Case Study 2: Improving E-Mail Marketing Response

Case Study 2: Improving E-Mail Marketing Response

MAT 510

Case Study Overview

A company wants to improve its email marketing process to increase in the response rate to email advertisements. The company decided to study the process by evaluating all combinations of two options of the three factors, which are email headings (Detailed, Generic), email open (No, Yes); and email body type (Text, HTML). These combinations were repeated on two different occasions. The factors and the response rates are summarized in the following table:

In this study, we will conduct a design of experiment and analyze the data shown in the table above. We will also reveal graphs to present the results calculated. Grasping from the analysis, we will then make recommendations on what actions the company can do to increase the response rate to the emails that are being sent out. Also, we will develop a process model for the company.

Design of Experiment

Using the data table, to conduct a design of experiment, we must utilize our resources to analyze primary data. In this case, we will organize the provided information through Microsoft Excel to analyze the response rates and conducting the average through the heading, body type, and whether the email has been opened or not. Please view the results below:

Run Heading Email Open X2 Body X1X2 X1X3 X2X3 X1X2X3 Response Rate 1 Repeat Rate 2 Average 50RR
1 Generic no text + + + _ 46 38 42 5.657
2 Detail no text + + 34 38 36 2.828
3 Generic yes text + + 56 59 57.5 2.121
4 Detail yes text + 68 80 74 8.485
5 Generic no HTML + + 25 27 26 1.414
6 Detail no HTML + 22 32 27 7.071
7 Generic yes HTML + 21 23 22 1.414
8 Detail yes HTML + + + + 19 33 26 9.899
    Heading Email Open X2 Body X1X2 X1X3 X2X3 X1X1X3
Sum+   163 179.5 101 168 152.5 126 145.5
Sum-   147.5 131 209.5 142.5 158 184.5 165
Avg+   40.75 44.875 25.25 42 38.125 31.5 36.375
Avg-   36.875 32.75 52.375 35.625 39.5 46.125 41.25
Effect   3.88 12.13 -27.125 6.375 -1.375 -14.625 -4.875

Using these tables, the best graphical display tool that we should use to present the results of the DOE should be the interaction effect charts. The rationale behind the recommendation is because this approach indicates the effect of each variable and what level of independence a variable is to another variable, according to the course textbook. Upon observing the low and high values of the heading, body, and whether the email is open or not, it shows that that heading type would have a mild interaction effect on the email marketing, whereas the opening of the email and body type having their independence to counteract the hypothesis that there is a dependence on the variables.


It would be recommended that the company should focus on the heading type, given that this has a slight effect on the case study based off of the averages, as shown in the table above. The best strategy for developing a process model for the company to increase the response rate of email advertising would be through a regression model. The regression model also sparks the regression analysis, which analyzes two-level factorial designs, providing the company with estimates of main effects and interaction effects. Regression grabs the high and low average values of the data associated with the variables to determine the interaction and dependency. For this case study and producing continuous positive results, the regression model is the approach that the company has to take for the process.


Hoerl, R., & Snee, R. (2012). Statistical Thinking: Improving Business Performance. Hoboken, New Jersey: John Wiley & Sons, Inc.

Place an Order

Plagiarism Free!

Scroll to Top