Assessor Model for Simulated Test Markets

Firms that produce consumer-packaged goods use simulated test markets to forecast the long-run market share of a new product before introducing it in the marketplace.
This Demonstration shows how the market share estimate varies as a function of the different trial and repeat values available from the simulated test market.


  • [Snapshot]
  • [Snapshot]
  • [Snapshot]


Market share is given by , where represents trial, the percentage of the target market that will eventually try the product, and represents repeat, the long-run share among the triers.
The trial can occur because of awareness and availability of the new product or because of sampling .
The trial due to awareness and availability is computed from , where is trial under complete awareness and availability, is the proportion of the market that will be eventually aware, and is the proportion of stores that will carry the product in the long run.
The trial due to sampling is given by , where is the proportion of the market that will receive a free sample and is the probability of usage of a free sample conditional on receiving it.
Thus .
The repeat component is given by , where is the competitor's switching rate and is the new brand's switching rate.
A. J. Silk and G. L. Urban, "Pre-Test-Market Evaluation of New Packaged Goods: A Model and Measurement Methodology," Journal of Marketing Research, 15(2), 1978 pp. 171–191.
    • Share:

Embed Interactive Demonstration New!

Just copy and paste this snippet of JavaScript code into your website or blog to put the live Demonstration on your site. More details »

Files require Wolfram CDF Player or Mathematica.