As part of my capstone journey in data analytics, I worked on a case study with Bellabeat, a wellness technology company that creates smart products for women, such as activity trackers, a hydration bottle, and a mobile app. The challenge was simple but powerful:

How can Bellabeat use smart device data to better understand customer habits and shape their marketing strategy?

To answer this, I analyzed Fitbit fitness tracker data (available publicly on Kaggle) that included daily activity, steps, calories, and sleep records from real users.

What I Did

I followed the full analytics process:

1. Ask → Define the goal: uncover smart device usage trends that could guide Bellabeat’s marketing.
2. Prepare → Gathered Fitbit data from Kaggle (see dataset here).
3. Process → Cleaned the data, removed errors, created new fields like weekdays and hourly trends.
4. Analyze → Looked at daily activity, sleep patterns, calorie burn, and correlations.
5. Share → Visualized usage patterns (activity by weekday, hourly calorie burn trends).
6. Act → Translated insights into business recommendations for Bellabeat.

Key Insights

• Weekdays vs Weekends: Users are most active at the start of the week (Monday) and Saturdays, with a drop on Sundays.
• Peak Hours: Early evenings (5–7 PM) show the highest activity — right after work.
• Sleep Patterns: Many users don’t meet recommended sleep hours, suggesting opportunities for wellness content.
• Engagement: A positive (but moderate) correlation between active minutes and calories burned.

Recommendations for Bellabeat

1. Reinforce Good Habits – Push motivational content on high-activity days (e.g., Mondays and Saturdays).
2. Encourage Consistency – Use mid-week and Sunday challenges to boost lower activity days.
3. Personalize by Time – Share hydration and nutrition tips during work hours, fitness content in the evening, and relaxation/sleep advice at night.

Why It Matters

This project showed me how raw fitness data can be turned into actionable marketing strategies. For Bellabeat, it means they can better engage users with the right content at the right time — and for me, it was another opportunity to apply R, data cleaning, and visualization skills to a real-world scenario.

Full Analysis & R Code on Kaggle:
Check out my Kaggle Notebook here