Choice Modeling: Testing Product and Service Bundles


A supplier of high-end technology products and services was interested in offering its products and services to a broader group of potential customers. In order to develop an effective sales strategy, the supplier needed to understand which product and service configurations were preferred by the mass market. A key objective was to develop bundles of products and services that could then be marketed to consumers who were down-market from the traditional luxury customers. Choice modeling was required in order to produce a list of the most desired services and bundles of services at attractive price points.

Strategic Issues

This client recognized (earlier than most in the industry) the great potential for enhanced electronic conveniences in the mainstream market. In order to stay ahead of the trend, it was vital to understand consumer preferences for these services. They recognized that the mainstream buyer could have different priorities and needs than those of luxury buyers. Separating the must-have functions from the nice-to-have functions and bundling/pricing them in a way that was attractive to mainstream buyers would be essential to the company’s success.

Research Objectives

The primary research objective was to develop a consumer-configured list of preferred products and service bundles to offer to mass and near-luxury consumers. The list was developed from a menu of approximately 20 possible services that would be available to the luxury market.

Research Design and Methods

A choice modeling design was used. An experimental design was optimized to measure consumers’ relative preferences for 20 different services and the impact of package price on demand. This design defined 3 package offers within each of 12 scenarios presented to each online survey respondent. For each scenario, the respondent selected which package was preferred, and then they stated their likelihood to purchase the preferred package.

The respondent’s choice and rating answers were used to estimate each respondent’s service preferences and sensitivity to package price, providing the data to build a market simulation tool. The resulting DecisionSimulator™ included an adjustment for awareness level (based on a marketing plan) and the ability to simulate a product line of 1 to 10 packages. The DecisionSimulator™ was used to run hundreds of market simulations to inform our client’s strategic decisions.


The research revealed 3 tiers of services that were ranked based on consumer preferences. Simulations of thousands of packages of varying sizes (1 to 10 services) suggested the optimal bundles of services. Additional simulations where package price was varied reported the revenue-maximizing price by package size. Realistic market take rates were reported for specific packages at specific prices. The client was able to use the research findings to build an optimal product-line offering.


Choice Modeling Services

If you would like more information about our Advanced Analytics or Choice Modeling Services, please contact Jerry W. Thomas, President/CEO (, or John Colias, Senior Vice President and Director of Advanced Analytics, (, at 1-800-ANALYSIS (262-5974) or 1-817-640-6166.