The design of experiments (DOE) is a systematic way of finding the cause-and-effect relationship between a business process and its output (Goos & Jones, 2011; Anderson & Whitcomb, 2000). In the case study, the e-mail marketing process comprises three two-level factors, namely, heading, email open, and email body as shown in Table 1 below. The design has six treatments (3 by 2), and therefore, a two-level factorial design will be used (Creswell, 2008). The design with the outputs (response rate) is shown in Table 2.
Table 1: The Factors and Associated Levels.
Table 2: Email Marketing Factors and Response Rates.
The ANOVA summary for the Email Response Rate is shown in Table 3 below.
Table 3: ANOVA Summary Table.
The experiment will use interaction plots as the graphical display tool. The rationale for using this tool is that interaction graphs display the effect of the “combined manipulation of the factors” being investigated (Pear, 2000, p. 57). From Table 3, the F-value is about 16.5. Therefore, any factor with an F-value exceeding the critical value (16.47) has a significant effect on the output (Montgomery, 2007). Therefore, one can conclude that ‘E-mail Open’ (F=34.31) and ‘E-mail Body’ (F=23.6) have significant effects on the e-mail response rate. The combinations of ‘open’ and ‘body’ and ‘heading’ and ‘body’ have negative interaction effects on the response. In contrast, the combination of two factors, namely, ‘heading’ and ‘open’, has a positive effect (+0.5) on the response rate.
A successful strategy to improve the response rate is creating emails with personalized subject lines and content. According to the Experian Marketing Services (2013), lifting the response rate requires e-mails that appeal to the personal characteristics and preferences of the target customers. A detailed e-mail heading and an HTML body are preferred to a generic heading and a text body. The rationale for recommending personalized heading and content is that such a strategy would facilitate relevant messaging to increase the likelihood of a response.
In this study, a model combining an open email and an HTML email body would have a significant effect on the response rate. The F-value for ‘Open and Body’ (37.7) is above the critical F-value (16.5). Therefore, the recommended model for increasing e-mail response is an open e-mail (Yes) with an HTML body. The rationale for proposing this model is that an optimal process output (response rate) is achieved when an open e-mail and an HTML body are combined as the critical input factors.
References
Anderson, M. & Whitcomb, P. (2000). DOE Simplified Practical Tools for Effective Experimentation. New York, NY: Productivity Inc. Web.
Creswell, J.W. (2008). Educational Research: Planning, Conducting, and Evaluating Quantitative and Qualitative Research. Upper Saddle River, NJ: Prentice Hall. Web.
Experian Marketing Services. (2013). How Email Marketers are Connecting, Engaging, and Inspiring their Customers. Web.
Goos, P. & Jones, B. (2011). Optimal Design of Experiments: A Case Study Approach. Hoboken, NJ: Wiley. Web.
Montgomery, D.C. (2007). Design and Analysis of Experiments. New York, NY: John Wiley & Sons. Web.
Pear, J. (2000). Causality: Models, Reasoning, and Inference. Cambridge: Cambridge University Press. Web.