BEHAVIORAL SEGMENTATION AND TARGETING AT ATX OUTERWEAR

The marketing director at ATX Outerwear, a small men's outdoor apparel retailer, is considering reaching out to existing customers in the company's database to offer customers the opportunity to purchase items from the company's Spring 2023 catalog. The database contains purchase history data for 96,512 customers, including how long since the customer's last transaction, how frequent the customer purchases (in a window of the last 24 months), and the monetary value of purchases (in a window of the last 24 months). The director is considering mailing the catalog to either the entire database or a subset of customers from the database. The cost of reaching a customer via a catalog mailing is $0.80. She decides to run an experiment to help choose which strategy to pursue. First, she has a database programmer conduct an RFM segmentation based on quintiles of the recency, frequency, and monetary spend data. This results in each of the 96,512 customers being assigned a 3-digit RFM code between 111 and 555 (see the class notes for more on this approach). She then takes a random sample of about 10% of the database resulting in a sample of 9,565 customers and mails each customer in this sample a catalog. After 4 weeks she observes which customer have made a purchase. The average spend on purchases made is $104. ATX Outerwear earns a 50% profit margin on purchases.

The Excel dataset ATX_Outerwear_Market_Rollout_Data.xlsx contains the data for the market test customers (9,565 customers) and the rollout customers (86,947 customers). The market test data contains the customer ID, a binary indicator (labelled "buy") of whether or not the customer in the test market placed an order, and the RFM code. The rollout data contains the customer ID and the RFM code. Using these data please answer the following questions.

1) What is the overall average buy rate among the market test sample? Provide a neatly formatted bar chart of the average response rate by 3-digit RFM code for the market test data.

2) Assuming the director decides to mail to all 86,947 customers in the rollout sample, how many customers can she expect to make a purchase?

3) What is the response rate required to break even on the mailing costs? Using this information, how many of the 86,947 customers will be targeted to receive a catalog?

4) Compute the expected profits from mailing to all 86,947 customers in the rollout data vs. the expected profits from mailing to your targeted subset of rollout customers. How well did you do? Provide some explanation for why your targeted approach differs from the approach of mailing to all customers.