Your task is to ‘get to know’ the data by conducting some statistical analysis using BBB’s customer database.

Description

Create a word file that (1) copy and paste the 10 questions above, (2) below each
question, copy and paste your code and output from R into the word file, (3) highlight
the part of the output that answers the respective question, (4) convert the word file
into a PDF and submit it using Canvas.
Note that if you are familiar with R Notebook/Markdown, you may directly output your
code/output as a PDF. If you are unfamiliar with Notebook, no worries as I will cover it
next time.

Don't use plagiarized sources. Get Your Custom Assignment on
Your task is to ‘get to know’ the data by conducting some statistical analysis using BBB’s customer database.
From as Little as $13/Page

Unformatted Attachment Preview

Kellogg School of Management
Northwestern University
The BookBinders Book Club: Basic Customer Analysis
BOOK CLUBS
Historically, book clubs offered their readers continuity and negative option programs that were
based on an extended contractual relationship between the club and its clients. Under a
continuity program, a reader signs up for an offer of several books for a few dollars each (plus
shipping and handling), and an agreement to receive a shipment of one or two books each
month thereafter. This is a “low maintenance” arrangement since a single contract guarantees
a sequence of sales. It is most common for children’s books, where parents are willing to
delegate the rights to make a selection to the book club, and in fact much of the club’s prestige
depends on the quality of its selection. In a “negative option” program, readers get to choose
which and how many additional books they would receive, but the default option is that the
club’s selection will be delivered to them each month. The club informs them of the monthly
selection and they are specifically required to mark “no” by a deadline date on their order form if
they do not want to receive it. Negative option programs sometimes result in customer
dissatisfaction and always give rise to significant mailing and processing costs.
In an attempt to reverse these trends and combat the success of superstores and online
retailers, some firms are beginning to offer books on a positive option basis, but only to selected
segments of their customer lists that are deemed receptive to specific offers. Thus, book clubs
are beginning to use customer analytics to work smarter rather than expand the coverage of
their mailings. They target individual customers based on data in their databases to select only
customers who are likely to be interested in their offers, and they differentiate their offers across
their customer population.
THE “BOOKBINDERS BOOK CLUB” – BBB
The BBB Club was established for the purpose of selling specialty books through direct
marketing involving a variety of channels, including media advertising (TV, magazines,
newspapers) and mailing. BBB is strictly a distributor and does not publish any of the books it
This case was originally prepared by Nissan Levin and Jacob Zahavi, Faculty of Management, Tel Aviv University
and subsequently modified by Professor Charlotte Mason and Melissa Martin of the University of North Carolina. The
case was further modified by Florian Zettelmeyer for use in his course and for use with Stata instead of SPSS.
Professor Song Yao revise the case for use in his course and for the applications of R. This case was developed to
provide material for class discussion rather than to illustrate either effective or ineffective handling of a business
situation. Names and data may have been disguised to assure confidentiality.
Copyright Ó 2013 by Charlotte Mason and Florian Zettelmeyer.
Page 2
sells. In anticipation of using database marketing, BBB made a strategic decision right from the
start to build and maintain a detailed database about its club members containing all the
relevant information about their customers. Readers fill out an insert that is returned to BBB
which then enters the customer into the database. By now the company has built a database of
500,000 readers through advertising in specialty magazines.
Initially, and as long as the customers’ database was still relatively small, BBB contacted all its
club members with each appeal to purchase books. Every other month, the company sent out
solo mailings for its latest offering (that is, each offer is for one book). BBB’s sales have grown
steadily, but profits began falling when the database got larger and when the company
diversified its book selection and increased the number of offers sent to customers. The falling
profits have led BBB to switch to customer analytics in order improve its mailing yields and stay
profitable. BBB’s management has decided to base customer analytics on the following
principles:







New members would be acquired by advertising in specialty magazines,
newspapers, TV.
Existing club members would be contacted by direct mail and telemarketing.
The mailing / telemarketing list would be chosen using database marketing
technology.
Customer response, whether purchase or no purchase, will be recorded and
maintained in the database, to be used in future targeting of audiences for promotion
(in fact, the management considers this database as the main asset of the
company).
Every new book would be offered to the club members first – prior to advertising it in
the media.
The price in media advertising would always be higher than the one in the direct
mailing/telemarketing offer.
Live market tests, involving a random sample of customers from the database, would
be conducted for new book editions in order to analyze customers’ response and
calibrate a response model for the current book offering. The response model’s
results will then be used to “score” customers in the database and select customers
for the mailing campaign.
The idea is to run simultaneous targeted campaigns where each target audience will receive
appropriate solo mailings.
Parallel to selling books, BBB management has also been taking advantage of its database to
offer its members non-book products. Based on the success that the company had in the past
with the non-book operation, it plans to continue this operation, even expand it, in the future.
THE DECISION PROCESS
Two core decisions are usually supported by customer analytics: targeting and prediction.
Targeting is concerned with selecting the audience that is most receptive to the current offering.
Prediction is concerned with forecasting the number of orders generated. The decision process
usually consists of several steps:


Testing
Response modeling
2
Page 3


Scoring
Decision making
The process starts with a TEST mailing, where a sample of customers are selected randomly
from the database and mailed the new offer.
The responses to the offer are then analyzed along with customers’ past purchase history,
demographics and other relevant data to determine how the response varies as a function of the
customers’ attributes and history. Purchase history includes variables such as: how recently
they have purchased (“Recency”), how often they have purchased (“Frequency”), and how
much money they have spent on buying the company’s products in the past (“Monetary”) (the
so-called RFM variables). Other pieces of purchase history are the number of books bought by
various categories, and whether or not a customer has bought specific related products in the
past. The resulting model is referred to as a Response model.
The response model calibrated on the basis of the TEST results is next used to assign a “score”
for each customer in the balance of the list (i.e., for customers who were not part of the TEST
mailing, which constitute the majority of the list), reflecting the customers’ “likelihood” of
purchasing the current product offering.
Finally, customers are selected for the promotion based on their expected likelihood of
purchase, the revenue generated by a sale, and the cost of mailing the offer. The final mailing is
referred to as the ROLL mailing.
3
Page 4
Exhibit 1
The BookBinders Book Club Dataset
Summary information about the BookBinders Book Club’s customers’ purchasing history and
demographics is in the dataset called BBB.csv.
Below is a listing of the variable names and descriptions of the data types:
Contents of BBB.csv – contains records for 50,000 customers
Variable name
Description
acctnum
Customer account number
gender
Customer gender – M=male, F=female
state
State where customer lives (2-character abbreviation)
zip
ZIP code (5-digit)
zip3
First 3 digits of ZIP code
first
Number of months since first purchase
last
Number of months since most recent purchase
book
Total dollars spent on books
nonbook
Total dollars spent on non-book products
total
Total dollars spent
purch
Total number of books purchased
child
Total number of children’s books purchased
youth
Total number of youth books purchased
cook
Total number of cook books purchased
do_it
Total number of do-it-yourself books purchased
refernce
Total number of reference books purchased
art
Total number of art books purchased
geog
Total number of geography books purchased
buyer
Did the customer buy “The Art History of Florence?”
(1=yes, 0=no)
training
Dummy variable that splits the dataset into a training
(“1” and validation (“0”) dataset. This variable is used
only later in the course.
4
Page 5
YOUR TASK
Your task is to ‘get to know’ the data by conducting some statistical analysis using BBB’s
customer database. Summary information about the BookBinders Book Club’s customers’
purchasing history and demographics is in the dataset called BBB.csv.
Exhibit 1 (see above) contains a listing of the variable names and descriptions of the data types
that you should use to answer the following questions:
1. What percent of BookBinders customers are female?
2. Which three states account for the largest percentage of BookBinders’s customers?
3. What is the average Total $ spent, the average Total # of book purchases, and the average
number of months since last purchase?
4. Calculate the correlation between customers’ total spending on books and their total
spending on non-book products.
5. Which book categories have sold the most books? Which have sold the least?
6. Create a bar chart showing the average total spending on books for males and females.
7. For both males and females, find their respective total number and also the percent who
bought “The Art History of Florence.”
8. For both males and females, determine the total number of purchases and the average
number of purchases by males vs. females.
9. Determine the minimum, the maximum, and the average number of months between
customers’ first purchase and their most recent purchase.
10. What percent of repeat customers (those with two or more total purchases) bought “The Art
History of Florence?”
11. Create a word file that (1) copy and paste the 10 questions above, (2) below each
question, copy and paste your code and output from R into the word file, (3) highlight
the part of the output that answers the respective question, (4) convert the word file
into a PDF and submit it using Canvas.
Note that if you are familiar with R Notebook/Markdown, you may directly output your
code/output as a PDF. If you are unfamiliar with Notebook, no worries as I will cover it
next time.
5

Purchase answer to see full
attachment