Summary

Bellabeat is a high-tech company that manufactures health-focused smart products informing and inspiring women from around the globe. This company empowers women with knowledge about their own health by collecting data on activity, sleep, stress, and reproductive health.

This data set was supplied from “Fitbit Fitness Tracker Data” made available through Mobius from Kaggle and holds personal fitness trackers from thirty, consented Fitbit users with output between the dates ‘2016-04-12’ to ‘2016-05-12 ’for daily physical activity, steps, heart rate, and sleep monitoring. I felt Bellabeat’s “Time” smartwatch would be a great comparison against Fitbit’s Tracker seeing they share some of the same qualities.

CBS News reported, Piper Jaffray estimated smartwatches “made up to 6 percent of new watch purchases in 2014 but could rise to 31 percent by 2018” (Brown, 2015). This would mean an increase averaging about 5.2% per year. I concluded that with the watch set to increase within the next couple of years, one thing we needed to evaluate was how often the participants were wearing their devices and what types of readings they were receiving.

The marketing analytics team and I have been asked to focus on one Bellabeat product and analyze their usage data in order to gain insight into how people are already using their smart devices. This would supply high-level recommendations for informing Bellabeat to help answer the following questions:

1. What are some trends in smart device usage?

2. How could these trends apply to Bellabeat customers?

3. How could these trends help influence Bellabeat marketing strategy?

Filtering

For this presentation, I used Microsoft Excel, Tableau, SQL, and R and searched through the data to begin the cleaning process below:

Excel

• Uploaded data to Excel, utilizing formula “=round” to round numbers to two decimal places

*Note: Attempts were made to separate time and date on some of the data using “=time,” but this was unsuccessful.

SQL

• Scanned the data I felt would be pertinent to answer their questions and began synchronizing daily activities and minutes, calories, steps, heart rate, sleep, MET and weight log to Bigquery.

• I typed a query to find the distinct number of people who wore their fitness trackers daily between the dates of April 12th to May 13th from daily_activities. I extended the date by a day just in case. Found there were 33 participants.

• Continued to query distinct number of people and date from the rest of the data of calories, steps, heart rate, sleep, MET and weight log

• Changed some columns from f0_ to more descriptive columns such as “Id” for clarification.

• Renamed columns and removed duplicates

Program R

• Uploaded cleaned data from SQL and installed packages and libraries of tidyverse, dplyr, janitor, and ggplot.

• Used “glimpse" on all datasets heart rate, sleep, MET, weight log and daily activities

*Note: You can click here for the full documentation

Analysis

After cleaning the data, it was time to apply my data visualizations. I noticed when they were asking about the trends for marketing strategy, I wanted to portray just how many people used certain Fitbit features along with their data. This would show just how much and how long they were wearing their devices during such activities.

• All 33 participants used their Fitbit for their daily activities of active minutes, distance and calories.

• 27 participants utilized the MET feature

• 24 participants used their tracker for sleep

• Decrease in weight log of 8 participants

• Total of 7 participants for the heart rate

Take a peak at my data viz and check out the "Data Explorer" area or feel free to view in Tableau.

Sharing Suggested Solutions

After analyzing the data presented, I came across a few helpful insights for Bellabeat.

• The “Time” smartwatch has features for daily activities, sleep, meditation, menstrual cycles, and stress. An added feature called, “fat-burn” could be beneficial. This would possibly allow more customers to check their heart rate during their activities as well as motivate them to increase their ability to be more active. Healthline states fat burning heart rate “is at about 70 percent of your maximum heart rate.” We could “determine the maximum heart rate by subtracting the age from 220.” (Marcin, 2019)

• 24 participants used the Fitbit for sleep; however, their data was inconsistent possibly because of having to recharge their devices. Therefore, an excellent quality battery would be helpful to allow users to be given a better opportunity for capturing sleep data.

• People are less likely to log their weight if this has to be done manually. We could introduce an additional product called "Bellabeat Scale" that could automatically log their weight (lbs. and kg) and BMI to their phones.

Worked Cited

• Brown, H. (2015, July 23). Good question: How many people still wear watches? CBS News. Retrieved August 7, 2022, from https://www.cbsnews.com/minnesota/news/good-question-how-many-people-still-wear-watches/

• Marcin, A. (2019, March 8). Fat-burning heart rate: What is it, how to calculate, and chart by age. Healthline. Retrieved August 7, 2022, from https://www.healthline.com/health/fitness-exercise/fat-burning-heart-rate#_noHeaderPrefixedContent