Derived / Stated Importance. There is often a difference between what people “say” and what they actually “do”. Thus, uncovering the true drivers of consumer behavior is important. Regression models provide a correlation between stated attitudes and actual service usage, and allow a researcher to calculate the derived importance of key market attributes. Derived importance can be plotted against stated importance to understand which key traits are high on both dimensions.

Discrete Choice Analysis. People choose services based on the value and utility that they place on service attributes. Discrete choice analysis offers respondents a choice of products with different combinations of attributes. Since the choice sets are created and analyzed using powerful modeling techniques, researchers can estimate with precision the relative importance of price as compared to other attributes, such as quality of service.

Discriminant Analysis. This multivariate technique is used to identify what types of traits are best able to discriminate why a person fits into a specific group. Discriminant analysis can be used in conjunction with cluster analysis to gain a better understanding of what types of characteristics or preferences make clusters or segments different from one another. For example, it might be important to know if certain types of attitudes make a person more likely to be in the segment that purchases cheaper health services.

Factor Analysis / Structural Equation Models. Businesses often need to reduce large numbers of consumer attributes into a smaller set of meaningful categories. For example, one might ask numerous questions about why a person prefers a specific brand. Factor analysis can reduce the large set of questions into a smaller set of “factors” that can then be labeled and interpreted. Structural equation models go one step further and allow a researcher to explain an outcome variable (e.g., brand choice) based on sets of factors.

Perceptual Maps. Most market research problems involve numerous attributes and stated rates of usage. Multivariate techniques such as correspondence analysis and multidimensional scaling allow a researcher to summarize multiple dimensions of data in a two dimensional plot. Once graphed, it is easy to search for relationships between variables such as different brands and attitudes, and identify potential market opportunities and challenges.

Price Sensitivity Analysis. Most firms need more information about what prices to set for their services. We use two techniques to address this issue. One is to test realistic price points among consumers to identify specific prices to sell services. The second is to determine an acceptable range of prices based on the perceived value of the item in question. These techniques can be used separately or together depending on the client’s objectives, such as sales penetration to gain market share or to increase revenue and ROI.

Regression Models. Regression models explain a specified outcome using a set of “predictor” variables. These models illustrate the relative importance of various traits in determining marketing outcomes such as brand preference or purchasing behavior. There are many types of regression models that researchers can use depending on the type of data that is being collected.