Data Analytics Case Study

How Can a Wellness Company
Play It Smart?

Analyzing fitness smart‑device data to shape the marketing strategy of Bellabeat, a high‑tech wellness brand for women.

Google Data Analytics Capstone • Python · pandas · seaborn • 33 users · 2016 FitBit data
See the insights Download the case study

01 — Ask

The business problem

As a junior data analyst on Bellabeat's marketing team, the goal was to analyze how consumers use non‑Bellabeat smart devices and translate the trends into concrete marketing recommendations for a Bellabeat product.

  • What are the trends in smart device usage?
  • How do they apply to Bellabeat customers?
  • How can they influence marketing strategy?

“An analysis of available consumer data would reveal more opportunities for growth.”

— Urška Sršen, Bellabeat cofounder & CCO

02 — Prepare

The data & its limits

Source: FitBit Fitness Tracker Data (Kaggle, CC0 Public Domain) — minute‑level activity, sleep and heart‑rate logs from 33 users, Apr 12 – May 12, 2016.

33activity users
24sleep users
8weight users
~31days of data
Limitations (ROCCC): small non‑random sample · no demographics (sex, age) — critical for a women's brand · data from 2016 · single‑month window. These are carried transparently into the final recommendations.

Methodology

The 6‑phase data analysis process

1

Ask

Define the business task and stakeholders before touching the data.

2

Prepare

Source the data, assess credibility (ROCCC) and document its limits.

3

Process

Clean in Python: drop duplicates, parse dates, isolate non‑wear days.

4

Analyze

Summary stats, correlations, usage segmentation and time patterns.

5

Share

Six brand‑aligned visualizations that communicate the key findings.

6

Act

Translate the insights into three concrete marketing recommendations.

04 — Analyze · 05 — Share

Key insights

8,280avg steps / daybelow the 10k recommended
15.8hsedentary / day79% of tracked time
6.9hsleep / nightat the minimum (7–9h)
Scatter plot: more steps lead to more calories burned
More steps → more calories (r ≈ 0.57). A concrete motivational feedback to surface in the app.
Scatter plot: more sedentary hours lead to less sleep
More sedentary time → less sleep (r ≈ −0.60). A lever to connect daytime activity and rest quality.
Bar chart of average steps by hour of day
Activity peaks at lunch & evening (12–2pm, 5–7pm). Ideal windows for notifications and campaigns.
Bar chart of device usage frequency
Users are loyal — most wear the device 21–31 days. Engagement must be nurtured, not won from scratch.
Pie chart of how the average day is spent
The average day is mostly sedentary — a huge marketing lever to encourage movement.
Bar chart of average steps by day of week
Saturday is the most active day, Sunday the least — useful for weekly campaign timing.

06 — Act

Top 3 marketing recommendations

Chosen product: the Bellabeat app (with a focus on the Leaf / Time tracker).

1

Anti‑sedentary campaign

Position Bellabeat as the app that gets you moving: smart reminders in the peak windows (12–2pm, 5–7pm) and realistic goals such as +1,000 steps/day.

2

Sleep tracking as a flagship feature

Only 24/33 users track sleep, yet it links to sedentary time. Communicate “move by day, sleep better at night” — an ideal lever for premium membership.

3

Data‑driven marketing + gamification

Send content at peak‑usage times and gamify the steps→calories feedback with badges and streaks to boost retention and upsell.

Deliverables

Explore the full case study

The complete, reproducible analysis is available in multiple formats — in English and Italian.

📓 Full notebook Commented analysis with charts (HTML) 📄 Written report 6‑phase walkthrough (Word .docx) 📊 Presentation Executive deck (PowerPoint .pptx) 🐍 Jupyter notebook Source code (.ipynb)

Italian versions: notebook · report · slide · .ipynb