Behavior Change Design: The Fish on the Beach Framework

Most people fail at behavior change design because they misunderstand the problem. You see someone struggling with a bad habit, and you assume they lack awareness or willpower. But that’s rarely the issue. The real insight comes from Stephen Wendel’s “fish on the beach” metaphor. Imagine you’re walking along a beach and you see a fish gasping on the sand. What do most people do? They yell “Get in the water.” They write articles about the benefits of water. They show charts about oxygen levels.

The fish already knows it needs water. It wants to be in the water. The gap isn’t knowledge or motivation. The gap is between intention and action. This is the central problem that behavior change design must solve, and most interventions fail because they target the wrong gap.

Speed Run Notes

• The intention-action gap is the real problem. Most people know what to do. They just don’t do it.
• Strategy 1 (Automation) is most effective but feels counterintuitive. Auto-enrollment jumps 401(k) participation from 37% to 86%.
• Strategy 2 (Simplify) removes skill or willpower requirements. Portable rules beat complex tools.
• Strategy 3 (Environment) gets most design resources but is often least effective long-term.
• Six diagnostic dimensions: Cue, Reaction, Evaluation, Ability, Timing, Experience.
• Automation works because it bypasses the brain entirely. You don’t need new neural pathways.
• Long-term change requires identity shift from “I do X” to “I am someone who does X.”
• Test your assumptions. PlayPump failed because it misdiagnosed the actual behavior problem.
• Gym memberships expose how we sell the feeling of change rather than designing for actual change.

Table of Contents

• The Three Strategic Responses
• Strategy 1: Remove Humans Entirely (Automating the Decision)
• Strategy 2: Simplify the Action Itself
• Strategy 3: Change the Environment (Architecture and Friction)
• Understanding the Diagnostic Framework
• The Long-Term Change Problem
• The Neuroscience Beneath the Strategy
• The Design Hierarchy
• The PlayPump Cautionary Tale
• Why Gym Memberships Expose Our Confusion
• Badges, Leaderboards, and Actual Accomplishment
• How Octalysis Maps the Three Strategies
• Connecting to Broader Gamification Principles
• Frequently Asked Questions

About the Creator of the Octalysis Framework

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 3,700+ more academic publications. Explore his books here.

The fish on the beach: it already wants to change, but the gap between intention and action remains

The Three Strategic Responses

Wendel’s framework gives us three ways to solve this intention-action gap. Each has different effectiveness, different costs, and different psychological implications. Understanding which one to use is the core skill of behavior change design.

1. Automate the decision — remove the human from the loop. Auto-enrollment, default settings, AI-driven choices.
2. Simplify the action — make the desired behavior easier than the current behavior. Bodyweight exercises, portable rules, smaller steps.
3. Change the environment — redesign the context the user moves through. Choice architecture, friction removal, defaults.

Strategy 1: Remove Humans Entirely (Automating the Decision)

The first strategy is counterintuitive: knock the fish out and carry it. Don’t ask the fish to swim. Don’t make swimming easier. Remove the fish from the equation entirely. In human behavior, this means automation and default settings. Remove the human decision-maker from the loop.

The most famous example is the auto-enrollment of 401(k) retirement plans. When companies made contributions automatic (with opt-out available), participation jumped from 37% to 86%. Nothing changed about education, motivation, or financial literacy. The system simply didn’t require a decision. The intention and the action became identical because the action happened without conscious choice.

Strava and Apple Fitness offer the contemporary version of this. The phone or watch passively tracks your run the moment it senses motion. RunKeeper pioneered the pattern in the early 2010s, and the pattern won: users who rely on automatic tracking complete substantially more runs than users who manually log workouts. Same goal, same fitness motivation, different friction. The automatic version wins decisively.

This is the most effective strategy, yet designers rarely use it first. Why? Because it feels like cheating. It circumvents the human mind rather than improving it. There’s a cultural bias toward strategies that respect autonomy and individual choice. But the data is clear: automation works. The resistance you feel is cultural, not evidence-based.

Myelinated neural pathways from childhood are 100x more efficient than new adult pathways

The most striking real-world case I’ve seen for layered automation is Caixa Bank’s adoption sweep, where my team’s redesign moved a key digital-engagement behavior from 10% adoption to 92%. The mechanism wasn’t education or motivation. It was Strategy 1 stacked on Strategy 3: default users into the desired path, then remove every avoidable friction step in the flow. Once you’ve combined the two correctly, the only users who don’t take the action are users who genuinely do not want it — which is the right floor, not 10%.

Strategy 2: Simplify the Action Itself

When automation isn’t possible or desirable, the second strategy is to make the target behavior dramatically easier. Pick the fish up and start moving it toward water in smaller steps.

Bodyweight exercises offer a clear advantage over gym-based routines. Same muscles engaged, same strength gains possible. But the friction is lower. You remove the need to commute, pay membership fees, or face social anxiety in a gym setting. People who can do push-ups at home are more likely to do them than people who must drive to a facility.

Stephen Wendel has shared his own regret about HelloWallet, a financial management tool his team built. The app required users to make sophisticated budgeting decisions and track spending across categories. In hindsight, he realized the approach was backwards. What users actually needed weren’t complex tools. They needed portable rules of thumb: “Always pay twice the minimum payment.” “Put 10% of your paycheck into savings.” Simple heuristics, not powerful applications.

Simplification works because it lowers self-efficacy demands. You don’t need to believe you’re capable of managing a budget spreadsheet. You just need to follow one rule. The burden shifts from execution skill to habit formation.

Strategy 3: Change the Environment (Architecture and Friction)

The third approach is to dig a channel. Change the environment so water flows toward the fish. This includes nudging, removing friction, and reshaping choices through what behavioral economists call choice architecture.

Placing fruit at eye level instead of vegetables at checkout increases fruit purchases. Automatic opt-in for organ donation versus opt-out changes donation rates from 10% to 90% across countries. These are environment changes, not character changes. The person’s internal motivation doesn’t shift, but their context does.

Interestingly, this is where most organizational energy and design resources flow. Yet it’s often the least effective of the three strategies. Why do designers gravitate here? The approach feels safe and respects human autonomy. No automation that might seem paternalistic. But the catch is that environmental changes require constant vigilance. Remove the nudge, and people revert. The behavior wasn’t internalized.

Understanding the Diagnostic Framework

Before choosing which strategy to deploy, you need to diagnose what’s actually blocking behavior change. I’ve found six diagnostic dimensions particularly useful.

Wendel labels these six dimensions with the acronym CREATE — Cue, Reaction, Evaluation, Ability, Timing, Experience. Naming the acronym matters because it lets you diagnose breakdowns precisely instead of waving at “motivation” as a single variable.

The first letter is the most underestimated. Wendel calls it Cue — does the person remember the desired behavior exists? — and most products fail right here. Attention is a scarce resource. Notice the language we use: we pay attention. It’s a budget you spend, not a faucet that flows. Designers who try to win attention by yelling louder (banner ads, push notifications, email blasts) are draining the user’s attention budget without earning it. The better move is to study the structure of the user’s day. When do they have mental bandwidth? When are they idle, primed, neurologically ready? Win the moment, not the megaphone.

The first is cue: does the person remember the desired behavior exists? Attention is a scarce resource, and most behavior change fails simply because the target action never enters conscious awareness.

The second dimension is reaction. Even when cued, does the person feel positive emotion toward the behavior? A gym-goer might remember their gym membership but feel dread at the thought of attendance. Emotion drives behavior more than rational calculation.

Third is evaluation. Can the person run a quick cost-benefit analysis and see the action as worthwhile right now? “Eating salad” might be good long-term, but “eating pizza” is better right-now, in this moment, in this context.

Fourth is ability. Does the person believe they can execute the behavior? Self-efficacy matters. If someone hasn’t run in five years, they might not believe they can run a 5K. Not because they’re physically incapable, but because they lack the self-trust.

Fifth is timing. When the moment arrives to act, is it positioned against procrastination triggers? A gym visit requires specific willpower and time allocation. If it competes with work stress or family obligations, it loses.

Sixth is experience. What’s the person’s history with this behavior type? Previous failures leave neural imprints. Someone who dieted and failed multiple times carries emotional weight into the next diet attempt that a first-timer doesn’t.

The Long-Term Change Problem

Here’s an uncomfortable truth: nobody has solved long-term behavior change past about six months. Wendel’s own framing in our conversations is bleaker than the academic version: “Once you get past six months, it all goes dark. We may be helping, but we just don’t know.” That’s twelve years of applied work at Morningstar and HelloWallet talking. Most interventions show declining compliance curves that drop steeply after the first few months, and the field still doesn’t have a reliable mechanism for what happens after that.

Three approaches exist for attacking long-term sustainability. The first is stable choice architecture. Keep the environment designed the same way indefinitely. If a 401(k) plan stays auto-enrolled, high participation stays. But this requires institutional continuity and gets expensive if you’re maintaining constant nudges.

The second approach is habit formation. The brain can wire repetition into automatic routines. But habits are brittle. Change the context, and the routine fractures. Go on vacation and miss your morning run routine. The habit doesn’t survive the disruption because it was never internalized.

The third approach is identity change. This is the holy grail of behavior change. When someone shifts from “I go running sometimes” to “I’m a runner,” the behavior survives vacations and context changes. Identity-based motivation persists because it’s connected to self-image. A runner reschedules missed runs. Someone who occasionally runs might not bother.

The Neuroscience Beneath the Strategy

Loretta Breuning’s work on neural pathways explains why automation is so effective. Early childhood wiring creates myelinated neural pathways that operate roughly 100 times more efficiently than new pathways. After puberty, attempting to rewire behaviors is like learning a foreign language. It’s possible, but cognitively expensive.

Automation works because it bypasses the brain entirely. You don’t need a new neural pathway if you never activate the choice in the first place. This is why “just willpower harder” fails so consistently. You’re asking someone to fight against decades of neural efficiency using fresh neural pathways that haven’t been myelinated yet. The game is rigged against willpower-based approaches.

This neuroscience also explains why simpler strategies succeed where complex ones fail. The brain prefers established pathways. Rules of thumb use existing neural hardware rather than requiring new architecture.

Breuning’s clearest applied example illustrates Strategy 2 in action. The standard intervention for smoking is to fill the smoker’s head with images of blackened lungs and early death. The intent is to attack the cigarette pathway. The actual neurological effect is the opposite: the fear images trigger a cortisol spike, and one of the fastest ways the brain knows to clear cortisol is to reach for a cigarette. The intervention reinforces the very pathway it tried to disrupt.

The Strategy 2 alternative builds a parallel pathway instead. “When I want a cigarette, I’ll give myself ten minutes of Netflix.” Same trigger, different action. The new pathway is shorter and pleasant, so the brain begins routing through it. Over weeks, the parallel becomes the default. You haven’t fought the old pathway. You’ve built a better road next to it.

The Design Hierarchy

Given these three strategies, which should you deploy first? My recommendation is to work down the hierarchy. Start with automation. If that’s not feasible, move to simplification. If that won’t work in your context, then invest in environmental design.

For long-term sustainability, the goal is always identity change, but that’s rare and requires deep cultural work. Focus first on making the initial behavior accessible, then worry about identity later. Too many programs fail at step one because designers are too focused on the aspirational vision of who people might become.

The PlayPump Cautionary Tale

Nothing teaches behavior change design better than failure analysis. PlayPump International raised over $60 million to install merry-go-rounds in rural African communities. The mechanism was clever: kids would play on the roundabout, and their motion would pump water. Build play into the system and solve the water crisis simultaneously.

The flaw: children don’t play all day. When the merry-go-round wasn’t in active use, there was no water pumping.

Communities eventually reverted to having adults manually push the roundabout, making it less efficient than the pumps it replaced. The designers had built an elegant solution to a problem they’d diagnosed incorrectly.

This illustrates a critical principle: testing matters more than techniques. The most beautifully designed intervention fails if it misunderstands the actual behavior change challenge. PlayPump failed not because automation, simplification, and environmental design are bad frameworks. It failed because the team didn’t validate whether their solution actually fit the behavioral gap they were trying to close.

Why Gym Memberships Expose Our Confusion

People value gym memberships to the point they pay every month. Yet they just don’t go. This contradiction reveals how disconnected we’ve become from actual behavior change design. We pay for the identity (“I’m someone who goes to the gym”), not the activity. We’ve constructed a system where people can purchase the feeling of being a gym-goer without having to actually work out.

This isn’t a character failure on the part of gym-goers. It’s a design failure on the part of the gym industry. If your model depends on people paying for something they don’t use, you’ve outsourced the behavior change problem to individual willpower. That’s not design. That’s just extracting money.

Better gyms might use Strategy 1: automatic enrollment in a group class at sign-up, with opt-out available. Strategy 2: mobile app training that requires no commute. Strategy 3: redesigning the gym layout so supportive community is the most visible element. Most gyms do none of these, which suggests selling memberships is more profitable than solving behavior change.

Badges, Leaderboards, and Actual Accomplishment

Within the Octalysis Framework, I’ve long examined how game mechanics intersect with real behavior change. One frequent mistake: designers award badges and points expecting them to sustain motivation. Just because you have a badge doesn’t mean people feel accomplished.

A badge tied to an external reward (showing status to others) can motivate short-term behavior. But it doesn’t create the internalized sense of competence that drives long-term change. This is why Core Drive 8: Loss & Avoidance relating to loss avoidance and identity-based motivation ultimately outperform shallow reward systems.

If you’re designing for behavior change, ask whether your game mechanics are creating genuine accomplishment or just simulating it through cosmetic systems. The distinction matters.

How Octalysis Maps the Three Strategies

Each of Wendel’s three strategies leverages a different cluster of Octalysis Core Drives, and once you see the mapping you understand why one strategy outperforms another in different contexts.

Strategy 1 (Automation) leverages Core Drive 8 (Loss & Avoidance) by sidestepping it. Auto-enrollment works because the default itself becomes the thing people would have to lose to act differently. The behavioral cost flips: opting out now feels like a loss, while staying enrolled feels like the safe default. You haven’t motivated anyone. You’ve reassigned the loss aversion that was already there.

Strategy 2 (Simplify) leverages Core Drive 2 (Development & Accomplishment) and Core Drive 3 (Empowerment of Creativity & Feedback). The reason bodyweight exercises beat gym memberships isn’t just lower friction — it’s tighter feedback loops. A push-up gives instant evidence of capability. A complex budgeting app drowns the same signal under setup screens. Wendel’s “pay twice the minimum” rule works for the same reason: every paycheck delivers a clean win.

Strategy 3 (Environment) leverages Core Drive 5 (Social Influence & Relatedness) and Core Drive 7 (Unpredictability & Curiosity). Choice architecture, nudges, and defaults work because they alter the social and contextual cues the user sees. Putting fruit at eye level changes what the user encounters first. Making giving the visible default at checkout shifts what feels normal. The intervention is environmental, but the lever is social — what other people seem to do, and what feels expected.

Identity-based long-term change activates Core Drive 1 (Epic Meaning & Calling) and Core Drive 5 (Social Influence & Relatedness). “I’m a runner” survives a vacation because it has joined the user’s sense of self and is reflected back by the running community. “I run on Tuesdays” doesn’t survive a Tuesday in Madrid. The Octalysis reading: behaviors that activate Core Drive 1 and Core Drive 5 outlast behaviors that depend only on Core Drive 2 and Core Drive 3.

Connecting to Broader Gamification Principles

Behavior change design doesn’t exist in isolation. It connects directly to how we structure motivation, feedback, and meaning. In my book Actionable Gamification: Beyond Points, Badges, and Leaderboards, I explore how the eight core drives shape human motivation across different contexts.

The fish on the beach framework intersects with Core Drive 6: Scarcity & Impatience and Core Drive 3: Empowerment of Creativity & Feedback. Making behavior change effortless through automation reduces the scarcity of willpower. Simplifying actions increases feedback loops that let people see progress. Changing environments activates development and accomplishment through visible results.

FAQ

Q: Doesn’t automating behavior change feel paternalistic or manipulative?

A: Consider this: if your goal is to help people achieve outcomes they want for themselves, then removing friction is generous, not manipulative. The person paying for a 401(k) wants to save for retirement. Automatic enrollment honors that desire. The objection to automation often says more about designers’ discomfort with their own power than about actual harm to users.

Q: Can these three strategies work together?

A: Yes, absolutely. You might auto-enroll someone in a simplified workout program that also uses environmental design to make exercising more social and visible. But they work together most effectively when the primary friction point is clear. People who aren’t acting because they forgot benefit most from automation. Those who remember but find the task too hard need simplification. Users who remember and can do it but lack motivation respond best to environmental changes. Diagnose first, then layer strategies.

Q: How long does identity-based motivation take to develop?

A: There’s no fixed timeline. Some people adopt a “runner” identity after a few months. Others take years. The speed depends on how the behavior integrates with social circles and whether feedback reinforces the identity. Community-based change (like running clubs) often succeeds where solitary change fails because the identity forms faster when reflected back to you socially.

Q: What should I do if none of the three strategies seem feasible?

A: That’s a signal to question whether behavior change is actually possible or desirable in your context. Not every behavior should be optimized for. Some resistance might indicate that the proposed behavior doesn’t actually align with someone’s values, despite what a program designer thinks they should want. Step back and validate whether the behavior change goal is real or imposed.

Q: How do I diagnose which CREATE dimension is the real bottleneck?

A: Test each one systematically. Remove friction on cue (add reminders) and see if behavior increases. Adjust environment to reduce negative emotion and test again. Offer skill-building to boost ability. Each test isolates which dimension actually matters. Most programs skip this diagnostic work and apply generic solutions. That’s why they fail.

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