Variable Reward Design (Tribe, Hunt, Self)
Three reward types that create craving through unpredictable reinforcement
Variable Reward Design is a framework for creating the craving that keeps users coming back to a product. Based on B.F. Skinner's research showing that intermittent reinforcement dramatically increases behavior frequency, the framework categorizes all variable rewards into three types: Rewards of the Tribe (social validation and connectedness), Rewards of the Hunt (pursuit of material resources and information), and Rewards of the Self (intrinsic motivation for mastery, competence, and completion).
The neuroscience behind the framework is compelling. Research by Stanford's Brian Knutson showed that the nucleus accumbens activates not when a reward is received but in anticipation of it. Variability multiplies this effect, creating a focused state that suppresses judgment while activating wanting and desire. This is why slot machines, social media feeds, and email are so compelling: the unpredictability of what comes next drives compulsive engagement.
Critically, the framework also identifies two failure modes. First, finite variability (experiences that become predictable after repeated use) leads to declining engagement, as Zynga's FarmVille clones demonstrated. Second, rewards must match the user's actual motivation: Mahalo failed by offering monetary rewards when users were motivated by social recognition. The framework emphasizes that variable rewards are not fairy dust to sprinkle on any product; they must align with genuine user needs.
- The brain's dopamine system responds most powerfully not to rewards themselves but to the anticipation of uncertain rewards.
- Three types of variable rewards map to fundamental human drives: social belonging (Tribe), resource acquisition (Hunt), and mastery (Self).
- Finite variability leads to declining engagement; infinite variability (often through user-generated content or social interactions) sustains interest over time.
- Variable rewards must match users' actual motivations; mismatched rewards (like paying people for social behavior) backfire.
- User autonomy must be preserved; threatening people's freedom to choose triggers reactance and rebellion against the product.
- Identify Your Users' Core MotivationBefore designing rewards, understand what truly drives your users. Interview at least five customers in open-ended sessions to discover what they find enjoyable, surprising, or satisfying about your product. Look for emotional drivers, not feature requests.Pro tipQuora succeeded where Mahalo failed because Quora understood that Q&A users are motivated by social recognition (upvotes from peers), not monetary rewards.WarningDo not assume you know what motivates users. Mahalo's founders assumed money would drive engagement; it did not.
- Design Rewards of the TribeCreate mechanisms for social validation, connection, and status. This includes likes, comments, upvotes, follower counts, badges that confer community status, and features that let users see how peers respond to their contributions. Leverage Bandura's social learning theory by showing users rewarded behavior from people like them.Pro tipLeague of Legends used Honor Points (peer-awarded kudos for sportsmanlike conduct) to combat toxic behavior while creating powerful tribe rewards that users coveted.
- Design Rewards of the HuntCreate the sensation of pursuing and finding valuable resources or information. This includes content feeds with unpredictable mixes of relevant and irrelevant items, search results, deals and offers, and any mechanism where users scroll, search, or browse in pursuit of something valuable.Pro tipPinterest uses partially cut-off images below the fold to create curiosity that drives scrolling. The infinite scroll ensures users never hit a natural stopping point.
- Design Rewards of the SelfCreate experiences of mastery, completion, and competence. This includes leveling up, skill progression, inbox zero, streak tracking, and any mechanism that lets users see their own growth or achieve a sense of completion. Codecademy turns the difficult process of learning to code into a gamelike progression system.Pro tipMailbox achieves a feeling of inbox zero through clever folder sorting, giving users the reward of completion even though emails are simply hidden and reappear later.
- Layer Multiple Reward TypesThe most habit-forming products combine all three reward types. Map your existing rewards to the three categories and identify gaps. Then brainstorm ways to add the missing types without cluttering the experience.Pro tipEmail uses all three: social obligation and connection (Tribe), career-relevant information (Hunt), and the satisfaction of an empty inbox (Self). This triple reward structure explains its extreme stickiness.WarningDo not force-fit rewards that don't align with the product's core purpose. Rewards should feel natural, not grafted on.
- Ensure Infinite VariabilityEvaluate whether your rewards exhibit finite or infinite variability. Content consumption (watching a TV show) offers finite variability. Content creation and social interaction offer infinite variability. Design systems where users themselves create the unpredictability that sustains engagement.Pro tipWorld of Warcraft maintains engagement for years because players interact with other humans whose behavior is inherently unpredictable. FarmVille, played in solitude, lost its appeal as patterns became predictable.WarningProducts with finite variability must constantly produce new content to keep pace, operating under the costly 'studio model' like Hollywood.
Stack Overflow generates 5,000 detailed technical answers per day from volunteer contributors. Users write time-consuming responses in anticipation of upvotes and badges that confer community status. The variable element (how many upvotes a response receives) keeps contributors engaged. Points represent genuine standing in the developer community, not empty gamification.
MyFitnessPal required rigid calorie tracking that felt like an obligation. Fitocracy used social engagement as the primary hook, sending kudos from community members and enabling advice-sharing. Instead of imposing new behaviors, Fitocracy leveraged existing social behaviors (gym talk among friends) and made them easier and more rewarding digitally.
The concept of variable reinforcement traces back to B.F. Skinner's 1950s pigeon experiments, where he discovered that randomizing food pellet delivery dramatically increased lever-pressing behavior compared to predictable delivery. Eyal synthesized this foundational research with modern neuroscience (Knutson's dopamine studies), evolutionary biology (persistence hunting theory), and self-determination theory (Deci and Ryan's work on intrinsic motivation) to create a practical taxonomy of reward types.
The three categories (Tribe, Hunt, Self) emerged from Eyal's analysis of hundreds of habit-forming products and map to deep evolutionary drives: social belonging, resource acquisition, and mastery of environment.