Scenario
It was founded in 2013, Bellabeat has grown rapidly and quickly
positioned itself as a tech-driven wellness company for women. The
company has 5 focus products: bellabeat app, leaf, time, spring and
bellabeat membership. Bellabeat is a successful small company, but they
have the potential to become a larger player in the global smart device
market. I have been asked to analyze smart device data to gain insight
into how consumers are using their smart devices. The insights we
discover will then help guide marketing strategy for the company.
Ask
Buisness Task:
Analyze Fitbit data to gain insight and help guide marketing
strategy for Bellabeat to grow.
Questions for Analysis:
The following three points are the questions needed to be answered
by this analysis.
What are some trends in smart device usage?
How could these trends apply to Bellabeat customers?
How could these trends help influence Bellabeat marketing
strategy?
Key Stakeholders
Primary stakeholders: Urška Sršen and Sando Mur, executive team
members.
Secondary stakeholders: Bellabeat marketing analytics team.
Prepare
The Data
The data source used for our case study is FitBit Fitness Tracker
Data with 18 CSV. This dataset is stored in Kaggle and was made
available through Mobius. Bellabeat Data Credibility
Only 30 user data is available. This is not an accurate
representation of the entire population but meets the central limit
theorem general rule of n≥30. Therefore, the sample size is sufficient
for analysis.
Importing Datasets and Libraries
install.packages(“tidyverse”) library(tidyverse)
install.packages(“lubridate”) library(lubridate)
install.packages(“ggplot2”) library(ggplot2) install.packages(“tidyr”)
library(tidyr)
install.packages(“janitor”) library(“janitor”)
install.packages(“ggpubr”) library(ggpubr)
Datasets
The original data sets provided contain 18 CSV files, but I will
only be using two of them: dailyActivty_Merged and sleepDay_merged. I
will also be renaming the CSV files so they are easier to work with.
“dailyActivity_merged has a majority of data we can use to perform the
analysis: calories, total, step and the active minutes.
Import dataset dailyActivty and sleepDay and rename file to make it
easier to work with.
library(readr) activity <- read_csv(“dailyActivity_merged.csv”)
sleep <- read_csv(“sleepDay_merged.csv”)
head(activity) head(sleep)
Process
Let’s check our data to see if it is ready for analysis
Checking Unique ID’s
First I will check to see how many unique Ids there are in our
data.
install.packages(“dplyr”) library(dplyr) n_distinct(activity\(Id) n_distinct(sleep\)Id)
Duplicates
Check for duplicates and remove them
sum(duplicated(activity)) sum(duplicated(sleep))
Here we double check that the duplicates were actually removed
sleep <-sleep %>% distinct() %>% drop_na()
install.packages(“tidyr”) library(tidyr)
Indeed, they were removed
Cleaning
clean_names(activity) install.packages(“janitor”) library(janitor)
clean_names(sleep)
Analyze
Summarize and Analyze
activity %>% select(TotalSteps, TotalDistance, Calories) %>%
summary()
The users spend a rounded up average of about 7 hours of sleep and
spend an average of 7.6 hours in their bed. The CDC recommends 7 or more
hours on average.
We have 4 categories to look at for our activity levels
Very active average: 21.26 minutes Fairly active average:
13.56 minutes Lightly active minutes: 192.8 Sedentary minutes
average: 991.2
Notice that our users spend a majority of their active time on
“light activity” which could include things such as general housework or
office work.
The majority of their day (16.5 hours) was spent in
“SedentaryMinues”
*We can use push notifications to remind users to get up and be
active
Merging Data
Now we merge the data for easy to understand visualizations
activity_sleep <- merge(activity, sleep, by= c(“Id”,“date”), all.x
= TRUE)
Share
Total Time Asleep vs. Total Time in Bed
ggplot(data= subset(activity_sleep,!is.na(TotalMinutesAsleep)),
aes(TotalSteps,TotalMinutesAsleep)) + geom_rug(position= “jitter”,
size=.08) + geom_jitter() + geom_smooth() + labs(title= “Total Steps
vs. Sleep Time”) + theme_minimal() install.packages(“ggplot2”)
library(ggplot2)
I do not notice a real correlation between Total Steps and Sleep
Time. Meaning that amount of sleep does not have any effect on activity
levels for our selected users.
Total Steps and Calories
ggplot(data=activity, aes(x=TotalSteps, y=Calories)) +
geom_point(color=“orange”) + geom_smooth(color=“blue”) +
labs(title=“Total Steps vs. Calories”) ### The correlation between the
amount of steps taken and the calories burned are very strong. ###Shown
here the more active the user is the more calories you will burn.
*Bellabeat can encourage users to get active trough notification during
the day
Act
Here are my findings:
The average user is receiving the minimum hours of sleep per night
with an additional time in bed but not asleep.
There is no correlation on how much sleep they get and how active
they are. I notice that the more steps you take, the more calories you
burn, even during light activity.We confirmed that a majority of users
spend most of their day sedentary.
Recommendations
Being stimulated and active before bedtime can lead to unhealthy
sleeping habits. To remedy this, Bellabeat can recommend users to add a
bedtime routine through the app and send push notifications for bedtime
reminders a couple hours or an hour before bed. This will include
reminders to wind down before bed.
Bellabeat can encourage users to be more active throughout the day
by sending reminders to get up and get active during their day.
Conclusion
There are many ways in which Bellabeat can help users improve their
lives with daily activity. Even normal daily housework or light work can
improve one’s health. The most simple way is by creating healthy habits
with gentle reminders to be active or calm down and rest.