Intervention Design Worksheet

Converts diagnosis into designed action. One per target behaviour.

The rule this enforces: no intervention without a named barrier. Anything that can’t name the barrier it addresses gets cut. This single discipline removes most of the waste from a change plan.

1 · The behaviour

Fill in

B-number: ___ Who: _______________ Does what, when: _______________ Instead of: _______________ Current frequency: _______________ Target frequency: _______________

2 · The barrier — from evidence

From Barrier Analysis — COM-B Interview Guide.

COM-B component	Score 1–5	Evidence
Physical Capability	`_`
Psychological Capability	`_`
Physical Opportunity	`_`
Social Opportunity	`_`
Reflective Motivation	`_`
Automatic Motivation	`_`

Primary barrier: _______________ Secondary barrier: _______________

3 · Which intervention functions can work

Use the mapping from COM-B and the Behaviour Change Wheel:

Physical Capability      → Training, Enablement
Psychological Capability → Education, Training, Enablement
Physical Opportunity     → Restriction, Env. restructuring, Enablement
Social Opportunity       → Restriction, Env. restructuring, Modelling, Enablement
Reflective Motivation    → Education, Persuasion, Incentivisation, Coercion
Automatic Motivation     → Persuasion, Incentivisation, Coercion,
                           Env. restructuring, Modelling, Enablement

Candidate functions for my primary barrier: _______________

4 · Generate options

Take each candidate function and generate 2–3 concrete tactics. Raid Nudge Library and run the sprint in EAST — Easy Attractive Social Timely.

#	Function	Concrete tactic	Effort
1
2
3
4
5
6

Force at least one “change the path” option

Every list should contain at least one option that changes the environment rather than the person — a default, a removed field, a switched-off legacy route, an agenda change. These are usually the cheapest and most durable, and they’re the ones a group of people brainstorming will systematically under-produce.

5 · APEASE filter

Score 1–5. See COM-B and the Behaviour Change Wheel.

#	Accept.	Practic.	Effect.	Afford.	Side-eff.	Equity	Total
1
2
3
4
5
6

Selected: ___________ Rejected, and why (write this down — you will be asked): ___________

6 · The intervention spec

Spec

What we will do: _______________ Who does it: _______________ To whom: _______________ Starting: _______________ How long: _______________ Cost / effort: _______________ What we expect to see, by when: _______________ How we’ll know it worked: _______________ What we’ll do if it doesn’t: _______________

That last line matters. Interventions that fail quietly consume budget indefinitely. Set a kill criterion now, while it’s not personal.

7 · Side-effects check

What might get worse? ___________
Who might this disadvantage — a shift, a site, part-time staff, a role? ___________
What might people do to game it? ___________
Does it conflict with an existing target or incentive? ___________

⚠️ The gaming question is not cynical, it’s necessary. If you measure a behaviour, people will optimise for the measure. Ask “what’s the laziest way to score well on this without doing the real thing?” — and design that out before launch.

The stack

Single interventions rarely shift a behaviour. Aim for a stack of three to four across different components:

Layer	Example	Purpose
Make it easy	Cut form from 12 fields to 4	Removes Physical Opportunity barrier
Make it normal	Weekly team-level adoption, positive framing	Social Opportunity
Make it cued	New behaviour attached to end-of-call	Automatic Motivation
Make it rewarded	System returns something useful within 24h	Reflective + Automatic Motivation
Make it required	Monday meeting runs off new system only	Restriction / Env. restructuring

Log the result in T — Intervention Note, then test it with Pilot Design and Experiment Card.

← Back to the Behavioural Change Toolkit

Last updated: 11 August 2026