TL;DR
Nir Eyal's Hooked Model describes a four-step cycle that turns infrequent users into habitual ones: Trigger, Action, Variable Reward, Investment. Products that complete this cycle repeatedly build intrinsic motivation that competitors find hard to displace.
What Is the Hooked Model?
In 2014, Nir Eyal published Hooked: How to Build Habit-Forming Products after spending years studying why certain consumer products achieved daily engagement while technically comparable products did not. His research covered Facebook, Twitter, Instagram, and Pinterest at their growth peaks. The Hooked Model was his synthesis.
The core observation: habits form through repeated cycles, not one-time experiences. A product earns a place in someone's daily routine by guiding them through a loop often enough that the behavior becomes automatic. Eyal called each pass through the loop a "hook." The goal is to fire enough hooks that users no longer need external prompting. The product itself becomes the trigger.
This framework is distinct from general UX or onboarding theory. It focuses specifically on the mechanics of behavioral repetition and the conditions that sustain it over time.
The Four Steps
Step 1: Trigger
Every habit begins with a cue. Triggers are the signals that initiate the behavior. Eyal divides them into two types.
External triggers are prompts from outside the user: a push notification, a badge count, an email, a friend's recommendation. Instagram's "Your friend just posted" notification is a classic external trigger. It requires no prior emotional state from the user. It simply presents itself.
Internal triggers are cues that come from within: emotions, thoughts, and situations. Boredom triggers a scroll to Instagram. Loneliness triggers a check of messages. Uncertainty triggers a search. The most powerful products attach themselves to pre-existing internal states. When a user feels bored, the product is already there.
The goal is to start with external triggers during early adoption and progressively shift users to internal triggers. External triggers are expensive to deliver at scale and easy for competitors to match. Internal triggers are automatic and nearly impossible to copy.
Step 2: Action
Once triggered, the user takes the simplest possible action in anticipation of a reward. Eyal draws on BJ Fogg's Behavior Model here: Behavior = Motivation x Ability x Trigger (B = MAT). All three must be present simultaneously.
In practice, this means your product must minimize friction. The action must be so easy that even low motivation can complete it. Instagram reduced the action to a scroll. TikTok eliminated even that choice with auto-play. Duolingo reduced the commitment to "tap one answer."
The design test: can a tired, distracted, mildly motivated user complete this action in under five seconds? If the answer is no, the Action step has too much friction. Reduce the steps, remove form fields, pre-populate defaults, and cut any confirmation dialogs that are not strictly necessary.
Motivation amplifies ability, but ability is more reliable. Users are not always motivated. They are almost always capable of a simple tap.
Step 3: Variable Reward
The action must be followed by a reward. But not a predictable reward. The slot machine effect is the mechanism here: variable, unpredictable payoffs create more sustained engagement than fixed payoffs. Neuroscience research shows that dopamine spikes are larger in anticipation of uncertain rewards than in receipt of certain ones.
Eyal identifies three types of Variable Rewards:
Rewards of the Tribe are social payoffs: likes, comments, shares, follower counts, leaderboard positions. Twitter's core loop delivered this type. The reward for posting was unpredictable: sometimes nobody noticed, sometimes the post went viral. That variability kept users posting.
Rewards of the Hunt are informational or material payoffs: a useful search result, a deal, new content in a feed, a gem drop in a game. Pinterest and Reddit operate primarily on this type. Each scroll might surface something valuable or nothing. Users keep scrolling to find out.
Rewards of the Self are intrinsic payoffs: completion, mastery, progress. Duolingo's XP system, streaks, and skill trees generate this type. The learner pursues the reward not for social recognition but for the internal satisfaction of advancement.
Strong habit-forming products often combine two or three types. The combination makes the reward system more durable because different users respond to different types, and the same user may respond to different types at different moments.
Step 4: Investment
The final step is what separates the Hooked Model from a simple reward loop. Investment is a small action the user takes that loads the next trigger and simultaneously increases the product's value to them.
When a user follows accounts on Instagram, uploads a photo, or adds a comment, they are increasing the product's relevance to their own life. The feed gets better as they invest. The content is more personalized. The next session is more rewarding.
Investment also creates switching costs. A user with 400 followers, five years of posts, and a curated following list has stored significant value in Instagram. Moving to a competing platform means abandoning that investment. This is one of the most powerful forces in consumer product retention.
The investment does not need to be large. A brief profile customization, a first follow, a single saved item. Small acts of personalization still increase product relevance and raise the cost of leaving.
Worked Example: Duolingo
Duolingo is one of the clearest modern applications of the Hooked Model. Each step is visible and deliberate.
Trigger. A morning push notification reads: "Your 47-day streak is at risk!" This is a classic external trigger, but it works because it connects to an internal state: loss aversion. The user already cares about the streak. The notification simply surfaces that concern at the right moment.
Action. The lesson interface opens immediately from the notification. One tap. The first question appears in under three seconds. The lesson itself is five minutes with no prior knowledge required. Fogg's ability variable is maximized. There is almost no barrier between the trigger and the behavior.
Variable Reward. Completing a lesson delivers a combination of all three reward types. XP points and gems (Rewards of the Hunt), a league leaderboard showing the user's rank against friends and strangers (Rewards of the Tribe), and a skill-progress bar moving toward a new crown or checkpoint (Rewards of the Self). Critically, how much XP a bonus chest contains is randomized. The chest animation creates anticipation before the reveal.
Investment. Every lesson adds to the streak counter, extends the vocabulary record, and builds the friend list through the league system. The user's progress history is stored in the product. A lapsed user considering re-engagement sees all accumulated progress waiting for them. That history is both motivation to return and a reason not to switch to a competitor.
Duolingo completes this cycle daily for active users. The habit loop fires automatically without conscious deliberation. That is the benchmark the model targets.
When to Use the Hooked Model
The Hooked Model is most applicable to products where repeat, unprompted engagement is the primary business driver. This includes social platforms, fitness apps, learning tools, casual games, news and content aggregators, and any consumer product where daily or weekly active use determines retention and monetization.
The framework is also useful as a diagnostic for existing products that have low engagement despite good activation. Walking through each step often reveals where the loop breaks. Weak variable rewards, high-friction actions, or no Investment phase are the most common failure points.
Use the Journey Mapper to map how your current user experience moves through these four stages, and the User Persona Builder to identify which internal triggers each user segment carries before they open your product.
When NOT to Use the Hooked Model
The Hooked Model is not appropriate for transactional products where the purchase is infrequent by design. Insurance, mattresses, mortgages, and enterprise software licenses have no need for daily habit formation. Applying Hooked logic here creates friction rather than removing it.
B2B SaaS tools that are used heavily but on a task-driven basis (invoice software, legal research tools, HR systems) may benefit from some elements, particularly reducing Action friction and adding Investment through data accumulation. But the full loop, especially Variable Rewards, often does not apply.
Applying Hooked to professional tools that users must engage with for work can also obscure real product problems. If a PM tool gets daily active use because managers are required to update it, that is not habit formation. It is obligation. The distinction matters for retention modeling.
Common Pitfalls
Over-reliance on external triggers. Teams send more notifications because push notifications are cheap to build. But notification fatigue degrades both the trigger effectiveness and the brand. Users who disable notifications have broken the loop entirely. The goal is to get to internal triggers, not to flood inboxes.
Predictable Variable Rewards. When users learn exactly what to expect from each action, the dopamine effect flattens. If every lesson in a learning app delivers exactly 10 XP with no variance, the reward phase stops sustaining engagement. Introduce genuine variability: bonus rounds, random gem drops, surprise content, or social events with uncertain outcomes.
Skipping the Investment phase. Many products nail Trigger, Action, and Reward but never ask the user to contribute anything back. Without Investment, the product remains replaceable. Any competitor with a marginally better Action step can displace it because the user has nothing stored. Build the Investment step early: first follow, first save, first custom setting, first piece of user-generated content.
Targeting the wrong internal trigger. Attaching a product to a negative internal state (anxiety, loneliness, boredom) can work in the short term but creates long-term ethical and brand problems. Eyal's subsequent book Indistractable addresses this directly. Products with durable engagement tend to connect to positive internal states: curiosity, aspiration, social belonging.
Hooked vs. Alternatives
Hooked vs. Octalysis (Yu-Kai Chou). The Octalysis framework maps eight core human drives and scores a product's design against each. It is broader and more granular than Hooked. Octalysis is useful for full gamification audits; Hooked is more actionable for product teams shipping iteratively.
Hooked vs. Behavioral Design (BJ Fogg). Fogg's model (B = MAT) covers the mechanics of any single behavior change. Hooked is specifically about repeating behavior over time through accumulating Investment. Fogg explains the first action; Hooked explains the hundredth.
Hooked vs. Jobs to Be Done. JTBD identifies what outcome a user is trying to achieve; Hooked describes how to build a product that delivers that outcome repeatedly without requiring the user to consciously decide to return. They are complementary. JTBD defines the internal trigger; Hooked describes the full loop that sustains the response.
Tools That Help
Building a habit-forming product requires understanding your users deeply before designing any loop. The User Persona Builder helps you identify the emotional states and situations that could serve as internal triggers for each segment. The Journey Mapper lets you trace the current path a user takes from first open to the Investment step, and surface where friction breaks the cycle.
Understanding activation rate is also essential when applying the Hooked Model. Activation is the prerequisite to habit formation. Users who never complete the first loop cannot develop a habit. Monitor activation alongside engagement frequency to diagnose whether low retention is a Trigger/Action problem or a post-activation problem.
Putting It Into Practice
Start by auditing your current product against each step. Write down: what triggers a user to open your product on day 7, day 30, day 90? If your only honest answer is "a push notification we sent," you have not yet achieved internal trigger attachment.
Then identify the simplest action a motivated user can take immediately after arriving. Count the taps from notification to first meaningful behavior. Every extra tap is attrition.
Map your current rewards. Are they variable or fixed? Do they serve the Tribe, the Hunt, or the Self? Do they serve more than one at once?
Finally, find where your Investment phase is. If you cannot identify a moment where the user contributes data, follows, saves, or customizes in a way that makes the next session more valuable, add one.
The Hooked Model does not guarantee retention. But it provides a precise vocabulary for diagnosing why users leave and a structured way to test interventions. That precision is what makes it useful for product teams beyond its initial appeal as a behavioral design concept.