Case Study: How Can a Wellness Technology Company Play it Smart?
Diane Howard
2024-07-24
Introduction
The business task is to analyze smart device usage data in order to gain insight into how consumers use non-Bellabeat smart devices. Then select one Bellabeat product to apply these insights to in your presentation.
Public data from Kaggle.com was used to explore the daily habits smart device users. This Fitbit Fitness Tracker Data (CC0: Public Domain, dataset made available through Mobius) data set contains personal fitness tracker from thirty fitbit users. Thirty eligible Fitbit users consented to the submission of personal tracker data, including minute-level output for physical activity, heart rate, and sleep monitoring. It includes information about daily activity, steps, and heart rate that can be used to explore users’ habits.
I chose to focus on the daily activity, steps and calories. The data was collected from 30+ users and over a period of 2 months. It should be noted that the results may be biased. Also, since the weather and/or season could affect bias and credibility.
- Analysis of Fitbit activity data
- Data from 3.12.16 to 5.12.16
- Goals: Clean, analyze, and visualize data to identify trends
Data Preparation
Load Datasets
daily_activity <- c("dailyActivity_merged.csv", "dailyActivity_merged2.csv")
fitbit_activity <- map(daily_activity, read_csv)## Rows: 457 Columns: 15
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): ActivityDate
## dbl (14): Id, TotalSteps, TotalDistance, TrackerDistance, LoggedActivitiesDi...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.
## Rows: 940 Columns: 15
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (1): ActivityDate
## dbl (14): Id, TotalSteps, TotalDistance, TrackerDistance, LoggedActivitiesDi...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.all_activity <- bind_rows(fitbit_activity)Data Cleaning
## Rows: 1,397
## Columns: 15
## $ Id <dbl> 1503960366, 1503960366, 1503960366, 150396036…
## $ ActivityDate <chr> "3/25/2016", "3/26/2016", "3/27/2016", "3/28/…
## $ TotalSteps <dbl> 11004, 17609, 12736, 13231, 12041, 10970, 122…
## $ TotalDistance <dbl> 7.11, 11.55, 8.53, 8.93, 7.85, 7.16, 7.86, 7.…
## $ TrackerDistance <dbl> 7.11, 11.55, 8.53, 8.93, 7.85, 7.16, 7.86, 7.…
## $ LoggedActivitiesDistance <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, …
## $ VeryActiveDistance <dbl> 2.57, 6.92, 4.66, 3.19, 2.16, 2.36, 2.29, 3.3…
## $ ModeratelyActiveDistance <dbl> 0.46, 0.73, 0.16, 0.79, 1.09, 0.51, 0.49, 0.8…
## $ LightActiveDistance <dbl> 4.07, 3.91, 3.71, 4.95, 4.61, 4.29, 5.04, 3.6…
## $ SedentaryActiveDistance <dbl> 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.0…
## $ VeryActiveMinutes <dbl> 33, 89, 56, 39, 28, 30, 33, 47, 40, 15, 43, 3…
## $ FairlyActiveMinutes <dbl> 12, 17, 5, 20, 28, 13, 12, 21, 11, 30, 18, 18…
## $ LightlyActiveMinutes <dbl> 205, 274, 268, 224, 243, 223, 239, 200, 244, …
## $ SedentaryMinutes <dbl> 804, 588, 605, 1080, 763, 1174, 820, 866, 636…
## $ Calories <dbl> 1819, 2154, 1944, 1932, 1886, 1820, 1889, 186…## Id ActivityDate TotalSteps TotalDistance
## Min. :1.504e+09 Length:1397 Min. : 0 Min. : 0.000
## 1st Qu.:2.320e+09 Class :character 1st Qu.: 3146 1st Qu.: 2.170
## Median :4.445e+09 Mode :character Median : 6999 Median : 4.950
## Mean :4.781e+09 Mean : 7281 Mean : 5.219
## 3rd Qu.:6.962e+09 3rd Qu.:10544 3rd Qu.: 7.500
## Max. :8.878e+09 Max. :36019 Max. :28.030
## TrackerDistance LoggedActivitiesDistance VeryActiveDistance
## Min. : 0.000 Min. :0.0000 Min. : 0.000
## 1st Qu.: 2.160 1st Qu.:0.0000 1st Qu.: 0.000
## Median : 4.950 Median :0.0000 Median : 0.100
## Mean : 5.192 Mean :0.1315 Mean : 1.397
## 3rd Qu.: 7.480 3rd Qu.:0.0000 3rd Qu.: 1.830
## Max. :28.030 Max. :6.7271 Max. :21.920
## ModeratelyActiveDistance LightActiveDistance SedentaryActiveDistance
## Min. :0.0000 Min. : 0.000 Min. :0.000000
## 1st Qu.:0.0000 1st Qu.: 1.610 1st Qu.:0.000000
## Median :0.2000 Median : 3.240 Median :0.000000
## Mean :0.5385 Mean : 3.193 Mean :0.001704
## 3rd Qu.:0.7700 3rd Qu.: 4.690 3rd Qu.:0.000000
## Max. :6.4800 Max. :12.510 Max. :0.110000
## VeryActiveMinutes FairlyActiveMinutes LightlyActiveMinutes SedentaryMinutes
## Min. : 0.00 Min. : 0.0 Min. : 0.0 Min. : 0.0
## 1st Qu.: 0.00 1st Qu.: 0.0 1st Qu.:111.0 1st Qu.: 729.0
## Median : 2.00 Median : 6.0 Median :195.0 Median :1057.0
## Mean : 19.68 Mean : 13.4 Mean :185.4 Mean : 992.5
## 3rd Qu.: 30.00 3rd Qu.: 18.0 3rd Qu.:262.0 3rd Qu.:1244.0
## Max. :210.00 Max. :660.0 Max. :720.0 Max. :1440.0
## Calories
## Min. : 0
## 1st Qu.:1799
## Median :2114
## Mean :2266
## 3rd Qu.:2770
## Max. :4900## # A tibble: 1,397 × 15
## Id ActivityDate TotalSteps TotalDistance TrackerDistance
## <dbl> <chr> <dbl> <dbl> <dbl>
## 1 1503960366 3/25/2016 11004 7.11 7.11
## 2 1503960366 3/26/2016 17609 11.6 11.6
## 3 1503960366 3/27/2016 12736 8.53 8.53
## 4 1503960366 3/28/2016 13231 8.93 8.93
## 5 1503960366 3/29/2016 12041 7.85 7.85
## 6 1503960366 3/30/2016 10970 7.16 7.16
## 7 1503960366 3/31/2016 12256 7.86 7.86
## 8 1503960366 4/1/2016 12262 7.87 7.87
## 9 1503960366 4/2/2016 11248 7.25 7.25
## 10 1503960366 4/3/2016 10016 6.37 6.37
## # ℹ 1,387 more rows
## # ℹ 10 more variables: LoggedActivitiesDistance <dbl>,
## # VeryActiveDistance <dbl>, ModeratelyActiveDistance <dbl>,
## # LightActiveDistance <dbl>, SedentaryActiveDistance <dbl>,
## # VeryActiveMinutes <dbl>, FairlyActiveMinutes <dbl>,
## # LightlyActiveMinutes <dbl>, SedentaryMinutes <dbl>, Calories <dbl>## [1] "id" "activity_date"
## [3] "total_steps" "total_distance"
## [5] "tracker_distance" "logged_activities_distance"
## [7] "very_active_distance" "moderately_active_distance"
## [9] "light_active_distance" "sedentary_active_distance"
## [11] "very_active_minutes" "fairly_active_minutes"
## [13] "lightly_active_minutes" "sedentary_minutes"
## [15] "calories"## Rows: 1,397
## Columns: 15
## $ id <dbl> 1503960366, 1503960366, 1503960366, 1503960…
## $ activity_date <date> 2016-03-25, 2016-03-26, 2016-03-27, 2016-0…
## $ total_steps <dbl> 11004, 17609, 12736, 13231, 12041, 10970, 1…
## $ total_distance <dbl> 7.11, 11.55, 8.53, 8.93, 7.85, 7.16, 7.86, …
## $ tracker_distance <dbl> 7.11, 11.55, 8.53, 8.93, 7.85, 7.16, 7.86, …
## $ logged_activities_distance <dbl> 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0…
## $ very_active_distance <dbl> 2.57, 6.92, 4.66, 3.19, 2.16, 2.36, 2.29, 3…
## $ moderately_active_distance <dbl> 0.46, 0.73, 0.16, 0.79, 1.09, 0.51, 0.49, 0…
## $ light_active_distance <dbl> 4.07, 3.91, 3.71, 4.95, 4.61, 4.29, 5.04, 3…
## $ sedentary_active_distance <dbl> 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0.00, 0…
## $ very_active_minutes <dbl> 33, 89, 56, 39, 28, 30, 33, 47, 40, 15, 43,…
## $ fairly_active_minutes <dbl> 12, 17, 5, 20, 28, 13, 12, 21, 11, 30, 18, …
## $ lightly_active_minutes <dbl> 205, 274, 268, 224, 243, 223, 239, 200, 244…
## $ sedentary_minutes <dbl> 804, 588, 605, 1080, 763, 1174, 820, 866, 6…
## $ calories <dbl> 1819, 2154, 1944, 1932, 1886, 1820, 1889, 1…Data Analysis
Summary Statistics
## # A tibble: 1 × 3
## avg_steps avg_calories avg_distance
## <dbl> <dbl> <dbl>
## 1 7281. 2266. 5.22Total Steps by User
## `geom_smooth()` using formula = 'y ~ x'
Total Steps Over Time
# Identify trends by total steps over time
# Clean column names
cleaned_activity <- clean_names(all_activity)
# Inspect column names
colnames(cleaned_activity)## [1] "id" "activity_date"
## [3] "total_steps" "total_distance"
## [5] "tracker_distance" "logged_activities_distance"
## [7] "very_active_distance" "moderately_active_distance"
## [9] "light_active_distance" "sedentary_active_distance"
## [11] "very_active_minutes" "fairly_active_minutes"
## [13] "lightly_active_minutes" "sedentary_minutes"
## [15] "calories"# Clean column names
cleaned_activity <- clean_names(all_activity)
# Inspect the format of activity_date
head(cleaned_activity$activity_date)## [1] "3/25/2016" "3/26/2016" "3/27/2016" "3/28/2016" "3/29/2016" "3/30/2016"# Convert activity_date to Date format (adjust format as needed)
cleaned_activity <- cleaned_activity %>%
mutate(activity_date = mdy(activity_date))
head(cleaned_activity)# Summarize total steps by day
daily_steps <- cleaned_activity %>%
mutate(activity_date = as.Date(activity_date)) %>%
group_by(activity_date) %>%
summarize(total_steps = sum(total_steps, na.rm = TRUE))## `geom_smooth()` using formula = 'y ~ x'
User Distribution by Activity Level
#pie chart
# Calculate average steps per day for each user
user_avg_steps <- cleaned_activity %>%
group_by(id) %>%
summarize(avg_steps_per_day = mean(total_steps, na.rm = TRUE))
# Categorize users based on average steps
user_categories <- user_avg_steps %>%
mutate(category = case_when(
avg_steps_per_day > 12000 ~ "Very Active",
avg_steps_per_day >= 7500 & avg_steps_per_day <= 12000 ~ "Moderately Active",
avg_steps_per_day >= 5000 & avg_steps_per_day < 7500 ~ "Fairly Active",
avg_steps_per_day >= 2500 & avg_steps_per_day < 5000 ~ "Lightly Active",
avg_steps_per_day < 2500 ~ "Sedentary",
TRUE ~ "Unknown"
))
# Summarize the count of users in each category
category_summary <- user_categories %>%
count(category) %>%
mutate(percentage = n / sum(n) * 100)
# Order factor levels for category
category_summary$category <- factor(category_summary$category,
levels = c("Very Active", "Moderately Active", "Fairly Active", "Lightly Active", "Sedentary"))# Create pie chart with percentages
ggplot(category_summary, aes(x = "", y = percentage, fill = category)) +
geom_bar(width = 1, stat = "identity") +
coord_polar("y") +
geom_text(aes(label = paste0(round(percentage, 1), "%")),
position = position_stack(vjust = 0.5),
color = "black", size = 3) +
labs(title = "User Distribution by Activity Level", x = NULL, y = NULL) +
theme_void() +
scale_y_continuous(labels = scales::percent_format()) +
scale_fill_manual(values = c("Very Active" = "purple",
"Moderately Active" = "green",
"Fairly Active" = "orange",
"Lightly Active" = "yellow",
"Sedentary" = "pink"))
Calories vs. Steps
# Summarize data to get average steps and calories
daily_summary_calories_steps <- cleaned_activity %>%
group_by(activity_date) %>%
summarize(
avg_steps = mean(total_steps, na.rm = TRUE),
avg_calories = mean(calories, na.rm = TRUE)
)# Plot calories vs. steps with a regression line
ggplot(data = daily_summary_calories_steps, aes(x = avg_steps, y = avg_calories)) +
geom_point(color = "blue") +
geom_smooth(method = "lm", color = "red") +
labs(title = "Calories vs. Steps", x = "Average Steps", y = "Average Calories") +
scale_x_continuous(labels = function(x) format(x, scientific = FALSE)) +
scale_y_continuous(labels = function(x) format(x, scientific = FALSE)) +
theme_minimal()## `geom_smooth()` using formula = 'y ~ x'
Conclusion
Observations
- Total Steps by User:
- No significant trend is observed in total steps by individual users.
- Total Steps Over Time:
- An increase in total steps over time as the season gets warmer.
- User Distribution by Activity Level:
- The distribution of users by activity level is relatively even, with similar proportions of users in the “Very/Moderately Active” categories and the “Fairly/Lightly Active” categories.
- Calories Burned vs. Steps:
- There is a positive correlation between average steps and average calories burned, indicating that more active users tend to burn more calories.
Trends in Smart Device Usage
- Seasonal Variation:
- Smart device usage, particularly activity tracking, tends to increase in warmer seasons. This could be due to more outdoor activities and a general increase in physical activity during these times.
- User Engagement and Retention:
- Users tend to have varying levels of engagement with their devices. High engagement is often linked to more frequent and consistent usage, while low engagement might indicate sporadic use or abandonment of the device.
- Health and Wellness Integration:
- Users are increasingly interested in holistic health data, which includes not just activity tracking but also sleep patterns, heart rate monitoring, and nutritional information.
- Social and Community Features:
- Social features that allow users to connect, share achievements, and participate in challenges are becoming more popular and can enhance user engagement and retention.
Application of Trends to Bellabeat Customers
- Seasonal Promotions and Features:
- Bellabeat can introduce seasonal promotions and features that encourage users to stay active during warmer months. For example, launching seasonal challenges or offering discounts on outdoor activity accessories.
- Personalized Engagement:
- Implementing personalized engagement strategies based on user activity levels can help retain and motivate users. Tailored notifications, goal setting, and personalized feedback can enhance user experience.
- Holistic Health Tracking:
- Expanding Bellabeat’s product features to include comprehensive health metrics such as sleep analysis, heart rate monitoring, and nutrition tracking can provide more value to users and position Bellabeat as a holistic health and wellness brand.
- Community Building:
- Enhancing social and community features within the Bellabeat app can foster a sense of community among users. Features such as leaderboards, group challenges, and social sharing can increase user engagement.
Influence on Bellabeat Marketing Strategy
- Targeted Marketing Campaigns:
- Use data on seasonal trends to create targeted marketing campaigns that resonate with users’ activity patterns. For example, promoting outdoor activity products in the spring and summer.
- Segmentation and Personalization:
- Segment users based on their activity levels and personalize marketing messages to address their specific needs and motivations. Sedentary users might be encouraged with beginner-friendly content, while very active users could be targeted with advanced fitness products and features.
- Emphasize Holistic Health:
- Highlight Bellabeat’s commitment to holistic health in marketing messages. Emphasize features that track various health metrics and promote the overall wellness benefits of using Bellabeat products.
- Leverage User Stories and Testimonials:
- Share success stories and testimonials from users who have achieved their health and fitness goals with Bellabeat. This can inspire other users and build trust in the brand.
- Promote Social and Community Features:
- Market the social and community aspects of Bellabeat products to appeal to users who value social connections and shared experiences. Highlight community challenges, achievements, and social sharing capabilities.
By aligning Bellabeat’s marketing strategy with these smart device usage trends, the company can better meet the needs of its customers, enhance user engagement, and drive growth through targeted and personalized marketing efforts.