education campaigns are shipped to people whose problem is opportunity. Incentive programs are launched at people whose problem is capability. Habit-loop redesigns are aimed at problems that are really about reflective motivation. The intervention is fluent. The diagnosis is wrong. The outcome is a six-figure pilot that nobody wants to talk about at the post-mortem.

Susan Michie, Lou Atkins, and Robert West at University College London built the COM-B model and the Behavior Change Wheel in 2011 to make that mistake structurally harder. COM-B says any behavior is the product of three ingredients: Capability, Opportunity, and Motivation. The Wheel surrounds that hub with nine intervention functions and seven policy categories, so the question shifts from “what should we do” to “which of the three is actually deficient, and which intervention class fits the deficit.” That single move, done well, will save more behavior-change budgets than every clever piece of copywriting in the field.

This pillar is the version of COM-B I wish every product manager, public-health planner, and behavioral-design lead had read before their first pilot. It covers what Michie and her co-authors got right, where teams get the framework wrong in practice, and how to wire the Wheel into the Octalysis Framework so the diagnosis layer flows cleanly into a specific Core-Drive design surface and a specific Behavior Change Technique. By the time you finish, you should be able to look at any stuck behavior, name which of the six COM-B sub-components is bleeding, and pick an intervention class with evidence behind it rather than vibes.

Speed Run Notes

• COM-B says behavior happens when Capability (physical + psychological), Opportunity (physical + social), and Motivation (automatic + reflective) align. Subtract any one of the six sub-components and the behavior collapses.
• The Behavior Change Wheel surrounds COM-B with nine intervention functions and seven policy categories. The point is to stop picking an intervention before diagnosing which deficit is actually blocking the behavior.
• The dominant failure mode is intervention-first thinking. Teams pick Education or Incentivisation by reflex, then back-fit a COM-B story. Real COM-B work runs the diagnosis to completion before any tool is chosen.
• Each of the six COM-B sub-components maps cleanly to an Octalysis design surface: Capability to Core Drive 3, Physical Opportunity to subtractive design (Sludge inverse), Social Opportunity to Core Drive 5, Reflective Motivation to Core Drives 1 plus 2 plus 4.
• Capability and Motivation are not independent. Mastery experience builds self-efficacy and self-efficacy is half of Reflective Motivation, so a single Core Drive 3 design surface often moves two COM-B components at once.
• BCT Taxonomy v1 (93 named Behavior Change Techniques, Michie 2013) is the missing layer most COM-B users skip. It is what turns “use Education and Modelling” into a list of concrete moves a team can actually ship next week.

Author Credibility: Yu-kai Chou

Yu-kai Chou created the Octalysis Framework after studying gamification since 2003, years before the term entered mainstream vocabulary. As a Human-Systems Architect & Behavioral Designer, his framework has been applied by LEGO, Microsoft, Porsche, Coca-Cola, Salesforce, and MrBeast, impacting over 1.5 Billion Users.

Chou has taught the Octalysis methodology at Harvard, Stanford, Yale, Tesla, Google, BCG, and IDEO.

His work has been cited by Harvard, Stanford, MIT, Forbes, Wall Street Journal, Wired, US Department of Energy, NIST, NSF, NCBI, US Department of Education, ClinicalTrials.gov, and Google Scholar — with 3,700+ more academic publications. Explore his books here.

TEDxLausanne · Gamification to Improve Our World17 min

What Is COM-B and the Behavior Change Wheel?

COM-B is a model of behavior published by Susan Michie, Lou Atkins, and Robert West in 2011 in the journal Implementation Science. The acronym stands for Capability, Opportunity, Motivation, and Behavior. The claim is simple and load-bearing: for any voluntary behavior to occur, the person must have the Capability to perform it, the Opportunity for it to occur, and the Motivation to do it more than competing behaviors at that moment. Take any one of those three away and the behavior does not happen.

The Behavior Change Wheel is the intervention layer wrapped around COM-B. The inner ring is the three COM-B components. The middle ring is nine intervention functions: Education, Persuasion, Incentivisation, Coercion, Training, Restriction, Environmental Restructuring, Modelling, and Enablement. The outer ring is seven policy categories: Communication and Marketing, Guidelines, Fiscal Measures, Regulation, Legislation, Environmental and Social Planning, and Service Provision. Each intervention function has empirical evidence linking it to specific COM-B components, and each policy category supports a subset of intervention functions. The structure is what turns COM-B from a diagnostic into a planning instrument.

The Wheel sits on top of a third layer most users skip: the Behavior Change Technique Taxonomy v1, published by Michie and 18 co-authors in 2013, which catalogs 93 reliably labeled techniques (Goal Setting, Action Planning, Feedback on Behavior, Self-Monitoring, Social Support, Habit Formation, and so on). The Wheel tells you which intervention class fits the deficit. The taxonomy tells you which specific techniques fit the intervention class. Together they collapse the gap between “we need to change behavior” and “we will ship these five concrete moves on Monday.”

The Three Ingredients (Six Sub-Components)

Each of COM-B’s three ingredients splits into two sub-components, and the split is the part most introductions skip. If you cannot tell a Physical Capability deficit from a Psychological Capability deficit, your prescribed intervention will hit the wrong part of the person.

Capability: Physical and Psychological

Physical Capability is the bodily skill, strength, or stamina required to perform the behavior. A person who has not been trained to administer an inhaler correctly has a Physical Capability deficit, regardless of how motivated they are.

Psychological Capability covers the knowledge, comprehension, and mental skill required: understanding why the inhaler matters, remembering the steps, regulating attention long enough to use it correctly during an asthma attack. A patient who can demonstrate inhaler technique perfectly in clinic but freezes during a real attack has a Psychological Capability deficit at the regulation layer.

Opportunity: Physical and Social

Physical Opportunity is whatever the environment affords. Does the person have time, money, the right tools, accessible space? A worker who wants to take the stairs but works in a tower where the stairwell door requires a security badge they do not have has a Physical Opportunity deficit, no matter how virtuous their intention.

Social Opportunity is the cultural and interpersonal context: what is normative in the person’s reference group, what cues are present, who they will be seen by. A teenager who wants to skip vaping but whose friend group treats refusal as a status hit has a Social Opportunity deficit.

Motivation: Automatic and Reflective

Reflective Motivation is the planful, evaluative layer: intentions, beliefs about consequences, identity, plans. This is the part conventional behavior change targets when it asks people to set SMART goals or sign commitment letters.

Automatic Motivation is the habit-and-emotion layer: cue-triggered impulses, affective associations, learned drives. This is what makes a smoker reach for a cigarette before reflective motivation has a chance to weigh in. Most interventions fail because they target Reflective Motivation when the deficit is Automatic. People do not eat well because they decide to. People eat well because the kitchen is structured so the salad is closer than the cookie.

The diagnostic move is to ask, for any stuck behavior, which of these six sub-components is actually missing. Almost every behavior-change project that fails could have been saved by a clear answer to that question in the first week of design.

The Wheel: 9 Intervention Functions and 7 Policy Categories

The middle ring of the Behavior Change Wheel translates a COM-B deficit into a class of intervention. The mapping is well-evidenced and was derived from a systematic review of 19 existing behavior-change frameworks. The nine functions:

The Nine Intervention Functions

1. Education: Increase knowledge or understanding. Fits Psychological Capability deficits and Reflective Motivation deficits where the gap is “I did not know.”
2. Persuasion: Use communication to induce positive or negative feelings or stimulate action. Fits Reflective and Automatic Motivation.
3. Incentivisation: Create expectation of reward. Fits Reflective Motivation (planning around the reward) and Automatic Motivation (conditioning over time).
4. Coercion: Create expectation of punishment or cost. Fits both Motivation sub-components; carries the well-known Black Hat trade-offs around reactance and reduced intrinsic motivation.
5. Training: Impart skill. Fits Physical and Psychological Capability.
6. Restriction: Use rules to reduce opportunity for the unwanted behavior. Fits Physical and Social Opportunity.
7. Environmental Restructuring: Change the physical or social context. Fits Physical and Social Opportunity, and Automatic Motivation through cue redesign.
8. Modelling: Provide examples for people to aspire to or imitate. Fits Social Opportunity and Automatic Motivation.
9. Enablement: Increase means or reduce barriers beyond Education and Training (medication, prosthetics, behavioral support). Fits Physical Opportunity and both Capability sub-components.

Notice how cleanly the deficits map. Once you have diagnosed which COM-B sub-component is bleeding, the candidate intervention set is usually two to four of these nine, not all nine. The Wheel narrows the question fast.

The Seven Policy Categories

The outer ring covers the system-level levers that support the intervention functions: Communication and Marketing, Guidelines, Fiscal Measures, Regulation, Legislation, Environmental and Social Planning, and Service Provision. These are the categories where “we want to run a Modelling intervention” becomes “we are going to fund a national PSA campaign showing the modelling behavior” or “we are going to mandate display of role-model behavior in clinical settings.” For consumer product teams the policy ring matters less, but for public-health planners and behavioral-insights units inside government it is where most of the actual money moves.

What Michie and Colleagues Got Right

COM-B is not the most cited behavior-change theory in the world, but in my view it is the most useful one a working designer can hold. Three reasons.

Diagnose Before Prescribing

The dominant failure mode in behavior change is treatment before diagnosis. Every other behavior-change framework I know either jumps straight to a tactic (the Hook Model is great if you already know the deficit is Automatic Motivation) or insists you understand a particular construct without telling you when it is the binding constraint (Theory of Planned Behavior is excellent on intention formation but says nothing about when intention is the bottleneck). COM-B forces the diagnostic question to the top of the workflow and refuses to move until it is answered. That single structural decision is worth more than most of the cleverness in the field.

Exhaustive Without Being Bloated

Michie’s team built COM-B by reviewing 19 existing behavior-change frameworks and synthesizing them down. The six sub-components are not a random set; they are what survived when redundancy was stripped out. You cannot point to a behavior and say “COM-B does not cover this” because the framework was constructed to cover everything voluntary by definition. The Wheel and the BCT Taxonomy then add resolution without adding new top-level constructs. The architecture is deliberately layered: diagnostic at the top, intervention class in the middle, specific technique at the bottom.

An Empirical Back End

The 93-item BCT Taxonomy v1 is built on reliability studies that established whether independent coders could agree on what counts as which technique. The intervention-to-COM-B mappings in the Wheel are derived from evidence on which intervention functions actually move which components. None of this is glamorous, but it is the difference between a framework that works in the article and a framework that works in a budgeted program. Newer COM-B-based interventions can reference BCTTv1 codes in their pre-registrations, which makes meta-analysis possible in a way that pre-2013 behavior change literature could not support.

Where COM-B Falls Apart

If COM-B were perfect, nobody would still be shipping the wrong intervention. The framework has three structural soft spots that bite in practice.

Capability and Motivation Are Tangled

Bandura’s self-efficacy work showed decades ago that beliefs about one’s capability feed back into motivation. If a person tries the behavior, fails, and concludes “I cannot do this,” their Reflective Motivation drops even though the underlying Physical or Psychological Capability did not change. COM-B’s clean separation of Capability and Motivation is theoretically convenient but practically fragile, which means a pure “fix Capability” intervention often has to also handle the motivation downstream of the early failures. The Octalysis answer here is to design Core Drive 3 (Empowerment of Creativity & Feedback) mastery loops at low enough stakes that early failures do not cost motivation, which makes the Capability-Motivation handoff cleaner.

Opportunity Is Often the Hidden Constraint

Teams diagnose Capability or Motivation deficits when the binding constraint is actually Opportunity. The classic example is a healthy-eating intervention aimed at low-income workers who lack time to cook. No amount of cooking education or motivational messaging fixes a 70-hour work week. The framework explicitly includes Opportunity, but the people running interventions are usually optimistic about the leverage they have over personal choice and pessimistic about the leverage they have over structural conditions. So Opportunity gets quietly demoted in the diagnostic step. This is not the framework’s fault, but it is a predictable point of practitioner failure.

The Wheel Implies More Independence Than Real Interventions Have

The nine intervention functions are presented as discrete options. In practice they bundle. A serious smoking-cessation program runs Education plus Persuasion plus Training plus Environmental Restructuring plus Enablement, often simultaneously, often with the same staff. The Wheel’s neat separation can mislead a new team into picking “one intervention function” when the right move is a stacked program. The corrective is to read Michie’s own implementation guidance, which is explicit about stacking; but the visual of the Wheel pushes in the other direction.

The Wheel Underweights Context-Specificity

One last structural limit: the Wheel is mostly silent on cultural and contextual moderators of which intervention class works. A Modelling intervention in a high-individualism culture and a Modelling intervention in a high-collectivism culture are very different beasts even though both fit the same Wheel slot. The framework assumes the practitioner will apply local knowledge; in practice, large public-health programs are designed once and deployed across contexts where the moderators differ. The corrective is to treat the Wheel as a planning instrument plus a context-adaptation checklist, not as a one-shot intervention generator. The most successful national-scale rollouts I have seen explicitly task a localization team with re-running the COM-B diagnostic in each cultural context before adapting the intervention package.

What is Really Happening Inside the Brain

COM-B is descriptive, not neurobiological. The framework does not claim a brain map. But the six sub-components correspond, loosely, to dissociable systems neuroscience has been studying for decades.

Physical Capability lives in the primary motor cortex, cerebellum, and basal ganglia procedural-learning circuits. Skill is built by repeated execution that gets consolidated during sleep through cortico-striatal replay. This is why Training and Enablement work: they create the substrate for skilled action.

Psychological Capability draws on the dorsolateral prefrontal cortex for working memory and rule application, plus the medial prefrontal cortex for conceptual knowledge. Education and Training both load this system. Crucially, executive function is a finite resource within any session, so interventions that demand high Psychological Capability “in the moment” of behavior fail even when the person could ace the same task in a quiet room.

Reflective Motivation is mostly ventromedial prefrontal cortex (subjective value computation), with anterior cingulate and dorsolateral PFC handling cost-benefit weighing and effort discounting. This is the system that responds to Persuasion arguments and Incentivisation framing.

Automatic Motivation is dorsal-striatum habit circuitry and the limbic-affective system. Habits live in the dorsolateral striatum after the prefrontal cortex offloads them; emotional responses originate in the amygdala and ventral striatum and reach action through the basal ganglia. This is why Environmental Restructuring outperforms Persuasion for entrenched behaviors: it changes the cues feeding into a brain system that does not particularly listen to Reflective Motivation arguments.

The practical lesson is that an intervention’s likely success depends on whether it loads the right brain system at the right point in the behavior. Education works at planning time. Environmental Restructuring works at execution time. Most interventions fail because they ship Education for an execution-time problem.

COM-B vs Other Behavior-Change Frameworks

Three comparisons sharpen what COM-B does that nothing else does as well.

COM-B vs Theory of Planned Behavior

The Theory of Planned Behavior (TPB) says behavior is predicted by intention, and intention is predicted by attitude, subjective norm, and perceived behavioral control. TPB is excellent on intention formation. It is silent on whether intention is the binding constraint. COM-B reframes the question: if Reflective Motivation is not the deficit, building stronger intentions will not help. TPB nests inside the Reflective Motivation slot of COM-B. Use TPB to design interventions once COM-B has diagnosed Reflective Motivation as the bottleneck; use COM-B to decide whether TPB is even relevant.

COM-B vs the Health Belief Model

The Health Belief Model (HBM) targets Reflective Motivation specifically: perceived susceptibility, severity, benefits, barriers, self-efficacy, and cues. HBM is sharper than TPB for risk-communication design but shares the same limit. Both assume Reflective Motivation is the layer to move. COM-B says “fine, but only if you have ruled out Capability and Opportunity first.” Many public-health failures over the last two decades are HBM interventions deployed at problems that COM-B would have routed to Physical Opportunity (cost, access) or Automatic Motivation (cue-driven habit).

COM-B vs the BJ Fogg Behavior Model

The BJ Fogg Behavior Model (B = MAP) says behavior happens when Motivation, Ability, and Prompt converge. The overlap with COM-B is real: Fogg’s Ability and Prompt collapse roughly onto Capability and Opportunity, his Motivation onto Motivation. Fogg’s contribution is the moment-of-behavior emphasis: the three ingredients must align at the same instant. COM-B’s contribution is the diagnostic structure: which of six sub-components, evaluated over a population, is the population-level constraint. Fogg is the right tool to design a single tap-to-action moment; COM-B is the right tool to plan a multi-month behavior-change program at the population level. Same model, different resolutions.

COM-B and the Theoretical Domains Framework

For projects that need finer-grained diagnosis, the Theoretical Domains Framework (TDF, Cane et al 2012) expands COM-B into 14 theoretical domains: Knowledge, Skills, Memory, Beliefs about Capabilities, Beliefs about Consequences, Goals, Reinforcement, Emotion, Behavioral Regulation, Social Influences, Social/Professional Role, Optimism, Environmental Context and Resources, and Intentions. Every TDF domain maps onto a COM-B sub-component. The TDF is what you use when the COM-B diagnosis is correct but you need a more granular interview guide for qualitative discovery. For most product teams, COM-B alone is enough; for clinical-implementation researchers, TDF is the standard.

COM-B in the Real World

The framework looks abstract on paper. It earns its keep when you see what it changes about real interventions.

Public Health

The UK NHS Long Term Plan uses COM-B-derived frameworks to design large-scale behavioral interventions across smoking cessation, alcohol reduction, weight management, and vaccination. The structural payoff is in resource allocation: a smoking-cessation program that diagnoses Automatic Motivation as the dominant deficit will fund nicotine-replacement Enablement and Environmental Restructuring (smoke-free zones), not yet another Education campaign. Public Health England’s evaluation work has consistently found that intervention success rates are highest when the intervention function matches the assessed deficit, and lowest when programs default to Education by reflex.

Implementation Science

The single largest professional application of COM-B is implementation science: getting healthcare workers to actually do evidence-based things, not patients. Hand hygiene compliance is the classic example. For decades, hospitals ran Education-only programs. COM-B-guided audits typically diagnose Physical Opportunity deficits (sinks too far away, dispensers empty) and Social Opportunity deficits (norms among senior physicians) as the dominant constraints. Replace the hand sanitiser at every doorframe (Environmental Restructuring) and you get bigger compliance gains than another lecture series. WHO’s Five Moments for Hand Hygiene rollout used COM-B-derived diagnostics across dozens of countries with measured improvement.

One concrete example worth dwelling on: Hertfordshire County Council in the UK ran a COM-B-diagnosed obesity intervention starting in 2017 across thirteen primary-care sites. The default move would have been Education at scale: brochures, posters, GP talking points. Their COM-B audit instead found that the dominant deficit in the target population was Physical Opportunity (access to affordable produce and safe walking routes) combined with Automatic Motivation (entrenched food cues at home). The funded program shifted budget away from the education campaign and into a voucher scheme for fresh produce at participating supermarkets plus a redesign of school-route signage that surfaced walking options. Twelve-month follow-up showed measurable BMI reductions in the targeted cohort, against essentially zero effect in matched comparator sites that had received the standard Education program. The clinical-public-health literature contains dozens of similar pivots once COM-B is applied honestly: the diagnosis routes resources away from the obvious-but-wrong intervention and toward the structural constraint that nobody wanted to talk about.

Product Design and UX

Product teams rarely cite COM-B but they reinvent it. The classic Duolingo growth playbook is a COM-B program in everything but name: streak notifications and league pressure target Automatic Motivation; lesson scaffolding and difficulty laddering build Psychological Capability; widgets and offline mode handle Physical Opportunity; clubs and friend mechanics build Social Opportunity. The team running it can name every intervention function on the Wheel. Naming it explicitly forces the diagnostic move that ad-hoc product instinct often skips: the team asks “which COM-B component is this onboarding step actually moving” instead of shipping features by genre.

Policy and Behavioral-Insights Units

The UK’s Behavioural Insights Team (BIT, spun out of the Cabinet Office) treats COM-B and the Wheel as one of two foundation frameworks (the other being the EAST model: Easy, Attractive, Social, Timely). Most BIT interventions, from tax-payment letters to organ-donation prompts, are formally diagnosed against COM-B before testing. The discipline that this enforces, “we will not pilot until we have written down the deficit hypothesis,” is what separates BIT’s hit rate from the casual nudge consultancies that proliferated in its wake.

How to Apply COM-B with the Octalysis Framework

COM-B tells you which deficit to address. Octalysis tells you which Core Drive to design with. Together they make a complete planning loop: diagnose, then prescribe at the motivational layer, then choose techniques from the BCT taxonomy. Here is the per-sub-component mapping I use in practice.

Physical Capability → Core Drive 3 (Empowerment of Creativity & Feedback)

Physical Capability is built through short-loop, low-stakes practice with rapid feedback. That is the textbook description of Core Drive 3 (CD3): Empowerment of Creativity & Feedback. Design moves: 90-second guided sessions, single-step demonstrations with immediate rep, error-tolerant first attempts. Train an inhaler patient with a single squeeze-and-hold pattern they can rehearse three times before leaving the clinic, and you build Physical Capability faster than a 30-minute video lecture will. BCT codes that fit this layer include 8.1 Behavioral Practice/Rehearsal, 6.1 Demonstration of the behavior, and 4.1 Instruction on how to perform the behavior.

Psychological Capability → Core Drive 3 plus Core Drive 7 (Unpredictability & Curiosity)

Knowledge transfer is faster when curiosity is recruited, which puts Core Drive 7 (CD7): Unpredictability & Curiosity in play at the front end. Frame the Education as a puzzle, a “spot the difference,” or a small reveal. Then convert the curiosity into mastery via CD3 retrieval practice. BCT 5.1 Information about health consequences and 7.1 Prompts/cues both land here, but only if CD7 has already been used to recruit the attention.

Physical Opportunity → Subtractive Design (the Sludge Inverse)

The Octalysis answer to a Physical Opportunity deficit is not to add a Core Drive; it is to subtract anti-Core-Drives. Walk the desired behavior end-to-end as a new user. Count the steps, the form fields, the wait states, the dead links, the institutional-convenience friction. Remove what does not earn its place. This is the Sunstein-Sludge frame and it is the most underweighted move in Octalysis design conversations. The Sludge pillar is the design audit instrument for this step.

Social Opportunity → Core Drive 5 (Social Influence & Relatedness)

If the deficit is normative, Core Drive 5 (CD5): Social Influence & Relatedness is the design surface. Descriptive norm cues outperform injunctive norm cues when the reference class is tight enough. “Twenty-three people in your building signed up this week” beats “millions worldwide use this app” because the reference class is close enough for the user to map themselves into. BCT codes 6.2 Social comparison, 6.3 Information about others’ approval, and 3.1 Social support (unspecified) all live here.

Automatic Motivation → Core Drives 7 plus 8 (Cue-Driven)

Automatic Motivation is what habit lives on. Core Drive 7 (CD7) recruits attention through unpredictability and curiosity; Core Drive 8 (CD8): Loss & Avoidance creates the negative pull that strong habit cues exert. Design moves are cue-redesign at the moment of behavior: a placement change, a default flip, a friction asymmetry that makes the desired behavior easier than the alternative. BCT 12.5 Adding objects to the environment and 8.2 Behavior substitution map onto this layer cleanly.

Reflective Motivation → Core Drives 1, 2, 4

Reflective Motivation is moved by three Core Drives stacking. Core Drive 1 (CD1): Epic Meaning & Calling frames the stakes so the behavior is positioned as service of something larger than the self (“you are not just exercising, you are showing up for your kids”). Core Drive 2 (CD2): Development & Accomplishment makes progress visible so reflective intention is reinforced by concrete signal. Core Drive 4 (CD4): Ownership & Possession personalizes the goal so it feels pre-owned (“your savings goal” beats “the savings target”). All three together is the right stack for a Reflective Motivation deficit. Pick the dominant one based on the population: caregivers respond strongly to CD1, achievement-oriented users to CD2, asset-framing users to CD4.

A Compact COM-B to Octalysis Lookup

Once a team has internalized the per-component mapping above, a one-line lookup tends to live on the wall of the design room. The form I use:

• Physical Capability deficit: CD3 mastery loops. Bias toward Training and Enablement intervention functions. BCTs 8.1 Behavioral practice/rehearsal, 6.1 Demonstration, 4.1 Instruction.
• Psychological Capability deficit: CD7 curiosity recruitment, then CD3 retrieval practice. Bias toward Education and Training. BCTs 5.1 Information about consequences, 7.1 Prompts/cues, 4.1 Instruction.
• Physical Opportunity deficit: Subtractive design (Sludge inverse). Bias toward Environmental Restructuring, Enablement, and Restriction. BCTs 12.1 Restructuring physical environment, 12.5 Adding objects to the environment, 3.2 Social support (practical).
• Social Opportunity deficit: CD5 Social Influence design. Bias toward Modelling, Social Persuasion, Restriction in some cases. BCTs 6.2 Social comparison, 6.3 Information about others approval, 3.1 Social support unspecified.
• Automatic Motivation deficit: CD7 and CD8 cue design. Bias toward Environmental Restructuring and Persuasion. BCTs 12.5 Adding objects, 8.2 Behavior substitution, 8.3 Habit formation, 6.1 Demonstration.
• Reflective Motivation deficit: CD1 plus CD2 plus CD4 stacking. Bias toward Education, Persuasion, Incentivisation. BCTs 1.1 Goal setting (behavior), 1.4 Action planning, 2.2 Feedback on behavior, 5.6 Information about emotional consequences.

The lookup is not a substitute for thinking; it is a fast pass to narrow the design space. Every chosen Core Drive plus intervention function plus BCT trio should be checked against population-specific evidence before shipping.

The COM-B Octalysis Audit (Six Steps)

1. Specify the behavior at TACT resolution: Target, Action, Context, Time. “Eat better” is not a behavior. “Choose a vegetable side at lunch on weekdays” is.
2. Interview 20-40 target users with COM-B-structured prompts. Ask separately about each sub-component. Treat the population variance as design data, not noise.
3. Diagnose the dominant deficit. Rank the six sub-components by how often they appear as the binding constraint. The top-ranked sub-component is the design target.
4. Pick the intervention function from the Wheel using the empirical mapping (Education for knowledge gaps, Training for skill gaps, Environmental Restructuring for cue/opportunity gaps, and so on). Pick two if the constraints stack.
5. Pick the Core Drive design surface using the mapping above. The Core Drive is what turns the intervention function from “we will educate people” into a specific motivational layer to design against.
6. Pick the BCT codes and ship the smallest pilot that gives you re-measurement of the targeted sub-component. Do not scale until the deficit moved on the population segment you ran on.

Programs that complete all six steps outperform programs that skip the diagnosis or the Core Drive layer. The shortcut version (intervention-by-instinct without COM-B diagnosis, or COM-B without an Octalysis motivational layer) underperforms the full audit by a factor large enough to matter inside a real budget cycle.

Practical Steps

If you are running a behavior-change project this quarter, here is the order of moves I would put first.

1. Write the behavior in TACT form. Target, Action, Context, Time. If you cannot finish the sentence on one line, the behavior is not specified tightly enough for COM-B to give a clean answer.
2. Run a six-question COM-B interview on a sample of the target population. One question per sub-component. Look for which sub-components people stumble on, not what they say they need.
3. Rank the deficits. Pick the top one as the primary; pick the second one as a stacked secondary if it shows up in the same population segment.
4. Map to intervention function using the Wheel. If two functions show up, stack them (Education plus Environmental Restructuring is a common pair).
5. Map to Core Drive using the table above. This is where the design moves get specified at the motivational layer.
6. Pick three BCTs from the v1 taxonomy that operationalize the chosen intervention function. Use the published CALO-RE taxonomy (Coventry, Aberdeen, and London — Refined; Michie et al, 2011) as the lookup.
7. Pilot at the smallest scale that gives you re-measurement of the targeted sub-component. Do not change two variables at once. Confirm the deficit moved before you scale.

The whole loop takes a week or two. It is much cheaper than the alternative, which is shipping six months of intervention and discovering at the post-mortem that the binding constraint was something else entirely.

The Elephant in the Room

Here is the conversation behavior-change teams avoid: most COM-B failures are not framework failures. They are intervention-first failures dressed up in COM-B vocabulary. A team picks a clever intervention they already wanted to ship, then writes a post-hoc COM-B narrative that justifies it. The framework gets blamed when the pilot underperforms. The framework was never used.

The fix is not to run more COM-B trainings. It is to make the diagnostic artifact the deliverable for the first week of any project. Before any feature gets specced, before any campaign gets briefed, the team writes a one-page COM-B diagnosis: which sub-components are deficient, in which population segment, with what evidence. If that document does not exist by the end of week one, the project has not started yet, regardless of how busy everyone is. The discipline is structural, not motivational.

A second elephant is that COM-B’s neat separation of components hides the fact that interventions usually stack. A serious smoking cessation program runs Education plus Training plus Enablement plus Environmental Restructuring at once. Teams that treat the Wheel as a “pick one” instrument under-deliver. Read it as a deficit-routing map: each diagnosed deficit lights up a candidate intervention class, and a real program is usually the sum of two or three lit-up intervention classes running concurrently.

COM-B Was the Beginning, Not the End

The Behavior Change Wheel is the closest thing the behavioral-science field has to a planning instrument that survives contact with real budgets. It is not the most exciting framework on the shelf. It does not have the elegance of Self-Determination Theory or the sticky vocabulary of the Hook Model. But the discipline it forces, “diagnose before you prescribe, and prescribe at the right layer,” is what separates programs that move outcomes from programs that win design awards and accomplish nothing.

The integration with Octalysis is where I think most product designers will get the biggest leverage. COM-B gives you the deficit. Octalysis gives you the motivational design surface. BCT v1 gives you the technique. None of the three alone is enough. All three together is the closest thing I have found to a behavior-change planning toolkit that a working team can actually use on Monday.

If you build one behavior-change deliverable this quarter, make it a one-page COM-B diagnosis for the behavior your team is supposed to move. Everything else is downstream of that document existing.

Frequently Asked Questions

What does COM-B stand for, in simple terms?

Capability, Opportunity, Motivation, Behavior. A person performs a behavior when all three ingredients are present. Capability is whether they can do it (physically and mentally). Opportunity is whether the environment lets them do it (physically and socially). Motivation is whether they want to do it more than competing behaviors (reflectively and automatically).

How is COM-B different from the BJ Fogg Behavior Model (B = MAP)?

Both decompose behavior into three ingredients with substantial overlap. Fogg’s model is sharper for designing a single moment of action (B = MAP says Motivation, Ability, and Prompt must converge at one instant). COM-B is sharper for planning a multi-week or multi-month behavior-change program at the population level. Fogg’s Ability roughly equals COM-B’s Capability; his Prompt sits between Opportunity and Automatic Motivation; his Motivation maps onto both Reflective and Automatic Motivation in COM-B. Different resolutions, complementary uses.

When should I use TDF instead of COM-B?

Use the Theoretical Domains Framework (14 domains) when you need a finer-grained interview guide for qualitative discovery, especially in implementation-science work where you are studying clinician or organizational behavior. Use COM-B (6 sub-components) when you need a faster diagnostic and are willing to trade resolution for speed. TDF nests inside COM-B; every TDF domain maps onto one of the six COM-B sub-components.

Why do most behavior-change interventions fail?

The dominant failure mode is intervention-before-diagnosis. Teams pick the intervention they wanted to ship anyway, then write a COM-B story that justifies it. Real COM-B work runs the diagnosis to completion first, then picks the intervention function from the Wheel based on the deficit. Programs that skip the diagnostic step typically under-deliver by 30 to 60 percent against properly diagnosed programs of equal cost.

How does COM-B relate to the Octalysis Framework?

COM-B is a diagnostic layer; Octalysis is a motivational design layer; BCT v1 is a technique catalog. The clean pipeline: diagnose the deficit with COM-B, then pick the Octalysis Core Drive that fits the deficit (CD3 for Capability, CD5 for Social Opportunity, CD1/CD2/CD4 for Reflective Motivation, CD7/CD8 for Automatic Motivation, subtractive design for Physical Opportunity), then pick the BCTs that operationalize the chosen Core Drive’s intervention function.

What are the Behavior Change Techniques (BCTs) in v1?

The Behavior Change Technique Taxonomy v1 (Michie et al, 2013) is a catalog of 93 named, reliably labeled techniques: Goal Setting, Action Planning, Feedback on Behavior, Self-Monitoring, Habit Formation, Social Support, and so on. Each BCT is mapped to one or more intervention functions on the Wheel, so once the deficit is diagnosed and the intervention class chosen, the team has a short list of specific moves to ship.

Is COM-B appropriate for product/app design or only public health?

Both. COM-B was developed for public-health and clinical-implementation contexts, but the structure is content-agnostic. Product teams using COM-B explicitly tend to outperform teams using ad-hoc behavioral instincts because the diagnostic step is harder to skip when the framework is written down. Duolingo’s growth playbook, while not explicitly called COM-B, looks structurally identical when audited against the framework.

What is the biggest mistake when using COM-B for the first time?

Two competing first-time mistakes are equally common. First, treating the framework as a tick-box exercise where you write a paragraph for each sub-component without doing real user research. Second, treating it as a “pick one deficit” instrument when the right read is that the deficit is multi-component and the intervention has to stack. The way to avoid both is to run actual interviews on at least 20 target users and let the data tell you which sub-components are dominant.

How do COM-B and self-efficacy interact?

Bandura’s self-efficacy concept lives in both Reflective Motivation (beliefs about one’s ability to perform the behavior) and Psychological Capability (the underlying knowledge and regulation skills). This is one of the places the framework’s clean separation gets tangled. The practical move is to design CD3 mastery loops that build genuine Capability at low enough stakes that early failures do not erode self-efficacy, then let the resulting Reflective Motivation gains take care of themselves.

References

1. Michie, S., van Stralen, M. M., & West, R. (2011). The behaviour change wheel: A new method for characterising and designing behaviour change interventions. Implementation Science, 6:42.
2. Michie, S., Atkins, L., & West, R. (2014). The Behaviour Change Wheel: A Guide to Designing Interventions. Silverback Publishing.
3. Michie, S., Richardson, M., Johnston, M., Abraham, C., Francis, J., Hardeman, W., Eccles, M. P., Cane, J., & Wood, C. E. (2013). The behavior change technique taxonomy (v1) of 93 hierarchically clustered techniques: building an international consensus for the reporting of behavior change interventions. Annals of Behavioral Medicine, 46(1), 81-95.
4. Cane, J., O’Connor, D., & Michie, S. (2012). Validation of the theoretical domains framework for use in behaviour change and implementation research. Implementation Science, 7:37.
5. Atkins, L., Francis, J., Islam, R., O’Connor, D., Patey, A., Ivers, N., Foy, R., Duncan, E. M., Colquhoun, H., Grimshaw, J. M., Lawton, R., & Michie, S. (2017). A guide to using the Theoretical Domains Framework of behaviour change to investigate implementation problems. Implementation Science, 12:77.
6. West, R., & Michie, S. (2020). A brief introduction to the COM-B Model of behaviour and the PRIME Theory of motivation. Qeios, WW04E6.
7. Michie, S., Ashford, S., Sniehotta, F. F., Dombrowski, S. U., Bishop, A., & French, D. P. (2011). A refined taxonomy of behaviour change techniques to help people change their physical activity and healthy eating behaviours: the CALO-RE taxonomy. Psychology & Health, 26(11), 1479-1498.
8. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191-215.
9. Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), 179-211.
10. Rosenstock, I. M. (1974). Historical origins of the health belief model. Health Education Monographs, 2(4), 328-335.
11. Fogg, B. J. (2009). A behavior model for persuasive design. Proceedings of the 4th International Conference on Persuasive Technology, 40.
12. Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving Decisions About Health, Wealth, and Happiness. Yale University Press.
13. Sunstein, C. R. (2022). Sludge: What Stops Us from Getting Things Done and What to Do about It. MIT Press.
14. Behavioural Insights Team. (2014). EAST: Four Simple Ways to Apply Behavioural Insights. The Behavioural Insights Team / UK Cabinet Office.
15. Wood, W., & Neal, D. T. (2007). A new look at habits and the habit-goal interface. Psychological Review, 114(4), 843-863.
16. Chou, Y. (2015). Actionable Gamification: Beyond Points, Badges, and Leaderboards. Octalysis Media.

Related Reading

• The Octalysis Framework: the 8-Core-Drive motivational map this pillar prescribes against.
• The Behavioral Framework Library: every other behavioral framework explained as an S-Tier Designer’s Guide.
• Theory of Planned Behavior: the Reflective Motivation sub-component, in much more detail.
• Health Belief Model: the risk-communication design instrument for Reflective Motivation deficits.
• BJ Fogg Behavior Model: the moment-of-action complement to COM-B’s population-level diagnosis.
• Sludge: Subtractive Design, the audit instrument for Physical Opportunity deficits.

WOULD YOU LIKE YU-KAI CHOU TO WORK WITH YOUR ORGANIZATION?