SELECTING PRODUCTS FOR CATEGORY APPRAISAL STUDIES — FEWER PRODUCTS DO ALMOST AS WELL AS MANY PRODUCTS

This paper analyzes the results of different strategies for selecting test products in category appraisals. The three strategies are random selection of products from the marketplace, selection on the basis of consumer sensory data, and selection on the basis of expert panel data. All three methods generate stable results for category appraisals. The stability of the results (e.g., in terms of 'drivers of liking') increases very quickly and then levels off, suggesting that the researcher does not have to work with a particularly large number of products in a category appraisal to understand the sensory-liking dynamics.
It appears in our case, that 18 products were an appropriate number to strongly reflect results for 50 products. Researchers need to be sensitive to the variation in each category. Some categories may require more or less than this number of products to cover all the sensory ranges within the specific category.