📘 Business Task

Bellabeat, a high-tech manufacturer of health-focused smart products for women, wants to leverage data to inform its marketing strategy. As a junior data analyst, your task is to analyze smart device usage patterns (using Fitbit data) and generate actionable insights that could be applied to Bellabeat’s product suite — particularly the Leaf wellness tracker.

📂 Data Source

Dataset: Fitbit Fitness Tracker Data
Files used: dailyActivity_merged.csv, sleepDay_merged.csv, weightLogInfo_merged.csv
Tools: R, tidyverse, ggplot2, janitor, lubridate

🧹 Data Cleaning

library(tidyverse)
library(janitor)
library(lubridate)
library(skimr)

daily_activity <- read_csv("dailyActivity_merged.csv") %>% clean_names()
sleep_day <- read_csv("sleepDay_merged.csv") %>% clean_names()

daily_activity <- daily_activity %>% mutate(activity_date = mdy(activity_date))
sleep_day <- sleep_day %>% mutate(sleep_day = mdy_hms(sleep_day))

daily_activity <- distinct(daily_activity)
sleep_day <- distinct(sleep_day)

🔍 Analysis & Visualizations

📊 Steps vs Calories Burned

ggplot(daily_activity, aes(x = total_steps, y = calories)) +
  geom_point(color = "#FF6F61", alpha = 0.6) +
  geom_smooth(method = "lm", se = FALSE, color = "#333333") +
  labs(title = "Steps vs Calories Burned", x = "Total Steps", y = "Calories Burned") +
  theme_minimal()

📌 Insight: There is a strong positive correlation between the number of steps taken and calories burned. This validates the use of step goals in health-tracking apps as an effective strategy for promoting physical activity.

😴 Sleep vs Very Active Minutes

activity_sleep <- inner_join(daily_activity, sleep_day, by = c("id", "activity_date" = "sleep_day"))

ggplot(activity_sleep, aes(x = total_minutes_asleep / 60, y = very_active_minutes)) +
  geom_point(color = "#6C5B7B", alpha = 0.6) +
  geom_smooth(method = "lm", color = "#333333", se = FALSE) +
  labs(title = "Sleep Hours vs Very Active Minutes", x = "Sleep Hours", y = "Very Active Minutes") +
  theme_minimal()

📌 Insight: The relationship between sleep duration and very active minutes is weak, indicating that high activity does not necessarily reduce sleep time. This highlights a potential for Bellabeat to encourage both physical activity and adequate rest without compromise.

📈 Activity Types Comparison

daily_activity %>%
  select(very_active_minutes, fairly_active_minutes, lightly_active_minutes, sedentary_minutes) %>%
  pivot_longer(everything(), names_to = "activity_type", values_to = "minutes") %>%
  ggplot(aes(x = activity_type, y = minutes, fill = activity_type)) +
  geom_boxplot(alpha = 0.7) +
  scale_fill_brewer(palette = "Pastel1") +
  labs(title = "Comparison of Activity Types", x = "Activity Type", y = "Minutes per Day") +
  theme_minimal()

📌 Insight: Users spend the majority of their time in sedentary activity, with very little time being spent in highly active states. This presents a key opportunity for Bellabeat to introduce motivational features encouraging users to break up long sedentary periods with movement.

📌 Key Insights

Users average ~7 hours of sleep, close to health standards.
Strong positive correlation between steps and calories burned.
High sedentary time offers marketing opportunities for Bellabeat.
Sleep and activity levels do not strongly correlate.

🧠 Recommendations

Market Leaf as a behavior-change tool, not just a tracker.
Encourage habits like hourly walking, sleep tracking, and hydration.
Use in-app challenges and progress visuals to boost engagement.
Leverage insights to create motivational marketing content.

✅ Conclusion

The analysis reveals consistent patterns in physical activity, sedentary time, and sleep among Fitbit users. These insights can guide Bellabeat in refining their product positioning and user engagement strategy — especially for the Leaf smart wellness tracker.

Bellabeat Smart Device Marketing Analysis

Sreerup Banerjee