Bellabeat Fitness Tracker Case Study

Google Data Analytics Professional Certificate Capstone Project

Overview

Analyzed public Fitbit Fitness Tracker Data to uncover usage trends in activity, sleep, sedentary behavior, and calories. Provided marketing recommendations for Bellabeat (women-focused wellness tech company).

Business Task: Identify smart device trends and apply insights to Bellabeat products (Leaf, Time, app) to guide marketing strategy.

Tools & Tech Stack

• SQL: PostgreSQL/pgAdmin — data import, cleaning, aggregation (minute_sleep → daily sleep), merging
• Python: Jupyter Notebook (pandas, seaborn, matplotlib) — EDA, stats, visualizations
• Tableau Public: Interactive dashboard for sharing insights
• Data Source: Fitbit Fitness Tracker Data (Kaggle, CC0 Public Domain)

Key Findings

• Average daily steps: ~6,547 (below 10k goal)
• Average sleep: 6.56 hours (short on many days)
• Higher activity & calorie burn on days with sleep tracked
• Mid-week (esp. Wednesday) shows best activity + sleep balance
• High sedentary time (~16–17 hrs/day) across most days

Recommendations for Bellabeat

1. Promote overnight wear of Leaf/Time → better sleep tracking unlocks activity insights
2. Personalized notifications for low-activity days → target patterns like Tuesday dips
3. Focus marketing on holistic wellness → emphasize sleep-activity link for women

Project Files

• bellabeat_analysis.ipynb: Python analysis & plots
• SQL.sql: SQL queries for data prep & merging
• Tablue_Dashboard 1.png: Screenshot of Tableau dashboard
• data/: Raw & processed CSVs

Tableau Dashboard

Live version: https://public.tableau.com/app/profile/desalegn.tilahun/viz/Tablue_Google_dataanalytics/Dashboard1?publish=yes

How to Run

1. Clone repo
2. Open bellabeat_analysis.ipynb in Jupyter
3. Ensure data files are in place
4. Run cells sequentially

License: MIT