Smart device usage
Tholoana
2025-11-01
Ask: Business Task & Objective
Summary of Business Task
The purpose of this project is to analyse the smart device usage to gain insights into how consumers use their smart devices. By focusing on the competitor dataset, we can track how consumers use these smart devices and make solid decisions about how are we going to improve our marketing strategy.
Client/Sponsor
Cofounder and Chief Creative Officer, Urska Srsen.
Data Source
The CCO provided a “Fitbit Fitness Tracker Data” that is made available through Mobius on kaggle.
Prepare: Data Quality Assessment
Data Quality Issues
The issue is that the data is not current, meaning it is now outdated.Thirty eligible Fitbit users consented to the submission of personal tracker data, including minute-level output for physical-activity, heart rate, and sleep monitoring. This data lacks key metrics important to Bellabeat’s goal, which is stress, reproductive health, and mindfulness.
Process: Data Cleaning and Organization
Tool Selection
I decided RStudio will be the best tool to perform all the necessary tasks since data is stored in frames and functions and formulas are readily available.
Process: Data Cleaning and Organization
# Loaded necessary libraries
library(tidyverse)
# 1. Loaded the data frames (Adjust file paths as needed for your setup!)
dailyActivity_merged <- read_csv("C:/CASE STUDY/Fitabase Data 3.12.16-4.11.16/dailyActivity_merged.csv")
minuteSleep_merged <- read_csv("C:/CASE STUDY/Fitabase Data 3.12.16-4.11.16/minuteSleep_merged.csv")
# 2. Clean/Formatting of the minuteSleep data
minuteSleep1_merged <- separate(minuteSleep_merged, date, into = c("date", "time"), sep = " ")
minuteSleep2_merged <- minuteSleep1_merged %>%
mutate(time = hms::as_hms(time), minutes = round(as.numeric(time)/60, 2))
print(minuteSleep2_merged)## # A tibble: 198,559 × 6
## Id date time value logId minutes
## <dbl> <chr> <time> <dbl> <dbl> <dbl>
## 1 1503960366 3/13/2016 02:39:30 1 11114919637 160.
## 2 1503960366 3/13/2016 02:40:30 1 11114919637 160.
## 3 1503960366 3/13/2016 02:41:30 1 11114919637 162.
## 4 1503960366 3/13/2016 02:42:30 1 11114919637 162.
## 5 1503960366 3/13/2016 02:43:30 1 11114919637 164.
## 6 1503960366 3/13/2016 02:44:30 1 11114919637 164.
## 7 1503960366 3/13/2016 02:45:30 2 11114919637 166.
## 8 1503960366 3/13/2016 02:46:30 2 11114919637 166.
## 9 1503960366 3/13/2016 02:47:30 1 11114919637 168.
## 10 1503960366 3/13/2016 02:48:30 1 11114919637 168.
## # ℹ 198,549 more rows# 3. Creating the Daily_Minutes table
minuteSleep3_merged <- head(minuteSleep2_merged)
dailyActivity1_merged <- head(dailyActivity_merged)
Daily_Minutes <- bind_cols(dailyActivity1_merged %>% select(ActivityDate, SedentaryMinutes), minuteSleep3_merged %>% select(date, minutes))
Daily_Minutes <- Daily_Minutes %>% rename(SleepMinutes = minutes)
print(Daily_Minutes)## # A tibble: 6 × 4
## ActivityDate SedentaryMinutes date SleepMinutes
## <chr> <dbl> <chr> <dbl>
## 1 3/25/2016 804 3/13/2016 160.
## 2 3/26/2016 588 3/13/2016 160.
## 3 3/27/2016 605 3/13/2016 162.
## 4 3/28/2016 1080 3/13/2016 162.
## 5 3/29/2016 763 3/13/2016 164.
## 6 3/30/2016 1174 3/13/2016 164.write.csv(Daily_Minutes, "Daily_Minutes1.csv", row.names = FALSE)Analyze: Trend Relationships
Daily Sedentary Minutes
The first plot shows the relationhip between date and Sedentary minutes. We can see that consumers use most of their time doing less.
library(ggplot2)
ggplot(data = Daily_Minutes, aes(x=ActivityDate, y=SedentaryMinutes)) +
geom_point() +
labs(title = "Daily Sedentary Minutes", x="Activity Date", y="Sedentary Minutes") +
theme(plot.title = element_text(size = 14, face = "bold"))
Daily Sedentary Minutes
The second plot being Daily Sleep Minutes also shows the relationship between date and minutes.
library(ggplot2)
ggplot(data = Daily_Minutes, aes(x=SleepMinutes, y=date)) +
geom_point() +
labs(title = "Daily Sleep Minutes", x="Sleep Minutes", y="date") +
theme(plot.title = element_text(size = 14, face = "bold"))
Daily Sleep Plot
Act: Business Impact and Final Conclusion
The analysis of the “Fitbit Fitness Traker Date” revealed two key insights regarding consumer usage: high Sedentary Minutes and consistently low Daily Sleep Minutes (less than 200 minutes). These trends suggest that consumers are generally inactive and are not prioritizing sleep, potentially leading to increased stress and poor health.
Recommendation: focus on Bellabeat Membership
To address these pain points and inform the Bellabeat marketing strategy, the key recommendation is to heavily promote and invest in the Bellabeat Membership. The product directly fulfills the customer’s need for motivation. encouragement, and guidance to reach their greatest potential.
Marketing Strategy
The marketing plan should focus on maximizing vizibility across various platforms: Investment: Allocate more resources to enhancing and promoting the membership product. Channels: Utilize social media platforms, radio advertisements, and billboards. Key Message: “Highlight that the membership offers users 24/7 access to fully personalized guidance. This guidance covers: Nutrition Activity Sleep Health and Beauty *Mindfulness
By emphasizing this comprehensive, personalized support consumers will feel the encouragement, comfort, and support that Bellabeat offers, driving new subscriptions.