Case Study Title:

Case Study 2: How Can a Wellness Technology Company Play It Smart?

Authored By:

Jayoda Jithmini Kulatunga

Date:

12/14/2024

A clear summary of the business task

Analyze smart device usage data to uncover consumer trends and provide actionable recommendations for enhancing the marketing strategy of Bellabeat's products, focusing on one specific product to maximize its appeal and market reach.

A description of all data sources used

1. FitBit Fitness Tracker Data

• Source: Kaggle
• Content: Physical activity, sleep monitoring, heart rate, and daily steps data for 30 Fitbit users.
• Licensing: CC0: Public Domain
• Limitations: Small sample size, unknown demographic diversity.

2. Complementary Data (Optional for addressing limitations):

• Additional datasets from open data repositories (e.g., WHO, national health surveys) can provide broader demographic insights for comparison and validation.

Documentation of the cleaning and manipulation of data

Data Cleaning and Manipulation Documentation

Tools Used

• R Programming Language

• Libraries: tidyverse for data manipulation and visualization.

Steps Taken to Clean and Prepare the Data

1. Data Import

• Loaded daily_activity and sleep_day datasets using read.csv().

2. Initial Exploration

• Explored datasets to understand structure and contents:

• Used head() to preview the first few rows.
• Listed column names with colnames().
• Verified data types using str().

3. Checking for Missing Data

• Verified the presence of missing values in both datasets using sum(is.na()).

• Result: No missing values found.

4. Removing Duplicates

• Checked for duplicate rows using distinct().
• Removed duplicate rows from both datasets:

• daily_activity <- daily_activity %>% distinct()
• sleep_day <- sleep_day %>% distinct()

5. Ensuring Consistent Date Formatting

• Reformatted dates to ensure consistency:

• daily_activity$ActivityDate <- as.Date(daily_activity$ActivityDate, format = "%m/%d/%Y")
• sleep_day$ActivityDate <- as.Date(sleep_day$SleepDay, format = "%m/%d/%Y")

6. Alignment of Column Data Types

• Verified column data types and ensured alignment using str().

7. Filtering Invalid or Extreme Values

• Removed rows with invalid or extreme outlier values:

• Ensured all values in TotalSteps, TotalMinutesAsleep, and SedentaryMinutes were ≥ 0.
• Used filter() to retain only valid data.

8. Merging Datasets

• Merged daily_activity and sleep_day datasets on Id and ActivityDate using merge():

• combined_data <- merge(sleep_day, daily_activity, by="Id")

9. Removing Redundant Columns

• Dropped unnecessary columns, such as SleepDay, after merging datasets:

• combined_data <- combined_data %>% select(-c(SleepDay))

10. Derived Metrics

• Created new metrics to enhance analysis:

• Sleep Efficiency: SleepEfficiency = TotalMinutesAsleep / TotalTimeInBed
• Activity-to-Sedentary Ratio: ActivityToSedentaryRatio = TotalSteps / SedentaryMinutes

11. Summarization by User

• Aggregated data to calculate average values per user:

• user_summary <- combined_data %>% group_by(Id) %>% summarise(AvgSteps = mean(TotalSteps, na.rm = TRUE), AvgSleepMinutes = mean(TotalMinutesAsleep, na.rm = TRUE), AvgSedentaryMinutes = mean(SedentaryMinutes, na.rm = TRUE), AvgSleepEfficiency = mean(SleepEfficiency, na.rm = TRUE))

12. Verification

• Verified cleaned data using summary statistics:

• summary(combined_data) provided min, max, mean, and median values for key metrics.

• Visualized relationships to confirm data consistency:

• Scatterplots: TotalSteps vs. SedentaryMinutes, TotalMinutesAsleep vs. TotalTimeInBed.

13. Saving Cleaned Data

• Exported cleaned datasets for further analysis:

• Combined data: write.csv(combined_data, "cleaned_combined_data.csv", row.names = FALSE)
• User summary: write.csv(user_summary, "user_summary.csv", row.names = FALSE)

Summary

The cleaning and manipulation process ensured the dataset’s integrity and consistency, enabling accurate analysis of consumer trends. Key steps included handling duplicates, aligning data formats, creating derived metrics, and verifying the results through visual and statistical checks. The cleaned and transformed data is now ready for insightful analysis to support Bellabeat’s marketing strategy.

A summary of the analysis

click this link

• The analysis revealed that many users exhibit sedentary behavior and experience suboptimal sleep efficiency, highlighting areas where Bellabeat’s products can provide value.
• These findings will inform marketing strategies tailored to user segments and product improvements to address user challenges.

Supporting visualizations and key findings

Top high-level content recommendations based on The analysis

1. Target Sedentary Users:

• Insight: Many users log high sedentary minutes with low daily steps.
  Recommendation: Position Bellabeat products as tools to motivate movement and reduce inactivity.

2. Enhance Sleep Monitoring Features:

• Insight: Some users spend significant time in bed but achieve suboptimal sleep efficiency.
  Recommendation: Introduce advanced analytics and guided relaxation techniques in the app to improve sleep.

3. Segmented Marketing Campaigns:

• Insight: Different user segments (active vs. sedentary, good sleepers vs. poor sleepers) have varying needs.
  Recommendation: Use segmentation to tailor marketing efforts and personalized product suggestions.

4. Retention Through Wellness Tracking:

• Insight: Active and balanced users are more likely to stay engaged with the app and products.
  Recommendation: Develop loyalty programs or premium features that reward consistent usage and engagement.