Analysis of Bellabeat Products

 Bellabeat Products Data Analytics Case Study

Introduction

The Bellabeat app provides users with health data related to their activity, sleep, stress, menstrual cycle, and mindfulness habits. This data can help users better understand their current habits and make healthy decisions. The Bellabeat app connects to their line of smart wellness products.

We have Ask, Prepare, Process, Analyze, Share and Act steps in data analysis to analyze any data.

Ask Phase:

Business Task - Need to identify trends in smart device usage and figure out how this will help Bellabeat's marketing strategy.

Prepare Phase:

To solve this business task I am using Fitbit Fitness Tracker Data From Kaggle, to identify trends in smart device usage. It includes data of daily activities, heart rate, sleep timings, calorie details and weight details of users calculated for a few days.

Process Phase:

Here, I am using Big query to sort, filter and remove null values from different datasets. I have imported dailyActivity_merged, sleepDay_merged and weightLogInfo_merged CSV files into the big query. dailyActivity_merged consists of data that is merged or aggregated from other hours and minutes databases.

As ID of the user is common in three databases. I have counted distinct IDs in three databases which give the following result.

dailyActivity_merged consists of 33 users

sleepDay_merged consists of 24 users

weightLogInfo_merged consists of 8 users

The result shows that data might be missing for weight and sleep details for few users or users are not interested in tracking their sleep and weight details. There are blanks in the Fat column of weightLogInfo_merged, but we are not using this column for analysis. Apart from this column as there are no blank or null values in these three databases we don't need to remove any rows.

Analyze Phase:

I have joined dailyActivity_merged, sleepDay_merged and weightLogInfo_merged in Tableau to infer relationships between different factors. As Sleep Day and Date columns are Datetime in both sleepDay_merged and weightLogInfo_merged databases it will throw an error while joining with dailyActivity_merged database where Activity Date is Date. I have changed them to date while joining them in Tableau.

I have visualized the Total Minutes Asleep, Calories Burnt and Total steps for all users for weekdays.

We can see from visualization that calories burnt and Total Minutes Asleep are the same for almost all days. But the Total steps have increased on Monday and Tuesday and again decreased from there, might be users are walking more on Monday and Tuesday as the week begins.

If we see the same details from the perspective of users we can see the below graph.

As we have merged three databases there are 6 common users in all three databases. This graph shows that there is a relation between total steps taken and calories burnt. The last person in the graph took more steps and burnt more calories.

The below graph shows the relation between the total distance and the average weight of users.

Here we cannot say that the total distance taken is related to their average weight. We are not sure if the user is underweight or overweight as we don't have the user's gender, age and height details.

The below graph shows the relation between the total time they are in bed and the total time they are asleep for all users.

Both are almost the same for some users but few users are spending time in bed before they are sleeping.

The below graph shows the relation between steps taken by all users at each hour.

Here, the highest peak is at 18 which is 6:00 PM from there again it is falling. The rising of the line started at 5:00 AM, which is normally wake-up time for many users.

Share Phase:

I can share the above visualizations which have been analyzed from data to stakeholders.

Act Phase:

• As we have few users, who are checking their sleep activity and weight information through their smart devices. Bellabeat products should signal the user to check their weight and sleep activity daily.

• Bellabeat smart device should collect their height, weight, gender, age and other details. Accordingly, Bellabeat's smart device should show how many steps they need to complete each day to become healthy. 

• As many users are spending time in bed before sleeping, the Bellabeat device should make an alarm sound to sleep after the user spends more than 20 minutes in bed.

• Within the peak hours that is from 5:00 AM to 6:00 PM, users are walking more. There is a fall at 15 which is 3:00 PM. In these cases, the Bellabeat device should signal the user to walk if the user is sitting idle for more than an hour.