Template

Behavioural Scientist — Competency Roadmap

Work towards being a behavioural scientist: understanding how people actually decide and act, and being able to design, run and evaluate interventions that change behaviour in the real world.

This roadmap provides an end-to-end curriculum for aspiring behavioural scientists across public policy nudge units, tech product teams, and research labs. You will build systematic grounding in cognitive psychology and behavioural economics, master experimental evaluation via randomized controlled trials and A/B testing, and apply structured intervention design frameworks like COM-B and EAST. Sized for a consistent 10 hours per week, this track moves progressively from theoretical mechanisms to end-to-end intervention briefs, registered trial protocols, and analysis scripts. You will finish with three complete portfolio-grade behavioural case studies and the technical fluency to diagnose behavioural friction, design targeted choice architecture, and evaluate causal impact.

By the end: You will be able to diagnose real-world behavioural bottlenecks using structured frameworks (COM-B, EAST), design contextual interventions, author pre-registered experimental protocols (RCTs/A-B tests), and analyze intervention data in R to measure causal effect sizes.

Starting levelBeginnerStyleA mix
10h / week11 phases25 tasks~154h total

This is the map — make this roadmap yours

It shows what this journey generally looks like. Tell Kaidoro your version of the goal and it builds the plan around where you are actually starting, what to do first, the hours you really have, and what you have already finished.

1

Foundations of Cognitive Psychology & Decision Science

Establish the foundational cognitive models of human judgment and decision-making under uncertainty, focusing on heuristics, cognitive load, and dual-process theories.

  • Read core literature on heuristics, biases, and dual-process models
    ~6hLearn

    Before attempting to change behaviour, you must understand the mental shortcuts and cognitive constraints that govern human decision-making.

    You'll learn

    • Bounded rationality — decision-making within cognitive and environmental limits
    • Dual-process theory — System 1 fast automatic thinking versus System 2 deliberate effortful thought
    • Availability heuristic — assessing probability based on how easily examples come to mind
    • Anchoring and adjustment — disproportionate reliance on the first piece of information offered

    Examine foundational papers on System 1 and System 2 cognitive processing, bounded rationality, and common heuristics including availability, representativeness, and anchoring. Take detailed notes linking theoretical mechanisms to everyday judgment errors.

    Done when: you have written a 3-page summary defining 12 core heuristics with everyday examples and identifying the cognitive limitations of dual-process theory.

    How to work through it

    1. Read Kahneman and Tversky's foundational 1974 paper on judgment under uncertainty
    2. Review contemporary critiques and refinements of dual-process cognitive theory
    3. Extract 12 primary heuristics and their underlying cognitive mechanisms into a structured matrix
    4. Write the comparative mechanism summary document
  • Conduct a heuristic analysis of an existing digital or public service
    ~5hApply

    Translating theoretical cognitive biases into observable friction points in real-world environments is the primary diagnostic skill of a behavioural scientist.

    You'll learn

    • Cognitive load theory — the working memory demand imposed by interface and message complexity
    • Salience bias — the human tendency to focus on items that are prominent and ignore plain details
    • Default effect — the disproportionate tendency to stick with pre-selected options

    Select an existing public application form or digital service flow and systematically evaluate how cognitive biases and high cognitive load create friction or lead users into systematic errors.

    Done when: you have produced a slide deck annotating 5 user screens or steps with identified cognitive biases, estimated cognitive load, and potential missteps.

    How to work through it

    1. Select an existing user journey such as applying for social benefits or registering an account
    2. Document each screen, form field, and prompt with screenshots
    3. Map user cognitive load at each step against known cognitive biases
    4. Annotate screenshots with specific judgment traps and compile into a slide deck
2

Behavioural Economics & Choice Under Uncertainty

Deepen your understanding of formal behavioural economics, contrasting neoclassical expected utility with prospect theory, intertemporal choice, and social preferences.

  • Work through core models of Prospect Theory and Loss Aversion
    ~7hLearn

    Prospect theory is the cornerstone of behavioural economics and explains risk attitudes across gains and losses.

    You'll learn

    • Prospect theory — behavioral model describing decisions between alternatives involving risk
    • Loss aversion — psychological pain of losing being roughly twice as impactful as equivalent gains
    • Diminishing sensitivity — marginal value of gains or losses decreasing as distance from reference point grows
    • Probability weighting — over-weighting small probabilities and under-weighting moderate-to-high probabilities

    Examine how value functions, loss aversion, probability weighting, and framing effects diverge from standard expected utility theory in financial, health, and policy decisions.

    Done when: you have solved 5 calculation sets showing the mathematical divergence between Expected Utility Theory and Prospect Theory outcomes.

    How to work through it

    1. Study the S-shaped value function and probability weighting functions in Kahneman and Tversky (1979)
    2. Compare expected utility calculations with prospect theory decision weights across lottery pairs
    3. Document how framing changes risk preferences from risk-averse to risk-seeking
    4. Compile and verify worked mathematical examples
  • Analyze time discounting and present bias in decision scenarios
    ~6hPractice

    Present bias drives some of the most critical societal challenges across health, personal finance, and retirement planning.

    You'll learn

    • Present bias — tendency to overvalue immediate rewards relative to future consequences
    • Quasi-hyperbolic discounting (beta-delta) — mathematical formalization of immediate preference vs constant long-run discounting
    • Commitment devices — arrangements that incentivize a self-control goal by imposing costs on failure

    Study hyperbolic and quasi-hyperbolic (beta-delta) discounting models. Explore how present bias explains procrastination, under-saving, and non-adherence in preventative healthcare.

    Done when: you have completed a memo modelling how a naive vs. sophisticated present-biased agent behaves across 3 savings and health commitment scenarios.

    How to work through it

    1. Compare exponential discounting with hyperbolic and quasi-hyperbolic discounting formulas
    2. Define the difference between naive and sophisticated agents in self-control problems
    3. Model behaviour across three scenarios: credit card debt payoff, pension enrolment, and gym attendance
    4. Write the analytical memo highlighting commitment devices
  • Write a comparative breakdown of social norms and peer effects
    ~5hApply

    Social norms are powerful intervention levers but carry known backfire risks if poorly calibrated.

    You'll learn

    • Descriptive norms — perceptions of which behaviours are typically performed
    • Injunctive norms — perceptions of which behaviours are typically approved or disapproved of
    • Boomerang effect — unintended consequence where below-average consumers increase undesirable usage after norm exposure

    Investigate descriptive versus injunctive norms, reciprocity, and public goods experiments. Detail how social proof can backfire if descriptive norms highlight undesirable baseline behaviours.

    Done when: you have completed a 1,500-word critique analyzing two real-world campaigns that used social norms (one successful, one backfired).

    How to work through it

    1. Differentiate descriptive norms (what is done) from injunctive norms (what ought to be done)
    2. Review classic experiments on energy conservation (e.g., Opower) and boomerang effects
    3. Select one successful social norm campaign and one that backfired
    4. Author the comparative critique explaining the underlying behavioural mechanisms
3

Behavioural Diagnosis & Problem Scoping

Learn to translate ambiguous business or policy goals into precise target behaviours and map barriers using frameworks like COM-B, the Behaviour Change Wheel, and EAST.

  • Define target behaviours using the 'Who, What, When, Where' standard
    ~4hPractice

    Behavioural interventions fail most often at the problem-definition stage when trying to change an outcome rather than a specific action.

    You'll learn

    • Target behaviour specification — defining the exact action, person, frequency, and context
    • Impact-feasibility matrix — prioritising behavioural targets based on potential effect size vs ease of adoption

    Practice converting vague outcome goals (e.g., 'reduce electricity consumption') into observable, measurable, discrete target behaviours.

    Done when: you have decomposed 3 broad organizational goals into a table of 6 specific, actor-centric target behaviours.

    How to work through it

    1. Select three broad outcome goals across health, sustainability, and digital engagement
    2. Identify the actor, the precise action, the contextual setting, and the exact timing required
    3. Filter behaviours by impact versus feasibility
    4. Document the target behaviour definitions
  • Build a comprehensive COM-B diagnostic map for a user journey
    ~7hBuild

    COM-B provides an exhaustive taxonomy to prevent leaping to solutions before identifying whether the barrier is knowledge, environment, or motivation.

    You'll learn

    • COM-B model — framework positing that Capability, Opportunity, and Motivation interact to produce Behaviour
    • Theoretical Domains Framework (TDF) — an integrative framework of 14 domains synthesizing behavior change theories
    • Sludge — cognitive and administrative frictions that make a process difficult without adding value

    Apply Michie's COM-B (Capability, Opportunity, Motivation -> Behaviour) model and the Theoretical Domains Framework (TDF) to diagnose behavioural barriers along a complex user flow.

    Done when: you have generated a complete COM-B diagnostic matrix identifying at least 6 distinct barriers across physical/psychological capability, physical/social opportunity, and reflective/automatic motivation.

    How to work through it

    1. Select a target behaviour (e.g., organ donation sign-up or completing continuous glucose monitoring logs)
    2. Map the current step-by-step user journey
    3. Categorize observed drop-off points against the 6 COM-B subcomponents
    4. Map identified barriers to the Theoretical Domains Framework to pinpoint psychological drivers
    5. Finalize the diagnostic matrix
4

Inferential Statistics & Power Analysis for Experiments

Master the statistical fundamentals necessary to design robust trials, run sample size calculations, and avoid false discovery in behavioural research. This phase can be studied concurrently with qualitative methods.

  • Set up an R environment for behavioural data analysis
    ~3hBuild

    R is the industry standard for experimental analysis in behavioural science research and public policy units.

    You'll learn

    • RStudio — integrated development environment for R
    • tidyverse — collection of R packages designed for data science and manipulation
    • pwr package — tool in R for computing statistical power and sample size requirements

    Install R and RStudio (or VS Code with R tools) and configure packages used in experimental analysis including tidyverse, broom, and pwr.

    Done when: you have executed an R script that imports a sample dataset, runs basic summary statistics, and outputs a clean exploratory plot.

    How to work through it

    1. Install R and RStudio Desktop
    2. Install core packages: tidyverse, pwr, broom, effsize, estimatr
    3. Write and run a starter script loading data, computing group means, and rendering a ggplot chart
    4. Verify all package environments compile cleanly
  • Calculate statistical power and minimum detectable effect sizes
    ~6hPractice

    Underpowered experiments waste resources and produce unreliable, unreplicable results.

    You'll learn

    • Statistical power — the probability of correctly rejecting a false null hypothesis
    • Minimum Detectable Effect Size (MDES) — smallest true effect an experiment has adequate power to detect
    • Cohen's d — standardized measure of difference between two means

    Master Type I and Type II errors, alpha levels, power (1 - beta), baseline conversion rates, and Minimum Detectable Effect Size (MDES) calculations for two-sample proportions and means.

    Done when: you have written an R script that calculates required sample sizes across 4 varying effect size (Cohen's d = 0.05 to 0.3) and power scenarios (80% vs 90%).

    How to work through it

    1. Review statistical power theory and trade-offs between sample size, effect size, alpha, and power
    2. Use the `pwr.2p.test` and `pwr.t.test` functions in R across varying baselines
    3. Generate power curves plotting sample size against detectable effect sizes
    4. Save the script and power curve charts
  • Run hypothesis tests and regression models on experimental data
    ~8hApply

    Regression-based experimental analysis allows you to control for baseline covariates and obtain more precise treatment effect estimates.

    You'll learn

    • Average Treatment Effect (ATE) — the average difference in outcome between treatment and control units
    • Robust standard errors — standard error calculations adjusted for heteroscedasticity
    • Covariate adjustment — inclusion of baseline variables to reduce error variance and increase precision

    Perform t-tests, chi-square tests, and Ordinary Least Squares (OLS) regressions with robust standard errors on simulated experimental trial datasets with covariates.

    Done when: you have authored an R Markdown report analyzing an A/B test dataset, reporting p-values, confidence intervals, and Average Treatment Effects (ATE).

    How to work through it

    1. Generate or load a mock trial dataset containing treatment assignment, covariates, and outcome variables
    2. Run two-sample t-tests and two-proportion tests
    3. Fit linear models with and without covariate adjustment using `lm_robust` from estimatr
    4. Compile the annotated R Markdown document into a rendered HTML report
5

Experimental Design & Field Trials (RCTs & A/B Testing)

Learn to design randomized controlled trials in the field and A/B tests in digital products, protecting against threats to internal and external validity.

  • Map unit of randomization and experimental variants
    ~6hLearn

    Selecting the wrong level of randomization can lead to experimental contamination that invalidates trial results.

    You'll learn

    • SUTVA — assumption that one unit's treatment assignment does not affect another unit's outcome
    • Cluster randomization — randomizing groups (classrooms, clinics, cities) rather than isolated individuals
    • Factorial design — testing two or more independent variables concurrently to detect interaction effects

    Study individual vs cluster randomization, factorial designs, and multi-arm trials. Learn how to identify and avoid spillovers, contamination, and interference between treatment and control.

    Done when: you have written a 2-page decision matrix comparing individual vs cluster randomization for 3 hypothetical interventions (school, workplace, digital app).

    How to work through it

    1. Examine Stable Unit Treatment Value Assumption (SUTVA) and spillover mechanisms
    2. Compare individual vs cluster-randomized trial requirements and design effects
    3. Evaluate 2x2 factorial designs to test multiple intervention components simultaneously
    4. Complete the decision matrix evaluating clustering needs
  • Audit and control for threats to internal and external validity
    ~5hPractice

    Field experiments are messy; anticipating data leakage and behavioral artifacts protects research integrity.

    You'll learn

    • Attrition bias — systematic difference in drop-out rates between treatment and control groups
    • Intention-to-Treat (ITT) — analyzing subjects based on assigned group regardless of actual protocol compliance
    • Hawthorne effect — alteration of behaviour by subjects due to awareness of being observed

    Investigate common trial failure modes: attrition bias, novelty effects, Hawthorne effects, demand characteristics, and p-hacking.

    Done when: you have written a risk mitigation checklist identifying 8 validity threats in field trials and the pre-planned methodological checks for each.

    How to work through it

    1. Define internal validity threats (attrition, history effects, instrumentation)
    2. Define external validity limitations (WEIRD sample bias, site-selection bias)
    3. Detail procedural safeguards: blind allocations, intention-to-treat analysis, and balance checks
    4. Synthesize into an actionable trial risk audit checklist
  • Draft a formal Pre-Analysis Plan (PAP) and registered trial protocol
    ~8hBuild

    Pre-registration is the gold standard for causal research, establishing credibility and preventing post-hoc hypothesis generation (HARKing).

    You'll learn

    • Pre-Analysis Plan (PAP) — formal document specifying all hypotheses and analytic procedures prior to data collection
    • HARKing — Hypothesizing After Results are Known, a key cause of false-positive literature
    • Primary vs secondary outcomes — pre-specifying key metrics to control family-wise error rates

    Create a complete, submission-ready pre-analysis plan for a proposed field experiment including primary hypotheses, power calculation, exclusion criteria, and statistical models.

    Done when: you have produced a complete 6-to-8 page Pre-Analysis Plan adhering to OSF / AEA Registry standards for an original intervention.

    How to work through it

    1. Select an intervention topic (e.g., improving tax compliance or increasing digital feature adoption)
    2. Define primary and secondary outcome metrics and exact measurement timeframes
    3. Specify sample size, power analysis, randomization procedure, and balance check tests
    4. Draft the complete pre-analysis plan and export to PDF
6

Qualitative & Mixed-Methods Behavioural Research

Master qualitative research techniques to discover underlying mental models, user contexts, and emotional drivers that quantitative data alone cannot reveal. This phase can proceed concurrently with quantitative phases.

  • Conduct semi-structured behavioural discovery interviews
    ~6hPractice

    People cannot reliably report their cognitive biases; interviewers must reconstruct actual past behaviour rather than asking for hypotheticals.

    You'll learn

    • Critical Incident Technique — qualitative interview method focusing on specific past actions rather than abstractions
    • Social desirability bias — tendency of interviewees to answer questions in a manner that will be viewed favorably by others
    • Discussion guide — structured outline of topics and open probes used to direct qualitative interviews

    Learn to write interview discussion guides focused on specific past episodes rather than abstract opinions, avoiding confirmation and social desirability biases.

    Done when: you have written an interview guide and conducted, recorded, and transcribed two 30-minute behavioural discovery interviews.

    How to work through it

    1. Draft a semi-structured interview protocol with open-ended, episode-specific prompts
    2. Avoid leading questions and abstract questions ('Why did you...?' vs 'Walk me through what happened next...')
    3. Conduct two 30-minute interviews with volunteer subjects on a specific decision journey
    4. Transcribe recordings and extract initial behavioral observations
  • Run a thematic analysis and synthesize behavioral personas
    ~6hBuild

    Behavioural personas focus on decision contexts and heuristics rather than demographic characteristics.

    You'll learn

    • Thematic analysis — qualitative method for identifying, analyzing, and reporting patterns (themes) within data
    • Behavioural personas — user archetypes defined by decision-making habits and mental models rather than demographics
    • Mental models — user internal explanations of how something works in the real world

    Code qualitative interview transcripts to extract recurring behavioural archetypes, frictions, and contextual triggers.

    Done when: you have completed a coded qualitative synthesis board and generated 2 behavioural personas highlighting mental models, friction points, and decision contexts.

    How to work through it

    1. Code transcript excerpts into thematic buckets using inductive coding
    2. Group codes into recurring friction patterns, context drivers, and baseline mental models
    3. Construct two behavioural personas articulating triggers, decision friction, and psychological barriers
    4. Publish the qualitative synthesis artifact
7

Choice Architecture & Intervention Design

Synthesize diagnostic findings into actionable interventions using structured frameworks like EAST, MINDSPACE, and choice architecture techniques.

  • Apply the EAST and MINDSPACE frameworks to ideate solutions
    ~5hPractice

    Structured ideation frameworks prevent teams from defaulting solely to information and awareness campaigns.

    You'll learn

    • EAST framework — core applied model: Make it Easy, Attractive, Social, and Timely
    • MINDSPACE — 9-factor checklist of behavioral influences (Messenger, Incentives, Norms, Defaults, Salience, Priming, Affect, Commitments, Ego)
    • Information deficit model — the mistaken assumption that providing more information alone will change behaviour

    Use the UK Behavioural Insights Team's EAST framework (Easy, Attractive, Social, Timely) and MINDSPACE mnemonic to systematically generate intervention candidates.

    Done when: you have generated an intervention matrix containing at least 16 candidate interventions mapped across EAST dimensions for a target problem.

    How to work through it

    1. Review each component of the EAST and MINDSPACE taxonomies
    2. Take the COM-B diagnostic map from Phase 3 and ideate 4 interventions per EAST pillar
    3. Filter solutions by policy/engineering feasibility and expected behavioral friction reduction
    4. Document the 16 intervention concepts in a comparative ideation matrix
  • Design a complete multi-arm intervention prototype
    ~7hBuild

    Behavioural scientists must produce concrete design artifacts and copy, not just theoretical recommendations.

    You'll learn

    • Choice architecture — the deliberate design of the environment in which people make decisions
    • Opt-in vs opt-out defaults — setting the desired path as the baseline without restricting free choice
    • Implementation intentions — if-then planning prompts connecting specific situational cues to actions

    Build production-ready prototypes for two intervention arms and one control condition (e.g., redesigned notices, modified digital user flows, or SMS reminder schedules).

    Done when: you have developed visual or textual assets for Control, Arm 1 (Friction Reduction), and Arm 2 (Social Norm + Timely Prompt) with an annotated design rationale.

    How to work through it

    1. Select the two highest-scoring intervention concepts from your ideation matrix
    2. Design the Control variant (status quo baseline)
    3. Draft Arm 1 focusing on default settings, friction removal, and simplified copy
    4. Draft Arm 2 combining social proof framing with targeted timing triggers
    5. Compile visual mockups or exact text assets with an accompanying theoretical rationale document
8

Applied Behavioural Science in Public Policy & Social Impact

Explore the structural applications of behavioural insights across government, civic tech, health policy, and environmental sustainability.

  • Analyze landmark public policy nudge unit case studies
    ~6hLearn

    Understanding how national nudge units navigate legislative constraints and cost-benefit assessments is vital for policy roles.

    You'll learn

    • Policy trial scaling — challenges in maintaining effect sizes when expanding from pilot to national rollout
    • Cost-effectiveness ratio — ratio of intervention cost to units of behavioural outcome changed
    • Heterogeneous treatment effects — variations in intervention impact across different demographic subgroups

    Study canonical policy trials run by BIT (UK Behavioural Insights Team), OES (US Office of Evaluation Sciences), and the World Bank eMBeD team across tax compliance, court appearances, and organ donation.

    Done when: you have written a 4-page comparative case analysis evaluating 3 major government trials on effect size, cost-effectiveness, and policy scaling.

    How to work through it

    1. Review published reports from the UK Behavioural Insights Team and US Office of Evaluation Sciences
    2. Extract trial sample sizes, baseline rates, absolute vs relative effect sizes, and implementation costs
    3. Assess how policy-makers handled scaling and heterogeneity across sub-populations
    4. Author the comparative evaluation document
  • Author an applied Policy Intervention Brief and Scaling Plan
    ~8hApply

    Policy leaders require concise, evidence-backed memos that clearly link behavioural theory to budgetary and operational reality.

    You'll learn

    • Policy memo writing — concise, evidence-based executive communication tailored to senior civil servants
    • Implementation fidelity — the degree to which an intervention is delivered as intended in real-world settings

    Write a professional policy brief proposing a behavioural intervention to a government department (e.g., boosting uptake of preventative childhood vaccinations or energy-efficiency retrofits).

    Done when: you have produced a 5-page Policy Brief containing an executive summary, COM-B diagnosis, proposed trial design, cost-benefit model, and operational delivery roadmap.

    How to work through it

    1. Select a target public policy problem and relevant government agency
    2. Write the problem definition and behavioural diagnostic summary
    3. Specify the proposed multi-arm RCT including sample size requirements and partner agency roles
    4. Model return on investment and outline the scaling roadmap
    5. Finalize the 5-page Policy Brief
9

Applied Behavioural Science in Technology & Product

Translate behavioural mechanisms to digital product environments, covering user onboarding, engagement, retention, habit formation, and digital experiment design.

  • Deconstruct digital habit loops and engagement mechanics
    ~6hLearn

    Tech product behavioural science focuses heavily on long-term retention, habit loops, and reducing interaction friction.

    You'll learn

    • Habit loop — cue, craving, response, reward cycle that forms automatic behavioural routines
    • Variable reinforcement — scheduling rewards at unpredictable intervals to increase engagement
    • Goal gradient effect — tendency to increase effort as one gets closer to completing a goal

    Study habit formation loops (Cues, Routines, Rewards / Variable Reinforcement) and digital friction models across consumer applications (e.g., Duolingo, fintech apps, digital health).

    Done when: you have created a teardown report breaking down the habit formation loops and choice architecture of 2 consumer applications.

    How to work through it

    1. Study literature on habit loops, contextual cueing, and variable reward schedules
    2. Select two digital apps in education, health, or fintech
    3. Map their notification timing, onboarding steps, friction reducers, and reinforcement mechanisms
    4. Write the comparative teardown report
  • Design a digital A/B test specification for an app flow
    ~7hBuild

    In industry product teams, behavioural scientists work directly with product managers and engineers to define telemetry and experiment logic.

    You'll learn

    • Guardrail metrics — metrics tracked to ensure an intervention does not cause unintended harm to product health
    • Telemetry specification — schema defining user actions and metadata logged by analytics engines
    • Funnel drop-off analysis — tracking user abandonment across progressive steps in a conversion pipeline

    Write a product experiment specification for a digital app flow (e.g., improving payment setup completion or onboarding activation), including event tracking telemetry and variant designs.

    Done when: you have produced a Product Experiment Spec documenting user telemetry events, primary/secondary/guardrail metrics, UI mocks, and sample size requirements.

    How to work through it

    1. Define the target conversion funnel and drop-off points
    2. Draft UI/UX specifications for Control, Variant A, and Variant B
    3. Define exact telemetry event tracking properties (triggers, payload, schema)
    4. Specify guardrail metrics (e.g., unsubscribe rate, error rate, latency) to detect negative externalities
    5. Package into a standard Product Experiment Spec document
10

Ethics, Dark Patterns & Behavioural Audits

Establish ethical frameworks for behavioural science, learning to identify dark patterns, evaluate nudge transparency, and navigate public scrutiny.

  • Apply the FORGOOD ethical framework to intervention design
    ~5hPractice

    Because behavioural techniques alter choices, practitioners must maintain rigorous ethical accountability and transparency.

    You'll learn

    • FORGOOD framework — systematic ethical rubric for evaluating behavioural public policy and nudging
    • Autonomy preservation — ensuring choice architecture does not coerce or eliminate meaningful alternatives
    • Paternalism in policy — government shaping environments for citizen welfare vs individual agency

    Evaluate behavioural interventions against Lades & Delaney's FORGOOD framework (Fairness, Openness, Respect, Goals, Opinions, Options, Delegation).

    Done when: you have audited 3 real-world behavioural interventions using the FORGOOD score matrix and written a 2-page ethical risk analysis.

    How to work through it

    1. Study the 7 dimensions of the FORGOOD framework
    2. Select 3 interventions (one public policy, one consumer tech, one marketing)
    3. Score each intervention against each dimension and document risks
    4. Write the synthesized ethical evaluation report
  • Conduct a Dark Patterns and Sludge audit on a commercial service
    ~6hApply

    Distinguishing between helpful nudges and manipulative dark patterns is a vital compliance and ethical responsibility.

    You'll learn

    • Dark patterns — user interfaces designed to trick users into doing things they might not otherwise do
    • Confirmshaming — guilting the user into opting into something by framing the decline option negatively
    • Roach motel — making a situation very easy to get into (subscription) but difficult to get out of (cancellation)

    Perform an audit of deceptive design patterns (confirmshaming, roach motels, hidden costs, forced continuity) and administrative sludge in a subscription or retail platform.

    Done when: you have created an annotated Sludge & Dark Pattern Audit identifying at least 5 manipulative practices, explaining their cognitive mechanism and drafting ethical alternatives.

    How to work through it

    1. Review legal and regulatory definitions of deceptive digital design patterns
    2. Select a commercial service known for high cancellation or sign-up friction
    3. Document each dark pattern with screenshots and pinpoint the underlying heuristic exploited
    4. Design ethical alternative flows that maintain user autonomy
    5. Publish the final audit document
11

Portfolio Assembly & Applied Case Study Readiness

Synthesize all prior work into a polished professional portfolio and prepare for behavioural science case study interviews across policy and industry sectors.

  • Assemble 3 end-to-end Behavioural Science Case Studies
    ~12hBuild

    Behavioural science hiring managers evaluate candidate thinking through structured end-to-end case evidence rather than resume claims.

    You'll learn

    • Behavioural portfolio curation — presenting problem scoping, diagnostic rigor, and causal evaluation clearly
    • Executive storytelling — structuring complex psychological methodologies into digestible business outcomes

    Format your previous diagnostic maps, intervention prototypes, and trial plans into 3 structured portfolio case studies (1 Policy, 1 Tech/Product, 1 Research/Methodology).

    Done when: you have published a clean portfolio website or PDF dossier showcasing 3 complete end-to-end case studies covering diagnosis, intervention, trial design, and evaluation.

    How to work through it

    1. Structure each case study into: Context & Target Behaviour, COM-B/EAST Diagnosis, Intervention Design, Trial Protocol, and Statistical Evaluation Plan
    2. Incorporate high-resolution visual assets, flowcharts, and annotated mockups
    3. Ensure reproducible R code snippets and power calculations are embedded or linked
    4. Publish the finished portfolio dossier
  • Simulate a live 45-minute behavioural design whiteboard challenge
    ~4hApply

    The standard interview format for behavioural science roles is a live prompt where you must structure problem diagnosis and experiment design on the spot.

    You'll learn

    • Live case structuring — organizing an unstructured behavioural problem in real time under time constraints
    • Five-stage behavioural framework — rapid problem breakdown: Define, Diagnose, Design, Test, and Scale

    Practice tackling live behavioral problem prompts (e.g., 'How would you increase organ donation registrations among young adults?' or 'How would you decrease cart abandonment on a mobile app?').

    Done when: you have recorded a 45-minute solo mock whiteboard presentation walking through the 5-stage framework (Define, Diagnose, Design, Test, Scale) for an unfamiliar prompt.

    How to work through it

    1. Select an unseen behavioral design prompt from policy or tech
    2. Set a 45-minute timer and open a digital whiteboard (e.g., Miro or FigJam)
    3. Work through the 5 steps: Target Behaviour Definition, Behavioral Diagnosis, Ideation, RCT Experimental Setup, and Ethical Checks
    4. Review the recording against evaluation rubrics used by behavioral teams

How the plan fits together

11 phases in 8 stages. Anything on the same row can be worked on at the same time.

An arrow points from a phase to the work it unlocks: before starting any phase, every phase with an arrow into it has to be finished first.

STARTSTAGE 2STAGE 3STAGE 4STAGE 5STAGE 6STAGE 7STAGE 81Foundations of CognitivePsychology & DecisionScience2 tasks · ~11h2Behavioural Economics &Choice Under Uncertainty3 tasks · ~18h3Behavioural Diagnosis &Problem Scoping2 tasks · ~11h4Inferential Statistics &Power Analysis forExperiments3 tasks · ~17h5Experimental Design &Field Trials (RCTs & A/BTesting)3 tasks · ~19h6Qualitative &Mixed-Methods BehaviouralResearch2 tasks · ~12h7Choice Architecture &Intervention Design2 tasks · ~12h8Applied BehaviouralScience in Public Policy &Social Impact2 tasks · ~14h9Applied BehaviouralScience in Technology &Product2 tasks · ~13h10Ethics, Dark Patterns &Behavioural Audits2 tasks · ~11h11Portfolio Assembly &Applied Case StudyReadiness2 tasks · ~16h
Solid arrow
Must be finished before the phase it points to
Dashed arrow
Same rule, but the prerequisite sits more than one stage back

Resources

24 in this plan's library, beyond the links on individual tasks.

Learning & Reference

Foundational texts, diagnostic frameworks, and methodology guides.

  • BE101x: Behavioural Economics in Action

    Contrasts standard neoclassical economic assumptions with behavioral models such as prospect theory and hyperbolic discounting.

    University of Toronto (Rotman School of Management / BEAR) on edX · Online Course · Free to audit

  • Deceptive Patterns: Exposing the Tricks Tech Companies Use to Control You

    Teaches practitioners how to audit digital products for deceptive UX, manipulative interfaces, and sludge.

    deceptive.design · Testimonium Ltd (Harry Brignull) · Book · approx. $15–$25

  • Designing for Behavior Change: Applying Psychology and Behavioral Economics

    Step-by-step roadmap for implementing behavioural intervention frameworks in digital products and software user journeys.

    O'Reilly Media · Book · ~£35-45 · Intermediate

  • Designing for Behavior Change: Applying Psychology and Behavioral Economics (2nd Edition)

    Details how to operationalize behavioral psychology, user onboarding, and retention mechanisms in digital products.

    oreilly.com · O'Reilly Media · Book · approx. $40–$55

  • EAST: Four Simple Ways to Apply Behavioural Insights

    Distills choice architecture and behavioral design into four operational levers: Easy, Attractive, Social, and Timely.

    bi.team · The Behavioural Insights Team (BIT) · Report · Free

  • Field Experiments: Design, Analysis, and Interpretation

    Serves as the methodological reference for designing, running, and analyzing randomized controlled trials in real-world settings.

    wwnorton.com · W. W. Norton & Company · Book · approx. $60–$90

  • Improving Your Statistical Inferences

    Focuses on preventing false discoveries, interpreting p-values correctly, calculating effect sizes, and conducting power analyses.

    coursera.org · Eindhoven University of Technology on Coursera · Online Course · Free to audit

  • Just Enough Research (2nd Edition)

    Provides a practitioner guide to contextual inquiry, stakeholder interviews, and qualitative bias mitigation.

    Erika Hall / mule.is · Book · approx. $24

  • Misbehaving: The Making of Behavioral Economics

    Traces the divergence from neoclassical rational-agent models to descriptive behavioural economics, covering prospect theory, mental accounting, and fairness.

    W. W. Norton & Company · Book · ~£12-20 · Beginner to Intermediate

  • Running Randomized Evaluations: A Practical Guide

    A step-by-step handbook on conducting field RCTs, covering randomization mechanics, attrition threats, and compliance monitoring.

    Princeton University Press · Book · ~£35 · Intermediate to Advanced

  • The Behavioral Scientist

    Offers applied breakdowns and real-world commentary to learn effective narrative framing for case study portfolios.

    behavioralscientist.org · Behavioral Science Press · Digital Magazine · Free

  • The Behaviour Change Wheel: A Guide to Designing Interventions

    Provides a foundational manual and step-by-step methodology for diagnosing behavioral problems using the COM-B model.

    behaviourchangewheel.com · Silverback Publishing · Book · approx. £20–£30

  • Thinking and Deciding (5th Edition)

    Establishes the formal cognitive architecture behind judgment and decision-making under uncertainty.

    Cambridge University Press · Book · approx. $55–$70

  • Thinking, Fast and Slow

    Essential foundational reading covering dual-process cognitive architecture, cognitive ease, and core heuristics.

    Farrar, Straus and Giroux · Book · ~£12-18 · Beginner

  • Tools and Ethics for Applied Behavioural Insights: The BASIC Toolkit

    Provides a step-by-step public policy methodology covering Behavior identification, Analysis, Strategy, Intervention, and Change.

    oecd.org · OECD Publishing · Guide · Free to read online

  • Trustworthy Online Controlled Experiments: A Practical Guide to A/B Testing

    Bridges field trial theory with digital experimentation, guardrail metrics, and statistical pitfalls in product environments.

    cambridge.org · Cambridge University Press · Book · approx. $35–$50

Tools & Software

Statistical environments, diagnostic templates, and survey tools.

  • G*Power

    A standalone power analysis tool used to compute required sample sizes for statistical tests prior to launching trials.

    psychologie.hhu.de · Heinrich-Heine-Universität Düsseldorf · Software · Free

  • JASP

    A statistical package providing frequentist and Bayesian hypothesis testing with a clean graphical interface.

    jasp-stats.org · University of Amsterdam / JASP Team · Software · Free

Trial Registries & Case Repositories

Open registries of public and industry behavioural experiments.

  • Deceptive Patterns Pattern Library

    Taxonomy and repository of real-world dark patterns and coercive choice architectures to study during behavioural ethics audits.

    deceptive.design · Deceptive Design (Harry Brignull) · Digital Archive & Taxonomy · Free · Beginner

  • OECD Behavioural Insights Knowledge Hub

    Searchable international repository of public policy interventions, trial designs, and evaluated behavioural insights across governments.

    OECD · Case Database · Free · Intermediate

  • OECD Behavioural Insights Knowledge Hub & Case Repository

    A global repository of real-world public sector and civic behavioral interventions with evaluation metrics.

    oecd-opsi.org · OECD Observatory of Public Sector Innovation · Repository · Free

  • The American Economic Association's RCT Registry (AEA RCT Registry)

    A searchable international registry of randomized controlled trials across economics, public policy, and social sciences.

    socialscienceregistry.org · American Economic Association · Registry · Free

Communities & Professional Networks

Societies and forums for applied behavioural scientists.