Skills / Product-design

Hook Model

Habits are defined as behaviors done with little or no conscious thought. The question for the product designer is: does the intended behavior occur often enough and is it sufficiently rewarding to form a habit?

The Hook Model is Nir Eyal's prescriptive four-phase framework — Trigger, Action, Variable Reward, Investment — for engineering habit-forming products. Each completed cycle deepens the user's internal association with the product until their own emotional state becomes the cue to return, without any external prompt. Built for product managers, UX designers, and founders who want to understand why some products achieve unprompted habitual use while others depend on paid re-acquisition that decays over time.

By Nir Eyal  ·  Free

Use this skill See it in action

Specimen 01 · Live diagnosisHook Model

Input

“I'm building a language learning app. We have decent Day 1 retention but users fall off after week two. People open it when we send a push…”

Diagnosis

Let's build your Hook Canvas.

Full transcript ↓

Calibrated referenceagent-skills.ai

The gap

Design the four-phase loop that brings users back without a nudge

The Hook Model sequences four phases into a self-reinforcing cycle. Phase 1 (Trigger) distinguishes external prompts — notifications, emails, ads — from internal triggers, which are emotions like boredom, anxiety, or FOMO already encoded in the user's memory; durable habits must ultimately be anchored to internal triggers. Phase 2 (Action) applies BJ Fogg's Behavior Model (B = MAT: Motivation × Ability × Trigger) and Eyal's six elements of simplicity — time, money, physical effort, brain cycles, social deviance, and non-routine — to ruthlessly reduce friction in the minimum behavior required for reward. Phase 3 (Variable Reward) delivers three reward types — Tribe (social validation and connection), Hunt (resource and information acquisition), and Self (mastery and completion) — with unpredictability engineered into the payoff to sustain dopamine-driven anticipation. Phase 4 (Investment) has the user contribute something of value — data, content, followers, preferences — that accumulates stored value and loads the next trigger, leveraging the IKEA Effect and escalating commitment to increase switching costs. Each completed cycle makes the internal trigger association faster and more automatic. An ethical overlay — the Manipulation Matrix, a 2×2 assessing whether you use your own product and whether it genuinely improves users' lives — guides designers toward the Facilitator archetype.

The problem

Most products never achieve unprompted habitual use — users open the app only when re-engaged by paid ads or push notifications that are increasingly ignored. The core failure is not product quality but habit architecture: teams optimize for first-time activation and paid return without ever answering how a user's own internal emotional state could become the trigger. Without a structured framework, engagement mechanics are assembled from cargo-culted patterns — streaks, notifications, referral bonuses — without understanding why some combinations create genuine habits and others create engagement that evaporates the moment marketing spend pauses.

The solution

Map every phase of your product's habit loop — and identify the exact trigger, friction point, reward mechanic, or investment gap that is preventing unprompted return — producing a complete Hook Canvas you can take directly into your next design sprint.

You bring

A product, feature, or user flow you want to make habitually used
The core user problem or emotional pain point your product addresses
An existing or proposed onboarding or engagement sequence to audit
Willingness to honestly assess the ethical posture of your engagement mechanics

You get

A filled-in Hook Canvas mapping all four phases for your specific product
A prioritized list of friction-reduction interventions in your action step using Fogg's six simplicity elements
A menu of Tribe, Hunt, and Self variable reward options matched to your product context
A Manipulation Matrix placement with honest assessment of your product's ethical posture and actionable guidance

In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01  —  The same situation, two ways

Specimens A · B

Specimen AVague

Unaided

“Your language app sends three push notifications a day. 22% of users open from a notification; 3% open unprompted. After week two, push open rates drop to 9%. You're spending $4.20 CAC on users who only return when you interrupt them, and your week-four retention is 11%. You've tried more notifications, animated streaks, and a referral program. Nothing moves the unprompted return number.”

×Vague — nothing concrete to act on.

Specimen BCalibrated

With Hook Model

“You've mapped the Hook Canvas: embarrassment-after-comprehension-failure is your internal trigger, paired to your app at every session. A zero-decision 'continue' entry eliminates brain-cycle friction. A variable reward system surfaces mystery cultural insights and social comparisons alongside streak progress. A personal 'moments I got stuck' log accumulates stored value that auto-generates the next session prompt. Week-four retention reaches 34%. Push notifications are a supplement, not the engine. Users report opening the app moments after a confusing real-world encounter — the internal trigger is firing without any nudge from you.”

◦Specific, grounded, and ends with one clear next step.

Fig.02  —  A real session, unedited

In-app · assistant

Hook ModelClaude

Hook Model

Let's build your Hook Canvas. The Hook Model has four phases — Trigger, Action, Variable Reward, Investment — and most products that fail to achieve habitual use have a break in one specific phase. Tell me about the product or feature you want to analyze: what does it do, and what is the core user behavior you want to make habitual?

I'm building a language learning app. We have decent Day 1 retention but users fall off after week two. People open it when we send a push notification, but we're not getting unprompted use.

That pattern — strong push-driven return, weak unprompted return — is a Phase 1 problem. You've built external trigger dependency without anchoring to an internal trigger. The question is: what emotional state does your user experience in everyday life that your app could become the habitual response to? Duolingo's answer was anxiety about a streak — the fear of losing a 14-day run became the internal trigger. What does your user feel in the moments right before they might practice? Boredom on a commute? Embarrassment after failing to understand something in their target language? Anxiety about an upcoming trip or work meeting?

Probably embarrassment — a lot of our users are learning for travel or work situations where they've felt caught off guard. That feeling lingers.

That's a strong internal trigger — embarrassment is specific, recurring, and motivating without being manufactured. The design challenge is associating your app with that emotional cue so that the next time the user feels it, even without a notification, your app is what fires. Now let's move to Phase 2: Action. What is the minimum behavior a user performs to get their first reward in a session? Walk me through the first 30 seconds after opening.

They pick a lesson from a list, then there's a short intro screen, then they start a five-question exercise.

There are at least two friction points before the reward lands. Fogg's Behavior Model tells us ability gaps kill behavior even when motivation is high — and after week two, novelty motivation has dropped. 'Pick a lesson from a list' requires brain cycles: the user has to evaluate options and make a choice before they've received anything. The intro screen adds more delay. I'd recommend a default 'continue where you left off' single-tap entry — zero decision, zero delay, reward in under 60 seconds. What does the reward look like at the end of a five-question exercise?

You get a score and a 'nice work' message. We have streaks but they're not very prominent.

That's a fixed, predictable reward — and fixed rewards habituate quickly. Phase 3 needs variability. You have access to all three reward types: Tribe rewards such as 'how did your score compare to others who practiced this topic today' introduce social comparison variability; Hunt rewards such as occasionally surfacing a surprising regional expression or cultural insight inject discovery; Self rewards like streak visibility are good, but introduce unpredictability in what gets unlocked — a mystery idiom, a rare phrase, an audio clip from a native speaker. Variability keeps anticipation elevated. Now the Investment phase: what does a user put into your app that accumulates over time?

Just their progress and streak. We don't really have social features or user-generated content.

Progress and streak are a start — stored value in the Self dimension. But the critical failure in your Investment phase is that it isn't loading the next trigger. One high-leverage mechanic: let users log a real-world embarrassment moment ('I couldn't understand the waiter in Spanish') as a practice target. This stores a personal record and creates a specific external trigger for the next session — 'You flagged a situation like this. Ready to practice it?' The stored memory becomes the trigger. The loop becomes self-loading. Let me pull this into your Hook Canvas: Phase 1 internal trigger is embarrassment-after-comprehension-failure; Phase 2 friction reduction centers on zero-decision session entry; Phase 3 needs variable rewards across Tribe, Hunt, and Self; Phase 4 investment should accumulate personal failure moments that prime specific next-session triggers. Want me to write the full canvas document?

Reply to Hook Model…

What the skill is doing

01Hook Canvas Generation

02Internal Trigger Identification

03Action Friction Analysis

Fig.03  —  The same skill, in the channel you already use

WhatsApp · live

9:41

HMHook Modeltyping…

Messages are end-to-end encrypted. No one outside of this chat can read them.

Today

I'm building a language learning app. We have decent Day 1 retention but users fall off after week two. People open it when we send a push notificatio…9:14✓✓

The Hook Model has four phases — Trigger, Action, Variable Reward, Investment — and most products that fail to achieve habitual us…9:14

Tell me about the product or feature you want to analyze: what does it do, and what is the core user behavior you want to make hab…9:15

Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

◦Reads the situation, names the pattern, returns one concrete next move.

◦Delivered in seconds, inside a conversation that already exists.

Specimen · WhatsApp Business API · live

Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

Hook Canvas Generation

Walks you through all four Hook Model phases for a specific product or feature and produces a completed Hook Canvas — a structured artifact mapping your current or proposed trigger, minimum action, reward type and variability, and investment mechanic. The canvas exposes gaps where the loop breaks and surfaces the weakest phase for prioritized improvement.

◦ MethodBased on Eyal's Hook Canvas worksheet from 'Hooked', which sequences the four phases as a design review tool applicable to both greenfield products and existing features under audit.

CapabilityC-02

Internal Trigger Identification

Helps you move beyond external triggers (notifications, emails) to identify the internal emotional states — boredom, loneliness, uncertainty, FOMO — your product can become associated with. Surfaces techniques for pairing your product's cue with an existing internal pain so users reach for it without being prompted.

◦ MethodEyal argues that connecting internal triggers to a habit is 'the brass ring of habit-forming technology' — external triggers are a scaffold; the design goal is the moment when an emotion itself fires the behavior without any external prompt.

CapabilityC-03

Action Friction Analysis

Audits the minimum behavior your user must perform before receiving reward, applying Fogg's six simplicity elements — time, money, physical effort, brain cycles, social deviance, and non-routine — to identify which friction type is the binding constraint on your action step and recommends targeted reductions ranked by expected impact.

◦ MethodBased on Eyal's application of BJ Fogg's Behavior Model (B = MAT) in Phase 2 of the Hook, which establishes that ease of action is often more leverageable than motivation for increasing behavior frequency, particularly as post-novelty motivation declines.

CapabilityC-04

Variable Reward Design (Tribe / Hunt / Self)

Generates a set of reward options across all three reward types — Tribe (social validation, connection), Hunt (information or resource acquisition), and Self (mastery, completion, competency) — tailored to your product category, and advises on how to introduce variability so that anticipation, not just satisfaction, drives return.

◦ MethodEyal's three reward types map to distinct neurological reward systems; variable ratio schedules keep dopamine-driven craving elevated between cycles in ways that fixed, predictable rewards do not, sustaining the compulsion loop.

CapabilityC-05

Investment Mechanic Ideation

Surfaces concrete ways to have users invest value — content, preferences, social connections, history, reputation — that accumulates stored value over time and primes the next external trigger. Distinguishes high-leverage investment mechanisms (those that both increase switching cost and set up a specific next trigger) from low-leverage ones.

◦ MethodEyal ties the Investment phase to the IKEA Effect (we value what we help create) and escalating commitment; each investment both loads the next hook cycle and increases the perceived value of continuing versus switching to a competitor.

CapabilityC-06

Manipulation Matrix Ethical Assessment

Guides you through Eyal's 2×2 Manipulation Matrix — assessing whether you use the product yourself and whether you believe it materially improves users' lives — to place your product in one of four archetypes (Facilitator, Peddler, Entertainer, Dealer) and surfaces honest design questions about the engagement mechanics you are building.

◦ MethodEyal introduces the Manipulation Matrix as a mandatory ethical checkpoint for Hook Model practitioners, arguing designers should build only in the Facilitator quadrant where they are building something they believe genuinely serves the user's long-term interests.

Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces

OutputD-01

Hook Canvas

A structured four-phase document mapping your product's trigger (external and internal), minimum action, reward type and variability mechanism, and investment mechanic — with gap analysis indicating which phase is the binding constraint on habitual use.

OutputD-02

Friction Reduction Audit

A prioritized breakdown of Fogg's six simplicity elements (time, money, effort, brain cycles, social deviance, non-routine) scored for your specific action step, with concrete interventions ranked by expected impact on behavior frequency.

OutputD-03

Variable Reward Menu

A categorized list of Tribe, Hunt, and Self reward options appropriate for your product context, with notes on how to engineer variability into each and avoid reward satiation that flattens engagement over time.

OutputD-04

Manipulation Matrix Scorecard

A completed 2×2 placement of your product with written rationale, honest identification of any Peddler or Dealer signals in your current design, and actionable recommendations for moving toward and staying in the Facilitator quadrant.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

Nir Eyal

Nir Eyal is the author of 'Hooked: How to Build Habit-Forming Products' (2014) and 'Indistractable' (2019). He taught at Stanford Graduate School of Business and the Hasso Plattner Institute of Design (d.school), and has built and sold two technology companies. His writing on behavioral design has appeared in Harvard Business Review, Forbes, and TechCrunch, and 'Hooked' has become a foundational text widely assigned in product management courses and cited by product teams at leading technology companies.

Status · Inspired by Nir Eyal’s work — not yet claimed. Are you Nir Eyal?

Primary sourceS-01

Hooked: How to Build Habit-Forming Products

by Nir Eyal

Stanford GSB and d.school lecturer; two-time tech founder; author of 'Hooked' (2014) and 'Indistractable' (2019); HBR and Forbes contributor.

Read the original ↗

Citationnirandfar.com

In the build queue

Be first to run it.

Hook Model is being built right now. Leave your email and we’ll tell you the moment it goes live.

Notify meEmail

At launchI want to audit my product's habit loop. My retention problem is [describe it — e.g., 'good Day 1 but users fall off by week three and only return when we send notifications']. Can we build a Hook Canvas and find where the cycle is breaking?