More than 80% of firms globally use email blasts to reach customers. Are they successful? Despite their widespread use, the impact on online and offline revenues in respect to email ads is not very well understood. Do email ads effects spillover into the “real” brick-and-mortar world? This study states a resounding yes. Researchers also wanted to know if the effect is homogenous or heterogeneous across different consumer segments, and if heterogenuous, which consumer segments are the most affected.
In this study, Mi Zhou, Vibhanshu Abhishek, Edward Kennedy, Kannan Srinivasan, and Ritwik Sinha teamed up to investigate. To conduct the study, they examined a unique, high-dimensional observational dataset from one of the largest retailers in the U.S. This dataset linked each customer’s online behaviors to their item-level purchase records in brick-and-mortar stores. Rather than running experiments to measure the marketing campaign’s effectiveness, which is the trend, researchers here applied cutting-edge statistical methods to improve the robustness and credibility of the causal inferences made from this big data.
These findings have direct implications for marketers to improve their digital marketing strategy design and for policy makers who are interested in the economic impact of prevalent email advertising.
The specific technique is called a a doubly robust estimator (DRE). What it does is incorporate causal estimation and nonparametric machine learning methods to work through observational data. By applying this novel approach, researchers discovered that on average, receiving email ads increased consumer’s spending in brick-and-mortar stores by around $1.49. What’s more, there is a heterogenous affect across a number of ads. Receiving one or two email does not significantly affect a consumer’s offline purchase amount, the study found. But receiving three email ads did increase a consumer’s offline spending by a significant number, approximately $11.82. This finding suggests a reminder effect of email.
Additionally, researchers found that the increased offline sales resulted from increased purchase probability and a wider variety of products being purchased. They also demonstratde that the effect of email ads is heterogeneous across different consumer segments. Interestingly, the effect is highest among consumers who have fewer interactions with the focal retailer recently (i.e., lower email opening frequency).
Read the full working paper.