← The Policy Reform Toolkit

20 October 2026

Motivation Isn’t Enough: The Three Phases of Behaviour Change

Behaviour Change

Lots of policies involve behaviour change. This research paper synthesised the evidence into a three-phase process.

The 3 phases of behaviour change. Phase 1, Motivation (wanting to change): believing the change is worth it and that you can do it. Sounds like: "I know I should do x. But is it really worth it, and could I even stick with it?" If it seems not worth the effort: incentives, information and education, authoritative endorsements, social norms, social accountability. If people doubt it's possible: growth mindset, subgoals, role models, advice giving, fresh starts, values affirmation, fostering social belonging. Phase 2, Follow-through (turning intention into action): actually acting on good intentions, when it counts. Sounds like: "I meant to do x. Then I forgot, it felt like a hassle, and I did something else." Forgetfulness: reminders, goal setting, planning, if-then plans, checklists, feedback. Laziness: defaults, convenience. Impatience: commitment devices, making it fun. Phase 3, Habit (making it stick): the behaviour becomes automatic and survives disruption. Sounds like: "I did x every week for a month. Then the holidays hit and the routine fell apart." Need for consistency: rewarding repetition. Need for flexibility: offering flexibility. Each phase is also linked to COM-B (an added interpretation, not from the paper). Source: Voelkel, Milkman and Duckworth (2027), "Behavior Change", Annual Review of Psychology 78, Table 1.

Key takeaways

  • Behaviour change has three phases, each with its own problems. The authors frame change as a process: first building motivation, then following through on intentions, and finally (for repeated behaviours) forming habits.
  • Motivation needs two things: enough perceived value, and belief that success is possible.
  • Incentives work, but how they’re framed matters.
  • Social levers often beat plain information. Information campaigns usually have modest effects. Endorsements from the right sources, social norms, and accountability work better.
  • Make change feel achievable. Growth mindset messages, smaller subgoals, relatable role models, values affirmation, and belonging interventions all boost self-efficacy.
  • Intentions don’t reliably become actions. Shifting intentions by 0.66 standard deviations only shifts behaviour by 0.36. The three main follow-through barriers are forgetfulness, laziness, and impatience.
  • Beat forgetfulness with reminders, goals, and if–then plans.
  • Beat laziness by making the good option the easy one. Defaults are among the strongest tools.
  • Beat impatience with commitment devices and by making it fun.
  • Habits need repetition and flexibility, and the evidence base here is thinnest.
  • Behaviour change is hard and highly context-dependent. The authors call for comparing interventions head-to-head, more work spanning all three phases, and more research on habits and on when strategies work for whom.

Comment

COM-B (Capability, Opportunity, Motivation) is a common framework used for behaviour change. The two overlap. I wonder if the two frameworks should be combined to deliver the best outcomes.

What next?

Are you taking into account these findings when designing policies involving behaviour change?

Worth a read: Useful links. Free guides, research and tools for policy reformers.

What's new: Toolkit update. New features and content in the Policy Reform Toolkit.

  • New: The toolkit is now live. If you have not yet done so, you can sign up here.
  • New: All the weekly newsletters will be put onto a public page here.

Tell me what you think

Email me at russell@policyreformtoolkit.com with your thoughts, questions or requests. I read every message, and your feedback shapes what the Toolkit does next.

Russell

← All issues