StreakMate 2.0: Turning My Smartwatch Into a Tiny Therapist

May 2025 | Indu Arimilli

The Problem: Motivation doesn’t vanish — it leaks

I’m obsessed with streaks. Not Snapchat streaks — the Duolingo kind. But as every student knows, the minute finals hit or sleep doesn’t, the streak dies.

My first attempt, StreakMate, helped me track habits — but it couldn’t read the room. It didn’t know when I was exhausted or when “study mode” had turned into “doom scroll mode.”

So for Summer 2025, I decided to build StreakMate 2.0 — the version that uses AI to detect motivation drops before they happen.

Because as any PM will tell you, reactive notifications are old news. Predictive ones are where the magic happens.

Research — Why do streaks fail?

I first user researched what was disguised as overthinking.

I asked 10 friends (and bribed them with coffee) to share screenshots of their broken Duolingo, workout, or Notion streaks. I mapped out when and why streaks died.Patterns emerged:

• Context switches: exams, travel, burnout.

• Emotional fatigue: the streak felt like pressure instead of pride.

• Feedback fatigue: the same old “Keep going!” notification stopped working.

A TechCrunch article on “emotionally adaptive interfaces” caught my eye — how wearables like Fitbit and Whoop are now context-aware, sensing fatigue through heart rate and micro-motion. I thought: what if I applied that to motivation?

MVP — StreakMate 2.0

Core Idea:

Your smartwatch notices your behavioral drift — subtle changes in motion, time spent idle, and even typing rhythm — and gives empathetic nudges before motivation collapses.

Features List:

1. Mood-motion analysis — AI detects micro-changes in activity that hint at low focus or burnout.

2. Pre-drop alerts — “Hey, streak break risk detected. Want to do a mini-version today?”

3. Streak-saving mode — auto-scales habits (10-min study instead of 1 hr).

4. Energy log — tracks your energy patterns alongside habit data.

5. Weekly insight dashboard — visualizes “energy vs. effort” trends.

User Stories:

• As a busy student, I want StreakMate to sense burnout early so I can adjust instead of crash.

• As someone who thrives on streaks, I want dynamic habit scaling so streaks feel sustainable, not stressful.

I then used this prompt within the Claude LLM:

“Create a smartwatch app simulation that tracks daily activity levels (mock accelerometer + heart rate data) and predicts motivation drop-off using a rolling average of ‘energy index’. Send friendly notifications when a downward trend is detected. Use a simple dashboard UI with energy trends and mini habit rescaling.”

Visual Concept:

Clicking the "save streak" button makes the user do a mini task to boost up their "energy" making them feel more productive and happier, kind of an incentive to keep moving!

Conclusion

Data empathy matters. Motivation isn’t binary — it’s gradient.

• Predictive design beats guilt design. Instead of “You failed,” it says “Hey, wanna adapt?”

• PM lens: Always design for the dip, not the peak.

My watch can’t fix burnout — but it can whisper: “You’ve done enough for today.”