Key Questions:

1. Exercise Minutes: How much exercise do Fitbit users typically log per week?
   This can be compared to the recommended 150 minutes of moderate activity for maintaining a healthy lifestyle.

2. Steps and Activity Levels: What is the average number of daily steps, categorized by activity levels?
   Understanding these patterns can reveal user preferences and highlight opportunities for wearables and coaching programs.

3. Activity Patterns: When are users most active throughout the day and the week?
   These insights can help identify peak activity periods and encourage more consistent physical activity.

   Analysis:

   Tools: R packages (tidyverse, scale)
   Data Source: The dataset is publicly available on Kaggle (Fitbit Fitness Tracker Data) and is classified as 100% complete, credible, and compatible.

   Sample Overview:
   The dataset contains activity data from 33 Fitbit users over one month. However, several limitations should be considered:

   • Small Sample Size: The dataset includes only 33 users, which limits the generalizability of insights.
   • Missing Contextual Information: There is no data on participants’ gender, occupation, or location.
   • Incomplete Tracking: Some users did not log their activity daily, leading to potential biases and missing data for certain days.

Low Activity Levels:

• Finding: A significant portion (45%) of users exhibit low activity levels.
• Insight: This highlights an opportunity for Bellabeat to target women who struggle to meet activity recommendations.
• Action: Develop marketing messages and products that cater to this segment, focusing on low-impact activities like brisk walking.

Compare the average to the recommended 150 minutes.

Highly active: Exercise minutes > = 295 min;
Mid Active: Exercise minutes 295min-150min;
Not Active: Exercise minutes  <150 min;

Steps Below Recommendation:

• Finding: The average user doesn’t reach the recommended 10,000 daily steps.
• Insight: This suggests a need for products or features that encourage users to be more active.
• Action: Highlight features in the app or wearables that promote increased activity, such as step challenges or gamification elements.

Daily Activity Patterns:

• Finding: Step counts rise for all groups after 6 pm. The more active groups have higher average steps throughout the day.
• Insight: This suggests potential for evening activity promotion and highlights the importance of encouraging consistent activity throughout the day for all users.
• Action: Consider offering evening activity suggestions or reminders within the app. Develop strategies within the app to motivate users to maintain activity levels throughout the day.

Weekend Activity Patterns:

• Finding: All activity groups see a decrease in steps on Sundays and higher steps on Saturdays (particularly in the low activity group).
• Insight: This suggests weekends offer more time for physical activity, especially for those with lower baseline activity during the week.
• Action: Develop targeted weekend activity challenges or content to capitalize on this increased leisure time.

https://www.kaggle.com/code/erika22/capstone-case-study-fitness-data-analytics