Our advanced techniques help you understand and predict consumer behavior in complex business environments. We maintain strong ties with academic researchers to ensure that our methods remain "state of the science" and are purposefully designed to provide actionable information.

Each project is approached individually. If your marketing problems do not fit neatly into one of our branded statistical or analytic techniques, we draw from our exclusive library of statistical tools to address your specific problem.

As advocates and users of research, we find it frustrating when providers cannot or will not reveal how they collected or analyzed data. We will explain and demonstrate what we plan to do, how we plan to do it, and why. We use no "black boxes" or proprietary tools whose validity you cannot assess for yourself.

Our discrete choice models determine the value or "utility" that consumers ascribe to product or service benefits. These models can help to determine price sensitivity for a product or service, compare pricing strategies (for example, "all you can eat" versus "à la carte"), and identify optimal feature or benefit configurations. We offer two versions:

• Choice-based conjoint analysis: product or service offers are constructed according to an experimental design and presented to respondents in sets of alternatives
• Revealed preference analysis: the model is based on actual observed choices in the marketplace

Key deliverable: Virtual Market Simulator, an easy-to-use interactive decision tool that allows you to perform "what if" scenario analysis.

Products or services that have many options and allow consumers some degree of customization (for example, automobiles, personal computers, and insurance policies) present problems for "conventional" choice-based conjoint analysis. For these situations, Harris Interactive offers a design-your-own-product interface that allows consumers to choose among many options to choose the preferred features of a product or service.

Discrete choice models provide an estimate of potential demand for new products or services. Our Litmus Simulated Test Market System takes the next step by providing sales volume estimates for the first one or two years following a new product or service launch. Works with estimates of consumer demand from discrete choice models as well as from simpler concept tests, and includes a module for marketing plan optimization.

Market segmentation analysis is a key tool for understanding what drives consumers’ purchase behavior and how their values and beliefs affect their behaviors. By identifying distinctive groups of consumers—their characteristics, needs, purchase patterns, attitudes and/or values—segmentation research can help firms develop offerings targeted to the most profitable consumers.

Perceptual maps graphically display the relationship between brands, perceptions on key attributes, and key consumer segments.

Bayesian statistical methods are particularly valuable in choice studies, increasing the power and robustness of such tools as our Virtual Buyer Discrete Choice models, and in situations where data may be "sparse," for instance, when missing data need to be imputed for respondents. Other applications include utility estimation for Design Your Own tools and missing data imputation.

In addition to our branded methodologies, we use a variety of standard statistical tools such as multiple regression, logistic regression, latent class analysis, canonical correlation analysis, factor analysis, discriminant analysis, multidimensional scaling, correspondence analysis, cluster analysis, and structural equation modeling to meet client needs.