When a company develops a new innovation to make the product better or wants to know how much it can charge for a brand name product over a generic one, the price should change. Thanks
to lessons learned in Professor George John’s pricing class, Carlson School MBA candidates— including Carlson Consulting Enterprise students—develop experience with the tools that put dollars to the value of a product feature.
Key to the process is determining how much a consumer wants to pay for the feature or brand. Gauging consumer reaction to specific features can prove a challenge, as consumers often rate many attributes as equally important. John teaches his students to use a method known as conjoint analysis, in which potential buyers are asked to choose between pairs of choices described along sets of features.
John gives the example of a Polaris ATV with a four-stroke engine, auto transmission, being compared to a Honda ATV with a four-stroke engine and manual transmission. Multiple such pairs are presented to test subjects and the choices are recorded. The basic output is a dollar value attached to a feature. For example, it is worth an additional $200 to have an automatic transmission over a manual transmission, or an additional $100 for the Polaris brand name over the Honda brand name.
“Obviously, these are immensely valuable pieces of information to a firm,” says John, who credits the late Amos Tversky with having developed the underlying theoretical breakthrough for conjoint analysis back in 1967. But in recent years, the tools available for practical implementation have been significantly updated so that the new choice conjoint models are closely tied to economic models of discrete choice, which provide two advantages.
First, the new models make it easy to include the no-buy option. That means consultants can estimate the incremental sales that result from adding (or subtracting) a feature. Second, researchers are able to fit these newer models to supermarket scanner purchase data, such as prices, features and volumes, aggregated to the store level. “For example, we have estimated from such data that about 60 cents out of a $1.50 retail price for a 32 ounce Gatorade bottle is due to its brand name,” says John of data gathered from the Chicago supermarket chain, Dominick’s Finer Foods.
“Our MBAs learn to use both commercial software such as Sawtooth and hand-rolled Excel spreadsheet routines to implement these models in their consulting projects,” says George John. “Their consulting experiences complement marketing courses that stress the underlying concepts and the proper interpretation of the numbers as decision aids. We try to get them to think beyond the current implementation, to focus on the enduring ideas.”
|
