Hello, I’m a junior data analyst working on the marketing analyst team at Bellabeat, a high-tech manufacturer of health-focused products for women. Bellabeat is a successful small company, but they have the potential to become a larger player in the global smart device market. Urška Sršen, Co-Founder and Chief Creative Officer of Bellabeat, believes that analyzing smart device fitness data could help unlock new growth opportunities for the company. I have been asked to focus on one of Bellabeat’s products and analyze smart device data to gain insight into how consumers are using their smart devices. The insights I discover will then help guide marketing strategies for the company. Go here to find the Bellabeat dataset that has been stored on Kaggle (CCO: Public domain, dataset made available through kaggle) https://www.kaggle.com/datasets/kyle007hendricks/bellabeat-dataset.
The best data sources contain all critical information needed to
answer questions and find solutions. I have researched every aspect
provided by this dataset so that I can make sure the fitbit’s were in
accurate standings of Bellabeats standards.
This data is still very useful since the data is current and relevant
and I have cited all my resources to prove that the information I am
providing is credible.
When I’m choosing a data source, I think about three
things;
* Who created the data set?
* Is my data a part of a credible organization?
* And when was the data last refreshed?
* The Bellabeat dataset that I am using was Last updated 3 years ago and
was created by Kyle Hendricks and yes, this data set is from a credible
online organization named Kaggle and has over 8.82 Usability. The owner
of this data set is keeping data anonymous which is the process of
protecting people’s private or sensitive data by eliminating identifying
information.
The data set provides information about each participants; Id,
Activity date, total steps, total distance miles, tracker distance,
logged activities distance, very active distance, moderately active
distance, light active distance, sedentary active distance, very active
minutes, fairly active minutes, lightly active minutes, sedentary
minutes, calories.
However, I included fake names for each membership id since
the data is vastly different across three categories. In
each category, the pictures also vary by a wide margin, almost as much
as they do between categories. Further, within a given compartment the
pictures may be different, but consistent similarities persist.
I used the excel spreadsheets to clean and process all of the
data.
Afterwards, I utilized The Microsoft office Document and PowerPoint to
bind all of my findings together and the programming language R to
present how I found relationships in Fitbit usage for April and May of
2016.
In most cases I would check for null values and dispose of them to keep
my data clean but, because my data is consisting mainly of numerical
information any and all null or zero values are extremely important. So
transposing my data was not necessary.
I split everything into compartments starting with
steps versus calorie loss, then distance by months, and
finally sedentary and logged entries.
After I formatted my data in Excel I replaced all of the ID numbers with
fake names to distinguish between participants. This allowed me to find
any null values and to prevent cleaning mistakes. So you will notice a
few things when going over this data.
* Data aggregation so it’s useful and accessible.
* Organized and formatted data.
* Calculations.
* Trends and Relationships.
Now when we view this table we are looking for relationships.
Relationships between April and May. Finding relationships
between steps and calorie loss. We are only using the product
to measure usage and so far we see that most people like to
track only strenuous activity.
Now you may ask why is that? Although if you remember the beginning of
the Fitbit campaign it would make a lot sense.
Everyone was running, walking, hiking, taking their dogs out on runs
not sitting around enjoying their cups of coffee. However, now due to
smartwatches being normalized we see that sedentary hours may outweigh
the active hours that a person logs while wearing their fitbits.
* The maximum amount of steps were mostly congruent with the
amount of calorie loss.
* There was a correlation between 497K Steps with 106K Calorie
Loss
* We notice trends in participant recordings and I have added Id numbers
and fictional names for filtering. Please review the Total Steps, Total
Distance and Very Active Minutes for Handy Mandy(8877689391), Little
Mermaid(3372868164) and Optimus Primes(4057192912). Fictional names were
added to distinguish between participants easier.
* Active Distance is the amount of measured miles a
person is active for. And as we can see here the average amount of miles
each participant walks is 1.5 miles.
Also as you have noticed our data is only including 2 months in 2016 so
it does help us predict future months and it helps that both of these
had holidays within them as well.
* Sedentary Distance is another way of saying how
far someone is going without being active. Movement at rest. That may
sound silly but we actually rack up a lot of sedentary distance everyday
in vehicular transportation.
Active distance is tracked in miles and through any activity that
elevates the heart rate and allows your body to lose calories.
I did find a surprising discovery when
analyzing my data and as you can see looking at the Image, I realized
that there are no Jedi(2347167796), Dexter(8253242879) or Optimus
Prime(4057192912) logs in total distances for the month of May.
* Strategy 1 : Use Fitbit to track rest. REST IS PRODUCTIVE. Not only is sleep important and essential but, when a fitbit records times of rest it can also be referring to someone at work or with their children reading, people in class etc. So Let’s add in options for the types of rest we are engaged in.
* Strategy 2 : Get fun competitions
going!What is a Bellabeat challenge? Well we can do multiple
different challenges but here are two fun ideas.
1). If you reach 10K steps 3 days in a row then you win a badge on your
profile after 60K steps you get to unlock 5% off. As for a fun
competition with others you can use the distance tracker and plot QR
does that people scan once they run/walk to each destination.
*If you can reach all of them before the end of 3 months either alone or
with others then you get a swag bag. If you can do it in 2 months you
get 25% of any in-store item and If you can accomplish this task in one
month you get a coupon to but new running shoes (Like 45% off [any
partners like Skechers] running shoes and the coupon has to be used in 6
months) and a swag bag.
2). To keep it short and simple it would be the equivalent of a game
named Pokemon Go. What people would do is walk, hike or run, to certain
locations alone or with others and track their progress through the app.
Once they reach each destination point they can add up points to get
5-10% off of other Bellabeat purchases.
* Strategy 3 : Have Fitbit purchase
incentives. Create benefits or promotions added to any
transactions. 7% off any multicolored fitbit bands.
Get competitive with other Bellabeat users. Join up and try to get mile
averages higher which will help people have more total steps and strive
to reach higher goals.
The main goal is to have fun, include others to keep
the app in use and to rest but track any rest with explanations so there
is no shame or un-logged entries.
Moreover, we also saw getting people to use their device
while sedentary would be extremely useful. When customers
record sedentary hours as; working, in class, breastfeeding, or resting
they would be providing more detailed data of Bellabeat’s consumer base
and feel less guilt for not being active.
Most people enjoy have a goal to reach so that they feel
accomplished.
Plus, There’s this fun quote that goes, “It’s about the
journey not the destination” but, there are a lot of people who
prefer a destination to reach.
So as a new campaign strategy I think it would yield
amazing results if we focused on the adventure and
the relationships and the fun people
have while using Bellabeat products. TRACK THE FUN NOT THE
STRESS.
Camping Bachelorette Party https://www.pinterest.com/pin/788763322272199335/
Walking Shoes https://www.pinterest.com/pin/3799980929357605/
Decorative Eggs https://www.pinterest.com/pin/204984220531740148/
Taco Tuesday https://www.pinterest.com/pin/68739642878/
Kaggle Bellabeat Dataset https://www.kaggle.com/datasets/kyle007hendricks/bellabeat-dataset
Published by Victoria De La Cruz