Tying Mystery Shop Surveys to Business Outcomes

• Activity Number: 537
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Marketing
• Author(s): Kurt Pflughoeft* and Sharon Alberg
• Companies: MaritzCX and MaritzCX
• Keywords: mystery shop survey ; linkage analysis ; gis ; random forest

Abstract

Mystery Shop surveys provide an important mechanism to determine if corporate outlets are following proper protocols. Though compliance to standards is important, managers often wonder does it impact business outcomes?

Linking Mystery Shop data to revenue is difficult since many other factors impact revenues and most outlets are in compliance. The keys to successful linkage analysis are:

1) Finding the relevant variables so that the model is not misspecified. 2) Employing powerful analysis techniques, such as Random Forests or Boosted Trees, for prediction and variable importance.

In this research, we examine a mystery shop survey for a gasoline chain. Survey variables are combined with the attributes of the gasoline station. External variables related to station's locality, such as number of competitors and average household income, are derived from a Geographic Information System.

The results show that adherence to corporate standards and higher levels of satisfaction positively impact revenue when modeled with external variables.