READ AND FOLLOW ALL INSTRUCTIONS CAREFULLY!
Goal is to perform a statistical and financial analysis of which customers should be targeted in a magazine marketing campaign and provide a taste of how teams operate in companies when working together on a project.
The CEO/Chairperson (Professor) has assigned students to teams at the end of this document.
Background
A Data Mining Example
A magazine reseller, MagWeb, is trying to decide what magazines to market to customers. In the “old days,” this might have involved trying to decide which customers to send advertisements to via regular mail. In today’s world, MagWeb wants to decide what magazine offers to include in emails to select prospective customers as part of an email marketing campaign. All of the emails that will be sent will go to customers that have previously bought a magazine subscription at MagWeb and who have not opted out of receiving emails.
The Sample – Obtaining the Data
Because all of the recipients of the emails have previously made a purchase at MagWeb, the company can match the data collected when the customer made their previous purchases with third party data (which can be purchased from data sources such as credit scoring agencies), so they have quite a lot of information about each customer. For example, they have data such as income, number of people in the household, and so on. This kind of merging of data from multiple sources to assemble a remarkably rich “profile” of each customer is becoming increasingly common.
Here are the variables that MagWeb has on each customer from third-party sources:
• Household Income (Income; rounded to the nearest $1,000.00)
• Gender (IsFemale = 1 if the person is female, 0 otherwise)
• Marital Status (IsMarried = 1 if married, 0 otherwise)
• College Educated (HasCollege = 1 if has one or more years of college education, 0 otherwise)
• Employed in a Profession (IsProfessional = 1 if employed in a profession, 0 otherwise)
• Retired (IsRetired = 1 if retired, 0 otherwise)
• Not employed (Unemployed = 1 if not employed, 0 otherwise)
• Length of Residency in Current City (ResLength; in years)
• Dual Income if Married (Dual = 1 if dual income, 0 otherwise)
• Children (Minors = 1 if children under 18 are in the household, 0 otherwise)
• Home ownership (Own = 1 if own residence, 0 otherwise)
• Resident type (House = 1 if residence is a single family house, 0 otherwise)
• Race (White = 1 if race is white, 0 otherwise)
• Language (English = 1 is the primary language in the household is English, 0 otherwise)
So how might MagWeb decide what magazines to market to each person; that is, what magazine ads to put in each email? One way would be to develop a formula/equation that predicts the probability that a customer will buy a particular magazine based on the data that the company has about the customer. Such an equation would be developed for each magazine that the company sells.
If MagWeb has such a model for each magazine that they sell, they can calculate the probability that the customer will buy each of the magazines they offer. MagWeb might also do more complicated things than just look at the predicted probabilities, such as looking at the expected profit from the sales.
In order to be able to develop an equation that predicts the probability that a customer will buy a particular magazine, the company would run an experiment in order to collect data on customer purchase behavior. One way to do this is to randomly select some customers from the customer database and then send them emails with randomly selected ads. Whether or not these customers buy the advertised magazines can provide the data necessary to estimate the equations that will be used to predict the probability that other customers purchase a particular magazine.
So, the problem of deciding what magazine ads to email to each prospective customer depends on developing an equation for each magazine that predicts the probability that a customer will buy. We are now going to focus on developing such an equation for one magazine (“Kid Life”) whose target audience are the parents of children. In the process of sending out the “experimental” emails, the ad for “Kid Life” was shown in emails to customers and the purchase behavior recorded.
In addition to the variables for each customer listed above (the ones obtained from 3rd party sources), MagWeb has the following variables from their own databases:
• Previously purchased a parenting magazine (PrevParent = 1 if previously purchased a parenting magazine, 0 otherwise).
• Previously purchased a children’s magazine (PrevChild = 1 if previously purchased a children’s magazine)
The dependent variable Y comes from the “experiment;” that is, from the emails to customers containing the ad for “Kid Life” and whether or not the customer purchased the magazine. That is, the dependent variable is
• Purchased “Kid Life” (Buy = 1 if purchased “Kid Life,” 0 otherwise)
The experiment data is presented in one of the attached spreadsheet tabs. The sales prospects are presented in another tab.
Goal
Determine which customer prospects should be targeted in the email campaign to maximize expected profit.
Project Assignment
- Prepare the best regression analysis predicting customer prospect buying behavior and explain why you think this is the best statistically sound analysis. Logistic regression would be the appropriate approach for this type of problem.
- The financial metrics that should be used to evaluate the expected profit of each customer prospect is CLV (customer lifetime value) and PLV (prospect lifetime value). Review the PowerPoint presentation on CLV PLV CONCEPTS.pptx.
- Perform market research to collect and estimate any missing data to complete your analysis. Make believe you were starting a new business and had to estimate pricing, revenue and costs associated with this business. This is the more difficult aspect of this analysis and will require that you research marketing data, industry data and financial statements from the school’s library resources.
- Prepare spreadsheet(s), statistical analysis and a PowerPoint presentation showing your approach, assumptions, results and sources of information.
Guidance
• Discuss the case with your team.
• Develop a common understanding of the case and project.
• Plan your work.
o Develop a schedule for your work.
o Assign roles and responsibilities, including “due dates”.
o Research websites, articles, sources of data and use of other spreadsheet models.
o Develop reasonable data assumptions.
o Alternative scenario and sensitivity analysis.
o Use relevant graphics.
• Non-verbatim copying and adapting work of others are permitted, provided that citation of all sources is given in footnotes or bibliography with page numbers and complete direct links to all webpages and internet documents. Besides the webpage links, you must highlight the exact data used (circle or color the data in a screenshot).
• Collaboration may require using communication technologies such as teleconferencing and/or webconferencing if you can’t meet in person.