The predictive accuracy of a widely used pre-test-market model (ASSESSOR) is analyzed. The standard deviation between pre-test-market and test-market shares is 1.99 share points before adjustments for achieved awareness, distribution, and sampling and 1.12 share points after adjustment. Sixty-three percent of those products tested passed the pre-test screen and 66% of these were subsequently successful in test market. A Bayesian decision analysis model is formulated and a “typical” case shows a positive value of information. Although some conditions are identified under which a test market may be bypassed, in the authors’ opinion both pre-test and test-market procedures should be used in all but exceptional situations.
Get full access to this article
View all access options for this article.
References
1.
BurgerP., and GundeeC. H., and LavidgeR. (1981), “COMP: A Comprehensive System for the Evaluation of New Products,” in New Product Forecasting, WindY., and MahajanV., and CardozoR. N., eds. Lexington, MA: Lexington Books, 269–84.
2.
Business Week (1973), “The Rebuilding Job at General Foods,” No. 2294 (August 25), 48–55.
3.
BuzzellR. D., and NourseR. E. M. (1967), Product Innovation in Food Processing 1954–1964.Boston: Division of Research, Graduate School of Business Administration, Harvard University.
4.
De GrootM. H. (1970), Optimal Statistical Decisions.New York: McGraw-Hill Book Company.
5.
EskinG. J., and MalecJ. (1976), “A Model for Estimating Sales Potential Prior to the Test Market,”Proceedings 1976 Fall Educators’ Conference, Chicago: American Marketing Association, 230–2.
6.
GoldJ. A. (1964), “Testing Test Market Predictions,”Journal of Marketing Research, 1(August), 8–16.
7.
GoodmanL. A. (1960), “On the Exact Variance of Products,”American Statistical Association Journal, 55(December), 708–13.
8.
HauserJ. R., and UrbanG. L. (1979), “Assessment of Attribute Importance and Consumer Utility Functions: von Newman-Morgenstern Theory Applied to Consumer Behavior,”Journal of Consumer Research, 25(March), 251–62.
9.
HauserJ. R., and UrbanG. L. (1982), “Prelaunch Forecasting of New Consumer Durables: Ideas on a Consumer Value–Priority Model,” working paper, Alfred P. Sloan School of Management, MIT.
10.
KalwaniM. U., and SilkA. J. (1980), “Structure of Repeat Buying for New Packaged Goods,”Journal of Marketing Research, 17(August), 316–22.
11.
KeeneyR. L., and RaiffaH. (1976), Decision Analysis with Multiple Objectives.New York: John Wiley & Sons, Inc.
12.
KeeneyR. L. (1979), “New Product Success Ratios 1977,” No. 1.
13.
PringleL. G., and WilsonR. D., and BrodyE. I. (1982), “NEWS: A Decision Oriented Model for New Product Analysis and Forecasting,”Marketing Science, 1(Winter), 1–30.
14.
RaiffaH., and SchlaiferR. (1961), Applied Statistical Decision Theory.Boston: Division of Research, Graduate School of Business Administration, Harvard University.
15.
RaoV. R., and WinterF. W. (1981), “A Bayesian Approach to Test Market Selection,”Management Science, 12(December), 1351–68.
16.
ShoemakerR., and StaelinR. (1976), “The Effects of Sampling Variation on Sales Forecasts for New Consumer Products,”Journal of Marketing Research, 13(May), 138–43.
17.
SilkA. J., and UrbanG. L. (1978), “Pre-Test-Market Evaluation of New Packaged Goods: A Model and Measurement Methodology,”Journal of Marketing Research, 15(May), 171–91.
18.
TauberE. M. (1977), “Forecasting Sales Prior to Test Market,”Journal of Marketing, 41(January), 80–4.
19.
The Nielsen Reporter (1971), “New Product Success Ratios,” No. 5.
20.
UrbanG. L. (1968), “A New Product Analysis and Decision Model,”Management Science, 14(April), 501.
21.
UrbanG. L., and HauserJ. R. (1980), Decision and Marketing of New Products.Englewood Cliffs, NJ: Prentice-Hall Inc., 386–4.17.
22.
YankelovichSkelly and White, Inc. (1981), “LTM Estimating Procedures,” in New Product Forecasting, WindY., and MahajanV., and CardozoR. N., eds. Lexington, MA: Lexington Books, 249–68.
