CraveSmart
CraveSmart
Building habit-forming healthier options
Building habit-forming healthier options
Problem Statement
How might we create a habit-forming experience that helps people choose healthier food during moments of craving and decision fatigue?
Problem Statement
How might we create a habit-forming experience that helps people choose healthier food during moments of craving and decision fatigue?
While using apps like Blinkit, Swiggy, and Zomato, I observed a destructive pattern: people doomscroll through thousands of options, become overwhelmed, and default to unhealthy choices repeatedly.
Key insight: These apps have already created a strong habit. My challenge was to hijack this habit loop and redirect it toward healthier outcomes.
While using apps like Blinkit, Swiggy, and Zomato, I observed a destructive pattern: people doomscroll through thousands of options, become overwhelmed, and default to unhealthy choices repeatedly.
Key insight: These apps have already created a strong habit. My challenge was to hijack this habit loop and redirect it toward healthier outcomes.
Research Approach
Disclaimer: This case study is based on self-observation I acted as the primary user persona, identifying patterns in my own behavior and conversations with others.
The unhealthy habit loop I observed:
Trigger: Late-night hunger or stress
Action: Open food delivery app
Variable Reward: Endless doom scrolling, anticipation
Investment: Order unhealthy food
Outcome: Temporary satisfaction, lasting guilt
This loop repeats nightly, strengthening the unhealthy habit.
Research Approach
Disclaimer: This case study is based on self-observation I acted as the primary user persona, identifying patterns in my own behavior and conversations with others.
The unhealthy habit loop I observed:
Trigger: Late-night hunger or stress
Action: Open food delivery app
Variable Reward: Endless doom scrolling, anticipation
Investment: Order unhealthy food
Outcome: Temporary satisfaction, lasting guilt
This loop repeats nightly, strengthening the unhealthy habit.
Research Approach
Disclaimer: This case study is based on self-observation I acted as the primary user persona, identifying patterns in my own behavior and conversations with others.
The unhealthy habit loop I observed:
Trigger: Late-night hunger or stress
Action: Open food delivery app
Variable Reward: Endless doom scrolling, anticipation
Investment: Order unhealthy food
Outcome: Temporary satisfaction, lasting guilt
This loop repeats nightly, strengthening the unhealthy habit.
Phase 1: Trigger
I designed multiple trigger points to intercept users at moments of vulnerability. External triggers include time-based push notifications that appear when cravings typically hit, such as 4 PM for afternoon snacks or 11 PM for late-night cravings. A smart home widget displays predicted cravings for the current time with one-tap access to healthy options, reducing friction to near-zero.
Beyond external triggers, I designed for internal emotional triggers. The app becomes associated with specific emotional states through empathetic, non-judgmental language. When users feel bored, stressed, or guilty, the app responds with understanding rather than lectures, making it feel like a supportive companion rather than a demanding taskmaster.
Phase 1: Trigger
I designed multiple trigger points to intercept users at moments of vulnerability. External triggers include time-based push notifications that appear when cravings typically hit, such as 4 PM for afternoon snacks or 11 PM for late-night cravings. A smart home widget displays predicted cravings for the current time with one-tap access to healthy options, reducing friction to near-zero.
Beyond external triggers, I designed for internal emotional triggers. The app becomes associated with specific emotional states through empathetic, non-judgmental language. When users feel bored, stressed, or guilty, the app responds with understanding rather than lectures, making it feel like a supportive companion rather than a demanding taskmaster.
Phase 2: Action
Following the Fogg Behavior Model, I maximized both motivation and ability to make the core behavior effortless. The entire flow takes 30 seconds and just three taps: open the app or tap the widget, see the predicted craving highlighted, tap the craving category, view the top three options instantly, and tap to order.
The key design decision was craving-based navigation. Instead of asking "What do you want to eat?" which requires cognitive effort, I ask "What are you craving?" with visual cards. This shifts the mental load from recall to recognition, making the choice feel instinctive. I limited results to the top three options, eliminating the paradox of choice. The AI handles the cognitive work of filtering and ranking, while users simply pick what appeals to them.
Phase 2: Action
Following the Fogg Behavior Model, I maximized both motivation and ability to make the core behavior effortless. The entire flow takes 30 seconds and just three taps: open the app or tap the widget, see the predicted craving highlighted, tap the craving category, view the top three options instantly, and tap to order.
The key design decision was craving-based navigation. Instead of asking "What do you want to eat?" which requires cognitive effort, I ask "What are you craving?" with visual cards. This shifts the mental load from recall to recognition, making the choice feel instinctive. I limited results to the top three options, eliminating the paradox of choice. The AI handles the cognitive work of filtering and ranking, while users simply pick what appeals to them.
Phase 3: Variable Reward
I designed for all three types of variable rewards to keep users engaged. Social rewards include streaks, achievements, leaderboards, and community challenges. Seeing "7 days healthy" with a fire emoji or knowing you're in the top 10% of healthy eaters creates powerful social validation.
Material rewards follow an unpredictable pattern. Sometimes users get a 10% discount, sometimes 30%, sometimes free delivery. The mystery health score adds anticipation: "Calculating... 87/100!" Surprise upgrades, like unlocking premium features for a week, create delight through unpredictability.
Personal rewards focus on mastery and progress. Daily macro rings fill up as users eat, goal achievements celebrate milestones, and before-and-after insights show tangible improvement over time. These rewards tap into intrinsic motivation, making users feel they're becoming better versions of themselves.
The first-time experience delivers an immediate win. When new users open the app and tap "Sweet," they discover three healthy options they didn't know existed. After ordering the Greek Yogurt Bowl, they receive instant feedback: "This has 2x the protein of ice cream and 60% less sugar. Great choice!" This aha moment cements the value proposition immediately.
The "Aha!" Moment
First-time experience:
"What are you craving right now?"
Taps "Sweet"
Sees 3 options they didn't know existed
Orders Greek Yogurt Bowl
Instant feedback: "2x the protein of ice cream, 60% less sugar. Great choice! 🎉"
Phase 3:
Variable Reward
I designed for all three types of variable rewards to keep users engaged. Social rewards include streaks, achievements, leaderboards, and community challenges. Seeing "7 days healthy" with a fire emoji or knowing you're in the top 10% of healthy eaters creates powerful social validation.
Material rewards follow an unpredictable pattern. Sometimes users get a 10% discount, sometimes 30%, sometimes free delivery. The mystery health score adds anticipation: "Calculating... 87/100!" Surprise upgrades, like unlocking premium features for a week, create delight through unpredictability.
Personal rewards focus on mastery and progress. Daily macro rings fill up as users eat, goal achievements celebrate milestones, and before-and-after insights show tangible improvement over time. These rewards tap into intrinsic motivation, making users feel they're becoming better versions of themselves.
The first-time experience delivers an immediate win. When new users open the app and tap "Sweet," they discover three healthy options they didn't know existed. After ordering the Greek Yogurt Bowl, they receive instant feedback: "This has 2x the protein of ice cream and 60% less sugar. Great choice!" This aha moment cements the value proposition immediately.
The "Aha!" Moment
First-time experience:
"What are you craving right now?"
Taps "Sweet"
Sees 3 options they didn't know existed
Orders Greek Yogurt Bowl
Instant feedback: "2x the protein of ice cream, 60% less sugar. Great choice! 🎉"
Phase 4: Investment
Every interaction increases the app's value to the user while raising the cost of switching. Through preference learning, users rate dishes, set dietary restrictions, and mark favorite cuisines, making AI recommendations progressively more accurate. Goal setting creates personalized meal plans based on health objectives, eating frequency, and calorie targets.
Social investment builds network effects. Inviting friends enables better challenges, following other users provides inspiration, and sharing achievements creates public commitment devices. Data accumulation happens automatically—every order is tracked, patterns are learned, and the AI becomes smarter over time.
After 30 days, users have built something irreplaceable. The AI knows their preferences intimately, 30 days of nutrition data provides unique insights, streaks create loss aversion, and social connections feel valuable. The switching cost becomes prohibitively high because users would lose all this accumulated value.
The Stored Value Loop
After 30 days:
AI knows preferences intimately
30 days of irreplaceable data
Streaks create loss aversion
Social connections built
High switching cost
Why: Investment loads the next trigger. Users return because they've built something valuable.
Phase 4: Investment
Every interaction increases the app's value to the user while raising the cost of switching. Through preference learning, users rate dishes, set dietary restrictions, and mark favorite cuisines, making AI recommendations progressively more accurate. Goal setting creates personalized meal plans based on health objectives, eating frequency, and calorie targets.
Social investment builds network effects. Inviting friends enables better challenges, following other users provides inspiration, and sharing achievements creates public commitment devices. Data accumulation happens automatically—every order is tracked, patterns are learned, and the AI becomes smarter over time.
After 30 days, users have built something irreplaceable. The AI knows their preferences intimately, 30 days of nutrition data provides unique insights, streaks create loss aversion, and social connections feel valuable. The switching cost becomes prohibitively high because users would lose all this accumulated value.
The Stored Value Loop
After 30 days:
AI knows preferences intimately
30 days of irreplaceable data
Streaks create loss aversion
Social connections built
High switching cost
Why: Investment loads the next trigger. Users return because they've built something valuable.
"Habits aren't built by adding features—they're built by removing friction, creating unpredictability, and making each use more valuable than the last. Design for Day 365, not Day 1."
"Habits aren't built by adding features—they're built by removing friction, creating unpredictability, and making each use more valuable than the last. Design for Day 365, not Day 1."