EAST Nudge Framework: S-Tier Behavioral Designer’s Guide

Every government in 2014 had an “evidence-based policy” team. Almost none of them had a checklist their delivery managers could actually remember at the desk where the decision got made. Nine-lever inventories sat in PDFs nobody opened twice. Ten-step protocols got reduced to “let’s send a letter and see.” The gap between the academic literature on behavioural science and the documents civil servants used to ship interventions was wide enough to swallow most of the field’s political capital.

The UK Cabinet Office Behavioural Insights Team closed that gap with a 53-page paper in April 2014. Owain Service, Michael Hallsworth, David Halpern, Felicity Algate, Rory Gallagher, Sam Sanders, Marcos Pelenur, Alex Tupper, Hugo Harper, Joanne Reinhard, and Elspeth Kirkman titled it EAST: Four Simple Ways to Apply Behavioural Insights. The four ways were the title. Make it Easy. Make it Attractive. Make it Social. Make it Timely. Each lever pointed to a category of intervention, and each category came with three sub-tactics drawn from the BIT’s first three years of field trials.

EAST shipped because MINDSPACE shipped first and was too big to use. Dolan, Hallsworth, Halpern, King and Vlaev’s 2010 nine-lever checklist (Messenger, Incentives, Norms, Defaults, Salience, Priming, Affect, Commitments, Ego) had been the BIT’s academic anchor since the unit’s founding. By 2013 the team had run hundreds of trials and learned which levers actually moved policy outcomes versus which sat in the inventory because the literature said they should. Four made the cut. The rest were either dominated by Easy and Timely or were too granular for a non-academic practitioner to deploy under deadline pressure.

The post below is the S-Tier Behavioral Designer’s Guide to those four levers — what Service and his co-authors actually wrote, where the framework sharpens every other behaviour model you already know, where it falls apart, and how to wire it into an Octalysis-grade build so the nudge earns its political budget instead of burning it.

Speed Run Notes

• EAST replaced MINDSPACE’s nine levers with four. Easy, Attractive, Social, Timely. The compression was deliberate, not lazy — the BIT had run hundreds of trials and learned which levers moved policy and which did not.
• Each lever is a category, not a tactic. Easy contains defaults, hassle reduction, and message simplification. Picking the right category before picking the tactic is what makes EAST scale across departments and contexts.
• Easy and Timely do the heavy lifting in field trials. Attractive and Social win at the margin. A nudge that does not first reduce friction or hit a receptive moment usually does not survive contact with a normal citizen’s inbox.
• The four levers stack rather than substitute. The HMRC tax letter that lifted payment rates from ~33% to ~39% used Social (peer norm), Easy (single-page simplification), and Timely (sent at the seven-day reminder window) in one envelope.
• EAST without an ethics filter becomes a Sludge generator. The same four levers that drive tax compliance drive exploitative subscription dark patterns and political microtargeting. Apply the publicity test from Libertarian Paternalism before you ship.
• Octalysis Crosswalk: Easy = Core Drive 8 (status-quo bias). Attractive = Core Drives 7 and 4. Social = Core Drive 5. Timely = Core Drive 6. The four EAST levers map onto a behavioural-design octagon that already exists.

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.

What Is the EAST Framework?

EAST is the 2014 Behavioural Insights Team paper that compressed behavioural science into a four-lever inventory designed to fit in the head of a policy officer at 4pm on a Friday. Owain Service was the BIT’s first managing director. Michael Hallsworth ran the team’s tax-compliance and health portfolios. David Halpern was the unit’s founding director, the author of The Hidden Wealth of Nations, and the bridge between Whitehall and the academic behavioural-science community. Felicity Algate, Rory Gallagher, Sam Sanders, Marcos Pelenur, Alex Tupper, Hugo Harper, Joanne Reinhard, and Elspeth Kirkman ran the trials that supplied the evidence. The eleven authors shipped one document under the Cabinet Office imprint and built it as a practitioner’s checklist rather than an academic monograph.

The argument has three parts. First, behavioural science has accumulated more practical levers than any one civil servant can carry. A nine-lever or eleven-lever inventory looks comprehensive on paper and becomes a paper-only artefact in practice. Second, the BIT’s field-trial portfolio between 2010 and 2013 had revealed a clustering pattern: a handful of lever categories produced almost all of the measurable policy effects, while the rest contributed mostly at the margin or only in laboratory conditions. Third, a practitioner who can remember four categories at the desk will deploy more behavioural insight than one who keeps a nine-lever reference card pinned to their monitor and never actually consults it.

EAST’s design genius is the ordering of the letters. Easy is first because the single largest predictor of whether a citizen completes a desired behaviour is whether the path of least resistance leads there. Attractive is second because once the path is open, salience determines whether the citizen sees the prompt at all. Social is third because once the prompt is visible, the norm conveyed alongside it is the strongest driver of whether the citizen acts on it. Timely is fourth not because it matters least but because timing multiplies everything before it — the same letter delivered seven days late is a different intervention from the one delivered at the receptive moment. Each lever does work the others cannot, and the stack is multiplicative rather than additive when the design is honest.

The framework also imports the BIT’s test-and-iterate culture. Every lever in the EAST inventory points to specific field trials run by the unit between 2010 and 2014, each with a randomisation design, a measured outcome, and a published effect size. EAST is not a literature review. It is a checklist of what the BIT had verified worked, in the field, at scale, by the time the paper went to press.

The Four Levers of EAST

Each of the four EAST letters expands into three sub-tactics drawn from the BIT’s trial portfolio. The structure is intentional: the letter is what a busy practitioner remembers, the sub-tactics are what a behavioural-insights analyst checks when designing an actual intervention. Read together they describe the design surface where most policy nudges live.

Easy: Reduce the Hassle Factor, Use Defaults, Simplify the Ask

The Easy lever owns the single most reliable finding in applied behavioural science: people will do the thing the design makes effortless and not do the thing it makes effortful, even when the effort gap is small and the stakes are large. EAST organises this into three sub-moves. Harness the power of defaults: when a citizen does nothing, what happens? Defaults convert non-decision into behaviour, and the auto-enrolment literature shows the conversion is durable across years. Reduce the hassle factor of an action: count every click, page, form field, and queue between the citizen and the desired behaviour, and remove the ones that do not earn their place. Simplify messages: language at a sixth-grade reading level, single-page letters, one call-to-action per envelope.

The Easy lever’s evidential anchor inside the EAST paper is the BIT’s 2013 trial of HMRC tax-self-assessment reminder letters and, behind the scenes, the cross-national auto-enrolment literature that culminated in the UK’s Pensions Act 2008 default-on pension scheme. The HMRC trial alone lifted on-time payment by roughly six percentage points (from ~33% in the control arm to ~39% in the highest-performing treatment) at zero marginal cost to the agency. Six points sounds small until you multiply by the tax base. The 2008 pension default has since enrolled over ten million UK workers into workplace pensions who would not otherwise have participated.

Attractive: Attract Attention, Design Rewards That Actually Pay

Attractive is the lever that turns the citizen’s eye toward the prompt and convinces them the prompt is worth a thirty-second decision rather than a swipe to the next thing. EAST splits it into two sub-tactics. Attract attention through personalisation, novelty, and salient design: address the citizen by name, use a colour or graphic that breaks the visual rhythm of the rest of their inbox, place the prompt where the eye lands first on a page or screen. Design rewards and sanctions for maximum effect: rewards should feel proportionate but immediate, and sanctions should feel certain rather than severe (the Becker-style severity-equals-certainty trade-off that the behavioural-insights tradition revises in favour of certainty).

The cleanest demonstration of the Attractive lever in the BIT canon is the 2012 trial of court-fine demand letters. A standard demand letter went to defaulting fine recipients in the control arm. A treatment arm received a personalised SMS message addressed by name, sent the day before the bailiff would have been deployed. Payment rates more than doubled, and the personalised SMS cost about fifteen pence per recipient versus several hundred pounds for the bailiff visit it replaced. The attention shift — from generic envelope to named text — was the entire intervention.

Social: Show the Norm, Use Networks, Name Commitments

The Social lever encodes the single most consistently underestimated force in behavioural design: what other people in your reference group do is the strongest predictor of what you will do, and explicit articulation of the norm in the prompt itself shifts behaviour with no other intervention. EAST splits Social into three sub-tactics. Show that most people perform the desired behaviour: descriptive norms expressed as accurate, locally-relevant percentages outperform generic appeals by a meaningful margin. Use the power of networks: peer messengers (a doctor delivering health advice, a fellow taxpayer delivering compliance advice) outperform institutional messengers. Encourage people to make a commitment to others: an explicit commitment to a named third party converts intention into behaviour at a higher rate than the same intention articulated privately.

The reference trial for the Social lever inside EAST is the HMRC tax-reminder experiment Hallsworth, List, Metcalfe and Vlaev published in the Journal of Public Economics in 2017 from BIT data. The most effective letter variant contained one new sentence: “Nine out of ten people in the UK pay their tax on time.” That single sentence, combined with a local-context variant that named the recipient’s town, lifted on-time payment from ~33% to ~39%. The intervention shipped at zero marginal cost and is now part of HMRC’s standard letter template.

Timely: Prompt at the Receptive Moment, Plan Immediate Consequences

The Timely lever encodes the insight that the same nudge delivered at different points in a citizen’s life-cycle has different effect sizes, and that the variance can swamp every other design choice. EAST splits Timely into three sub-tactics. Prompt people when they are likely to be most receptive: at moments of transition (moving house, starting a job, having a child), in response to a trigger event (a court summons, a medical diagnosis), or at the natural beginning of a new period (a new year, a new month, a new pay cycle). Consider the immediate costs and benefits, not just the long-run ones: a citizen weighing whether to enrol in a pension responds to the felt cost of the next paycheck more than to the abstract benefit of a retirement four decades away. Help people plan their response: implementation intentions (“When X happens, I will do Y”) convert intention into action at a higher rate than intention alone.

The Timely lever’s canonical BIT trial is the loft-insulation uptake experiment run with British Gas in 2012. Earlier trials had offered subsidised loft insulation directly. Uptake was poor. The Timely-lever variant bundled the insulation offer with a loft-clearance service, on the theory that the binding constraint was not the insulation cost but the cognitive cost of emptying the loft first. Bundling lifted uptake by roughly five-fold. Same insulation, same subsidy, same letter — different timing relative to the citizen’s actual decision-making sequence.

What Service, Hallsworth & Halpern Got Right

The compression is the contribution. MINDSPACE’s nine-lever inventory was a fine academic taxonomy, and it sat in PDFs while practitioners reinvented the wheel. EAST’s four letters are remembered, taught at orientations, and applied at the desk. The framework’s adoption curve across the OECD between 2014 and 2020 outpaced every prior behavioural-science checklist in the same window. Memorability is not a soft virtue; it is the difference between an inventory that ships and one that does not.

The second thing the authors got right is the stack-ability claim. EAST is not a substitution menu where the designer picks one lever. It is a multiplicative checklist where the strongest interventions hit all four in one envelope. The HMRC letter that lifted on-time payment was Easy (single page), Attractive (personalised salutation), Social (peer-norm sentence), and Timely (sent at the seven-day reminder window). The framework explicitly invites this stacking, which is what allows trial designers to compose interventions rather than choose between mutually exclusive options.

The third right move was pairing EAST with Test, Learn, Adapt as canonical companion documents. EAST tells you what levers exist. Test, Learn, Adapt tells you how to discover which one actually works in your context. Neither document is sufficient alone — an inventory without a method generates plausible-but-untested interventions, and a method without an inventory generates rigorous trials of underspecified theories. The BIT shipped both papers within a two-year window precisely because they need each other.

The fourth right move, often missed, is the deliberate humility about effect sizes. The EAST paper does not promise transformational results from any single lever. It promises measurable, replicable, low-cost effects that compound when stacked and scale when applied across the millions of citizen interactions a typical government runs each year. That framing has kept EAST credible in a decade when behavioural science’s reputation has otherwise been bruised by the replication crisis. The BIT promised modest, durable wins. It delivered them.

Where EAST Falls Apart

Every framework that fits on a postcard pays a price for the compression. EAST’s price is paid in three currencies, and a designer who relies on it without seeing the bill ends up shipping interventions that look behavioural-science-grade and behave like marketing copy.

Critique 1: Four Levers Create the Illusion That All Problems Are Nudge-Shaped

The first failure mode is the most political. EAST is a behaviourally-informed-letter toolkit. Its lever set assumes the binding constraint between a citizen and the desired behaviour is some combination of friction, attention, social context, or timing. When that assumption holds, EAST is the right inventory. When the binding constraint is something else — insufficient income, lack of childcare, geographic distance from a service, a structural rule that disqualifies the citizen, a market failure upstream of the citizen’s decision — EAST quietly recasts the problem as a nudge problem and ships an intervention that addresses none of it.

The clearest example is the long-running British debate about job-seeker outcomes. Several BIT trials applied EAST-style interventions to Jobcentre Plus interactions in the mid-2010s with measurable but small effect sizes. The structural diagnosis — that the binding constraint for many long-term unemployed citizens was not the design of the job-seeker prompt but the available local job market, childcare costs, transport infrastructure, and skills mismatch — sat outside the lever inventory and therefore outside the trial. EAST did not get the diagnosis wrong; it got the diagnosis quietly avoided.

Critique 2: Effect Sizes Are Usually Small and Rarely Scale Linearly

The second failure mode is statistical. Most EAST-style nudges in the published BIT portfolio deliver effect sizes in the two-to-six percentage-point range. That is real, it is replicable, and on a tax base of millions it pays for the unit several times over. But two failure modes follow. First, a two-point effect from a single trial does not translate into a two-point effect when the same intervention is shipped to a fifth of the population without retesting — the trial population, the message wording, the channel, and the receptive moment all interact with effect size in ways that do not survive transplantation. Second, multiple stacked EAST nudges do not always compose multiplicatively; the second nudge often subsumes the variance the first one already captured, and the third nudge sometimes works against the first.

The cleanest demonstration is the Energy Behaviour Change trial portfolio. Early descriptive-norm interventions (showing households their energy use relative to neighbours) delivered roughly a two percent reduction in consumption. Scaling that intervention nationally produced a substantially smaller effect, and stacking it with additional nudges (planning prompts, gain-frame framing) produced returns that diminished sharply after the second lever. The four-lever stack is real, but it is not linear, and EAST does not warn the practitioner about the diminishing-returns curve.

Critique 3: The Toolkit Does Not Tell You What Not to Nudge

The third failure mode is ethical. EAST inventories the levers a behavioural-insights practitioner can pull. It is silent on the levers a behavioural-insights practitioner should not pull, and on the citizen-relationships those levers corrode when pulled aggressively. The result, observed across the 2018-2024 commercial-sector adoption wave, is that EAST has become a standard playbook for subscription dark patterns, microtargeted political advertising, and engagement-loop design in apps that would not pass a public-sector ethics review.

Sunstein’s 2022 Sludge formalises the diagnostic test the original EAST paper omits: any time you can pull a lever to make the desired behaviour easier, the same lever can be pulled in reverse to make the undesired behaviour harder. The EAST lever that simplifies pension auto-enrolment is the same lever that buries a subscription cancellation page behind seven clicks. Service, Hallsworth and Halpern’s paper assumes the practitioner’s intent is pro-citizen; that assumption survives in a Cabinet Office unit accountable to elected ministers and breaks the moment EAST is taught at a marketing conference.

What’s Really Happening Inside the Behavioural Insights Team

EAST’s structure tracks a specific organisational story the paper does not narrate. The Behavioural Insights Team was founded inside the Cabinet Office in July 2010 as a two-person experiment under David Cameron’s Coalition government. By 2014 it had grown to roughly thirty staff and was generating measurable departmental savings. By 2014 it had also begun spinning out: in February 2014 the BIT formally became a mutual joint venture, partly owned by its employees, by the UK government, and by Nesta. The EAST paper landed in the same year as that spinout. It is not a coincidence.

The four-lever architecture is partly an organisational artefact of the spinout. A unit moving from inside-government to inside-Nesta needed a productisable methodology — something that could be taught to client governments and corporate clients without requiring them to hire BIT-grade academics. Nine MINDSPACE levers do not productise. Four EAST levers do. The paper is simultaneously a research summary, a sales document, and a training manual, which is why it reads more as a checklist than an academic monograph and why each lever is paired with a concrete trial rather than a citation cluster.

The second organisational dynamic visible in the EAST architecture is the BIT’s deliberate tilt away from the laboratory and toward the field. By 2014 the unit had abandoned the practice of grounding interventions in lab-validated phenomena and adopted a “we only ship what we have field-tested in this department” rule. The four EAST levers all have field-trial provenance. The MINDSPACE levers that did not make the cut (Priming, Affect, Ego) had stronger laboratory pedigrees but thinner BIT-field portfolios, and the replication crisis was beginning to hit the laboratory literature behind several of them. EAST is partly a behavioural-science framework and partly the BIT’s public statement about which parts of behavioural science it had stopped trusting.

The third dynamic is the relationship between Halpern and his political principals. Inside the Nudge Unit, Halpern’s 2015 memoir, makes clear that the unit’s survival under the 2010-2015 Coalition and the 2015-2020 Conservative governments depended on producing legible, defensible policy wins on roughly an annual cadence. EAST’s four-lever structure is what a policy advisor produces when the ask is “give us a methodology that the Permanent Secretary can explain to the Cabinet in three minutes.” It is engineering for a political constraint, not pure science.

EAST vs Other Frameworks

EAST sits in a crowded inventory of behaviour-change frameworks. Comparing it to its closest neighbours is the fastest way to see what it does and does not do.

EAST vs MINDSPACE

MINDSPACE (Messenger, Incentives, Norms, Defaults, Salience, Priming, Affect, Commitments, Ego) is EAST’s direct predecessor and the framework EAST was deliberately built to replace. MINDSPACE has nine levers and a stronger academic pedigree; EAST has four levers and a stronger field-trial pedigree. The two map onto each other roughly as follows: Easy = Defaults + Salience-of-friction. Attractive = Salience + Incentives + Affect. Social = Messenger + Norms + Commitments + Ego. Timely is the one EAST lever with no MINDSPACE equivalent — the original 2010 framework had no explicit timing dimension, and the absence was a real gap. EAST’s addition of Timely is the most consequential analytical move in the compression, not the elimination of Priming or Affect.

EAST vs Test, Learn, Adapt

Haynes, Service, Goldacre and Torgerson’s 2012 Test, Learn, Adapt is EAST’s methodological sibling. Where EAST is a lever inventory, Test, Learn, Adapt is the randomised-controlled-trial protocol that tells you how to determine which lever actually worked in your context. The two papers are designed to be used together: EAST tells you what intervention to design, Test, Learn, Adapt tells you how to design the trial that proves it worked. A team that adopts EAST without Test, Learn, Adapt produces plausible interventions and never knows if they are doing anything. A team that adopts Test, Learn, Adapt without EAST produces rigorous trials of underspecified theories. The full BIT stack is both documents plus the publicity test from Libertarian Paternalism.

EAST vs COM-B

Susan Michie’s COM-B (Capability + Opportunity + Motivation = Behaviour) is the diagnostic framework EAST silently assumes. COM-B answers the upstream question: which dimension of the citizen’s behaviour is broken? EAST answers the downstream question: given that the relevant dimension is broken, which lever do you pull to fix it? Easy maps roughly onto Opportunity (Physical). Social maps onto Opportunity (Social). Attractive and Timely map onto Motivation (Reflective and Automatic respectively). Capability sits mostly outside the EAST inventory, which is why EAST struggles with interventions where the citizen lacks the skill or knowledge to act regardless of lever choice. Use COM-B to diagnose; use EAST to prescribe.

EAST vs the Octalysis Framework

The Octalysis Framework is the behavioural-design system I introduced in 2015 and developed across Actionable Gamification. Where EAST is a four-lever practitioner’s checklist drawn from public-sector field trials, Octalysis is an eight-Core-Drive motivational design system drawn from game-design canon plus behavioural-science integration. Each Octalysis Core Drive (Epic Meaning, Development, Empowerment, Ownership, Social Influence, Scarcity, Unpredictability, Loss Avoidance) wraps the motivational psychology that makes an EAST lever land. EAST’s four levers fit cleanly inside the Octalysis octagon, which is what the Apply-Octalysis section below makes explicit. EAST gives you the lever; Octalysis tells you which Core Drive the lever is pulling on and where the design will burn out if you over-rely on it.

EAST in the Real World

EAST is the most-cited behavioural-insights framework in field deployment because it is grounded in actual field trials. The five case studies below are drawn from the BIT’s 2010-2018 published portfolio and from companion academic publications. Each one is a verified, replicated nudge with a measured effect size; none is a press-release win.

HMRC Tax-Reminder Letters

The HMRC tax letter is the canonical EAST stack. The pre-trial control letter was a standard institutional demand: roughly two pages, formal language, no personalisation. The treatment letter introduced four changes simultaneously. Easy: single page, sixth-grade reading level, one call-to-action. Attractive: addressed by name, payment deadline highlighted at the top. Social: the sentence “Nine out of ten people in the UK pay their tax on time” was added, with a local-context variant naming the recipient’s town. Timely: sent at the seven-day reminder window when behavioural-economic literature predicts maximum receptivity. The Hallsworth, List, Metcalfe and Vlaev Journal of Public Economics 2017 publication, drawing on BIT trial data, reports a roughly six-percentage-point lift in on-time payment rates and an effect that compounds across the tax base into hundreds of millions of pounds collected earlier each year. The intervention is now standard HMRC practice.

GP Antibiotic-Prescribing Letters

The Hallsworth, Chadborn, Sallis and colleagues 2016 Lancet publication reports a BIT trial that addressed England’s antibiotic-overprescription problem. High-prescribing General Practitioners received a letter from England’s Chief Medical Officer naming their practice as being in the top twenty percent of prescribers in their local area. The Social lever (peer norm via authoritative messenger) and the Attractive lever (named personalisation) did the heavy lifting; Easy contributed via the single-page format and Timely via delivery at the start of the prescribing year. The trial reduced antibiotic prescribing by roughly three percent across the treated practices, sustained over the six-month measurement window. Three percent at NHS prescribing volume translates to hundreds of thousands of fewer antibiotic prescriptions per year, with measurable downstream effects on antimicrobial-resistance trajectories.

Court-Fine SMS Reminders

The Haymarket Magistrates’ Court trial (2012) targeted defaulting fine recipients with a personalised SMS message addressed by name, sent the day before the bailiff visit. The Timely lever (the day-before window) and the Attractive lever (named personalisation in a channel typically used by family) did almost all of the work; Easy contributed via the one-tap reply format and Social was minimally engaged. Payment rates more than doubled relative to the standard letter-only control arm. The intervention cost approximately fifteen pence per SMS, against several hundred pounds per bailiff visit avoided. The trial scaled across Her Majesty’s Courts and Tribunals Service and is now part of standard fine-recovery practice in England and Wales.

Organ Donor Registration

The organ-donor lever debate in the EAST literature centres on the question of opt-in versus opt-out defaults and on the design of the prompt at the moment of driver’s-licence renewal. Johnson and Goldstein’s 2003 Science cross-national comparison showed registration rates of roughly fifteen percent under opt-in and over eighty-five percent under opt-out across otherwise-comparable countries. The BIT’s 2013 trial applied an EAST-stack intervention to the UK’s opt-in driver-licence-renewal page: Easy (one-click registration), Attractive (a salient prompt at the moment of renewal), Social (the descriptive norm sentence “Every day thousands of people who need an organ transplant…”), and Timely (delivered at the licence-renewal transition moment). The trial increased registration rates by roughly one-third over the control arm. The structural opt-out shift came later, with the UK Organ Donation (Deemed Consent) Act 2019, but the EAST intervention demonstrated that even within an opt-in regime, lever-stacking produced material lift.

Pension Auto-Enrolment

The UK Pensions Act 2008 instituted automatic enrolment of UK workers into workplace pension schemes — the largest pure-default intervention in modern policy history. Thaler and Benartzi’s 2004 Save More Tomorrow programme is the academic ancestor; the UK Act is the Easy-lever industrial-scale deployment. Roughly ten million UK workers have been enrolled into a workplace pension who would not otherwise have participated under the prior opt-in regime. Opt-out rates have stabilised at roughly nine percent, well below the participation gap under opt-in. The intervention pulls Easy almost exclusively (no behavioural-economic prompt at all is required — the default does the work) with Timely contributing through the moment-of-employment trigger. It is the cleanest demonstration in the EAST canon that a single lever, pulled at industrial scale, can outperform every clever multi-lever stack ever published in a journal.

The Elephant in the Room

EAST is morally neutral. The same four levers that drive HMRC tax compliance drive Netflix free-trial-to-paid conversion, the same four levers that drive antibiotic-prescribing reduction drive opioid-prescribing patterns when the pharmaceutical-marketing team gets hold of them, and the same four levers that drive organ-donor registration drive political-advertising microtargeting at scale. The EAST paper assumes a Cabinet Office unit accountable to elected ministers will only pull the levers in pro-citizen directions. That assumption survives the original organisational context and does not survive the framework’s diffusion into commercial and political settings where the practitioner’s intent is not subject to the same accountability.

The publicity test imported from Thaler and Sunstein’s libertarian-paternalism doctrine is the canonical ethics filter the EAST paper should have included and did not. Before deploying a four-lever stack, ask: if the citizen knew about every lever you were pulling, would they consent to the nudge? If the answer is yes, the design is doing pro-citizen work and the EAST inventory is a tool. If the answer is no, you are designing what Sunstein later named Sludge, and the EAST inventory is a weapon. The publicity test is not a deep ethical theory; it is a practical filter a busy practitioner can apply at the desk in thirty seconds.

An EAST-validated intervention that fails the publicity test is more dangerous than an unvalidated one. The published effect-size literature lends moral cover to the rollout. The trial registration looks legitimate. The peer-reviewed paper sits behind the marketing team’s deck. The citizen on the receiving end has no idea that the same techniques that auto-enrolled them into a pension are now auto-enrolling them into a recurring subscription their elderly parent cannot find the cancellation link for. The framework that compressed behavioural science for civil servants compressed it for everyone, and the elder fraud, the political microtargeting, and the engagement-loop design now ship inside the same four-letter inventory.

How to Apply EAST with the Octalysis Framework

EAST tells you which lever to pull. The Octalysis Framework tells you which Core Drive that lever is operating through — which lets you predict where the design will burn out, which complementary Core Drives can extend its reach, and which substitute drives could carry the same outcome through a different psychological route. The crosswalk below maps each EAST lever onto the Core Drive it primarily activates, with two additional Throughout rows covering the design dimensions every EAST stack quietly assumes.

Easy: Core Drive 8 (CD8): Loss & Avoidance

The Easy lever pulls on Core Drive 8 (CD8): Loss & Avoidance more than any other Core Drive. The default-bias literature is, mechanically, the literature of CD8: citizens stay enrolled in the default because changing it triggers the felt loss of whatever the default provides. Reducing the hassle factor is removing CD8 friction (the perceived loss of time and cognitive effort the citizen weighs against the desired action). Message simplification is removing the cognitive-load tax that triggers the CD8 reflex to abandon. A designer using Easy is, behaviourally, removing CD8 tax in some places and routing the citizen toward a default whose abandonment would activate CD8 in others. Both moves are the same Core Drive in opposite directions.

Attractive: Core Drive 7 (CD7): Unpredictability & Curiosity + Core Drive 4 (CD4): Ownership & Possession

The Attractive lever pulls on two Core Drives in tandem. Core Drive 7 (CD7): Unpredictability & Curiosity carries the attention shift — novelty, salient design, and the visual interruption pattern that breaks the citizen out of inbox-scan mode. Core Drive 4 (CD4): Ownership & Possession carries the personalisation effect — the citizen’s name in the salutation, the local-context variant, and the address-line specificity trigger the felt-ownership response that makes the prompt feel directed at the citizen rather than at a demographic. The two drives stack: CD7 wins the eye, CD4 holds the attention long enough for the actual ask to register. An Attractive design that hits only one of the two Core Drives is the marketing equivalent of an envelope that gets opened and immediately discarded.

Social: Core Drive 5 (CD5): Social Influence & Relatedness

The Social lever pulls on Core Drive 5 (CD5): Social Influence & Relatedness almost exclusively. Descriptive norms are CD5 in its purest form: the citizen’s sense of what people-like-me do shapes what they do next. Peer-messenger interventions are CD5 routed through the source-credibility channel: a doctor delivering a health message is CD5 by professional identity, a fellow taxpayer delivering a compliance message is CD5 by membership identity. Public commitments are CD5 by accountability: the citizen’s articulated promise to a named third party converts intention into behaviour because the cost of breaking the promise is a CD5-mediated reputational loss. The HMRC peer-norm sentence is the cleanest single-sentence CD5 deployment in the public-policy literature.

Timely: Core Drive 6 (CD6): Scarcity & Impatience

The Timely lever pulls on Core Drive 6 (CD6): Scarcity & Impatience. Receptive-moment prompting is scarcity in time: the prompt arrives at a window that will close, and the closing window converts the abstract benefit of acting into the felt urgency of acting now. Implementation intentions are CD6 routed through structured patience: the if-then plan converts a future moment of decision into a pre-committed action, removing the impulsive defection CD6 would otherwise produce. Immediate-cost-and-benefit framing is CD6 in its consumer form: the felt cost of the next paycheck dominates the abstract benefit of a retirement four decades away, and a designer using Timely is restructuring the citizen’s perceived scarcity to align with the policy outcome.

Throughout (Wrapping): Core Drive 1 (CD1): Epic Meaning & Calling

Every EAST stack that earns its political budget wraps the four levers in a Core Drive 1 (CD1): Epic Meaning & Calling frame. The HMRC letter works partly because paying tax is framed as contribution to a collective public good; the antibiotic-prescribing letter works partly because the doctor frames their prescribing decision as part of a collective antimicrobial-stewardship mission. CD1 is what distinguishes a pro-citizen nudge from a manipulative one. The publicity test, behaviourally, is asking whether the CD1 frame would survive disclosure to the citizen. A nudge that needs to hide its CD1 frame is not a nudge; it is a deceptive marketing pattern.

Throughout (Progression): Core Drive 2 (CD2): Development & Accomplishment

EAST stacks compound when paired with a Core Drive 2 (CD2): Development & Accomplishment progression structure. Pension auto-enrolment scales partly because the citizen can see their pension balance grow over time, and the visible progress reinforces the Easy-lever default. Antibiotic-prescribing feedback letters work partly because the GP receives quarterly updates showing their prescribing trajectory relative to peers, converting a one-shot Social-lever nudge into a CD2 development loop. A designer treating EAST as a one-shot intervention misses the compounding effect a CD2-wrapped progression structure produces.

Throughout (Agency): Core Drive 3 (CD3): Empowerment of Creativity & Feedback

The Core Drive every EAST stack should respect and many do not is Core Drive 3 (CD3): Empowerment of Creativity & Feedback. CD3 asks: does the citizen feel like an agent making a choice, or like a target being managed? A nudge that pulls Easy, Attractive, Social and Timely without leaving the citizen room to exercise judgement is mechanically efficient and ethically corrosive. The publicity test, again, sits on top of CD3: a design that preserves the citizen’s agency would survive disclosure; a design that strips it would not. The libertarian-paternalism doctrine is essentially the requirement that every EAST stack also pass a CD3 audit.

Practical Steps: The EAST × Octalysis Audit

The audit below operationalises the EAST framework with the Octalysis Core Drive crosswalk and the libertarian-paternalism ethics filter. Run it before designing the trial, not after the trial has shipped.

1. Diagnose the binding constraint with COM-B first, then reach for EAST. Before assuming the problem is nudge-shaped, ask: is the constraint Capability, Opportunity, or Motivation? If Capability (the citizen lacks the skill, knowledge, or physical resource to act regardless of lever choice), EAST will produce theatre. Fix Capability first. EAST applies when Opportunity or Motivation is the binding constraint.
2. Run the publicity test before drafting the intervention. If you cannot describe the four levers you are about to pull to the citizen and have them consent, the design fails the libertarian-paternalism ethics filter and you should stop. The publicity test takes thirty seconds at the desk and prevents the Sludge pattern that EAST’s commercial diffusion enabled.
3. Stack the levers honestly, not maximally. Four levers in one envelope is the design target only when each lever does work the others do not. A second Social lever stacked on a first Social lever subsumes variance rather than adding it. Pre-register the trial with effect-size predictions for each lever individually so the stack’s diminishing returns are measurable.
4. Use Test, Learn, Adapt as your trial protocol. EAST tells you which levers to pull. Test, Learn, Adapt tells you whether they worked. Adopting EAST without the BIT’s randomised-controlled-trial method generates plausible but untested interventions. Adopting both is the full BIT stack.
5. Wrap the stack in Core Drive 1 and design Core Drive 2 progression alongside. Single-shot nudges decay. EAST stacks paired with a CD1 mission frame and a CD2 progression structure compound across the citizen’s engagement window. The HMRC letter scaled partly because each year’s tax cycle reinforced the prior year’s nudge.
6. Audit Core Drive 3 last. Before shipping, re-read the intervention and ask: does the citizen still feel like an agent? A design that lifts measured outcomes by suppressing CD3 is a design that erodes the institutional trust the BIT spent fifteen years building. The publicity test and the CD3 audit are the same audit run twice.

EAST Was the Beginning, Not the End

EAST is the most-adopted behavioural-insights framework in policy practice because it solved the compression problem the field needed solved in 2014. The four levers fit on a Post-it, the case studies have published effect sizes, and the methodology pairs cleanly with Test, Learn, Adapt for the rigorous-trial half of the stack. A decade later, the framework is taught in every public-sector behavioural-insights unit in the OECD and most of the commercial behavioural-design teams in the Anglophone consulting world.

The decade also exposed what the four-lever compression left out. Capability sits outside the inventory, which is why EAST struggles when the binding constraint is skill or structural access rather than friction or attention. Effect sizes do not scale linearly across populations or stack multiplicatively across levers, which is why second-order trial results often disappoint relative to first-order publications. The framework is morally neutral by design, which is why commercial adoption has produced as much Sludge as it has produced citizen-benefit. Each of these is a real critique. None invalidates the framework; all sharpen it.

The next BIT-grade behavioural-design system will need to do three things EAST does not. First, embed a Capability dimension alongside the four levers so the diagnostic step is forced before the prescriptive step. Second, ship with an ethics filter built into the methodology rather than imported from a sister paper. Third, integrate the Octalysis Core Drive crosswalk explicitly so the practitioner sees which motivational route each lever is operating through and where the design will burn out. The Octalysis Framework, paired with the BIT’s trial method and the libertarian-paternalism ethics filter, is one path to that next system. EAST was the beginning, not the end, of how applied behavioural science gets translated into policy that works.

Frequently Asked Questions

What is the EAST framework in one sentence?

EAST is the UK Behavioural Insights Team’s 2014 four-lever practitioner’s checklist — make the desired behaviour Easy, Attractive, Social, and Timely — designed as a compressed, field-tested successor to the MINDSPACE nine-lever inventory.

Who wrote the EAST framework?

The 2014 EAST paper was authored by Owain Service, Michael Hallsworth, David Halpern, Felicity Algate, Rory Gallagher, Sam Sanders, Marcos Pelenur, Alex Tupper, Hugo Harper, Joanne Reinhard, and Elspeth Kirkman of the UK Cabinet Office Behavioural Insights Team. David Halpern, the BIT’s founding director, is the most frequently cited single author when the framework is attributed to one person, though the team authorship reflects the unit’s actual practice.

How is EAST different from MINDSPACE?

MINDSPACE is the BIT’s 2010 nine-lever academic taxonomy (Messenger, Incentives, Norms, Defaults, Salience, Priming, Affect, Commitments, Ego). EAST is the BIT’s 2014 four-lever practitioner’s compression of the same field, with one important addition: Timely, the timing dimension MINDSPACE lacked. EAST eliminates Priming and Affect (whose laboratory evidence was weakening) and consolidates the remaining MINDSPACE levers into Easy, Attractive and Social.

How is EAST different from Test, Learn, Adapt?

EAST and Test, Learn, Adapt are companion documents. EAST is the lever inventory: which intervention to design. Test, Learn, Adapt is the trial protocol: how to discover whether the intervention actually worked. Using EAST without Test, Learn, Adapt produces plausible untested interventions. Using Test, Learn, Adapt without EAST produces rigorous trials of underspecified theories. The full BIT stack is both.

What is the most common failure mode of an EAST deployment?

The most common failure mode is the all-problems-are-nudge-shaped misdiagnosis. EAST is the right inventory when the binding constraint between the citizen and the desired behaviour is some combination of friction, attention, social context, or timing. When the binding constraint is something else — income, childcare, geographic distance, a structural rule, an upstream market failure — EAST quietly recasts the problem and ships an intervention that addresses none of it.

Which EAST lever has the strongest evidence base?

Easy, by a meaningful margin. The default-bias literature, the auto-enrolment cross-national comparison, and the UK Pensions Act 2008 industrial-scale deployment all point to Easy producing the largest, most durable effect sizes in the EAST inventory. Timely is second. Attractive and Social produce smaller but reliably replicable effects, especially when stacked on Easy.

Can EAST be used in commercial settings or only in public policy?

EAST can be used in commercial settings, but the absence of a built-in ethics filter is a real problem. The four levers that drive tax compliance drive subscription dark patterns and engagement-loop addiction when the practitioner’s intent is not pro-citizen. Commercial EAST deployment without a publicity-test ethics gate is one of the larger sources of the Sludge pattern Cass Sunstein formalised in his 2022 book.

How does EAST relate to the Octalysis Framework?

Each EAST lever pulls on a specific Core Drive in the Octalysis octagon. Easy = Core Drive 8 (Loss & Avoidance, via default-bias and friction). Attractive = Core Drives 7 and 4 (Unpredictability & Curiosity plus Ownership & Possession). Social = Core Drive 5 (Social Influence & Relatedness). Timely = Core Drive 6 (Scarcity & Impatience). The Octalysis crosswalk lets a practitioner see which motivational route each EAST lever is operating through and design complementary or substitute Core Drives accordingly.

What is the ethics filter EAST should pass before deployment?

The libertarian-paternalism publicity test imported from Thaler and Sunstein’s Nudge: if the citizen knew about every lever you were pulling, would they consent to the nudge? If yes, the design is doing pro-citizen work. If no, you are designing Sludge. The publicity test takes thirty seconds at the desk and prevents the most common ethical failure mode of EAST deployment.

What is the canonical first intervention for a team adopting EAST?

A reminder letter or message intervention is the canonical first deployment because it stacks all four levers cleanly, costs almost nothing at scale, and produces measurable effect sizes within a one-quarter trial window. The HMRC tax-reminder letter is the BIT’s reference case; the GP antibiotic-prescribing letter and the court-fine SMS reminder are the other two canonical first-deployment templates a new team can adapt to local context.

References

Service, O., Hallsworth, M., Halpern, D., Algate, F., Gallagher, R., Sanders, S., Pelenur, M., Tupper, A., Harper, H., Reinhard, J., & Kirkman, E. (2014). EAST: Four Simple Ways to Apply Behavioural Insights. London: The Behavioural Insights Team / UK Cabinet Office.

Dolan, P., Hallsworth, M., Halpern, D., King, D., & Vlaev, I. (2010). MINDSPACE: Influencing Behaviour Through Public Policy. London: Cabinet Office and Institute for Government.

Halpern, D. (2015). Inside the Nudge Unit: How Small Changes Can Make a Big Difference. London: WH Allen.

Haynes, L., Service, O., Goldacre, B., & Torgerson, D. (2012). Test, Learn, Adapt: Developing Public Policy with Randomised Controlled Trials. London: Cabinet Office Behavioural Insights Team.

Thaler, R. H., & Sunstein, C. R. (2008). Nudge: Improving Decisions About Health, Wealth, and Happiness. New Haven: Yale University Press.

Sunstein, C. R. (2022). Sludge: What Stops Us from Getting Things Done and What to Do About It. Cambridge: MIT Press.

Hallsworth, M., List, J. A., Metcalfe, R. D., & Vlaev, I. (2017). The behavioralist as tax collector: Using natural field experiments to enhance tax compliance. Journal of Public Economics, 148, 14–31.

Hallsworth, M., Chadborn, T., Sallis, A., Sanders, M., Berry, D., Greaves, F., Clements, L., & Davies, S. C. (2016). Provision of social norm feedback to high prescribers of antibiotics in general practice: A pragmatic national randomised controlled trial. The Lancet, 387(10029), 1743–1752.

Thaler, R. H., & Benartzi, S. (2004). Save More Tomorrow: Using behavioral economics to increase employee saving. Journal of Political Economy, 112(S1), S164–S187.

Madrian, B. C., & Shea, D. F. (2001). The power of suggestion: Inertia in 401(k) participation and savings behavior. Quarterly Journal of Economics, 116(4), 1149–1187.

Johnson, E. J., & Goldstein, D. (2003). Do defaults save lives? Science, 302(5649), 1338–1339.

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).

Cialdini, R. B. (2008). Influence: Science and Practice (5th ed.). Boston: Pearson.

Schultz, P. W., Nolan, J. M., Cialdini, R. B., Goldstein, N. J., & Griskevicius, V. (2007). The constructive, destructive, and reconstructive power of social norms. Psychological Science, 18(5), 429–434.

Allcott, H. (2011). Social norms and energy conservation. Journal of Public Economics, 95(9–10), 1082–1095.

Bicchieri, C. (2017). Norms in the Wild: How to Diagnose, Measure, and Change Social Norms. Oxford: Oxford University Press.

Sanders, M., Snijders, V., & Hallsworth, M. (2018). Behavioural science and policy: Where are we now and where are we going? Behavioural Public Policy, 2(2), 144–167.

John, P. (2018). How Far to Nudge? Assessing Behavioural Public Policy. Cheltenham: Edward Elgar.

UK Department for Work and Pensions (2017). Automatic Enrolment Evaluation Report 2017. London: DWP.

Chou, Y. (2015). Actionable Gamification: Beyond Points, Badges, and Leaderboards. Fremont: Octalysis Media.

Related Reading

• MINDSPACE Framework (Dolan, Hallsworth, Halpern, King & Vlaev, 2010) — the nine-lever predecessor EAST compresses.
• Test, Learn, Adapt (Haynes, Service, Goldacre & Torgerson, 2012) — the BIT’s companion randomised-controlled-trial protocol.
• Libertarian Paternalism (Thaler & Sunstein) — the ethics filter EAST should pass before deployment.
• Sludge (Sunstein, 2022) — EAST’s evil twin and the diagnostic for when the four levers are being pulled against the citizen.
• COM-B / Behavior Change Wheel (Michie et al., 2011) — the diagnostic framework EAST silently assumes.
• The Behavioral Framework Library — the full inventory of behavioural-design frameworks every S-tier designer should know.
• The Octalysis Framework — the eight Core Drives that explain which motivational route each EAST lever is operating through.

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