Guidelines for Consumer Testing - guidance from ESN members

In this series ESN members give their solutions to the most frequently asked questions from product developers and marketers.


What is the Best Combination for Multiple Products?

With a product line that is successful, many manufacturers extend the product line and introduce different versions or varieties of the product in an attempt to maintain consumer interest and extend product appeal. Most manufacturers have difficulty deciding on the optimum number of product variants to introduce, and on the nature of those variants.

Huguette Nicod, Research Director at the French consumer research company ADRIANT explains: ‘’In a product line with several variants, there are variants that are complementary and variants that compete with each other. Complementary variants are those that the consumers of the product line will buy together or consecutively, and competitive variants are those that will compete with each other for the already loyal consumer. In order to optimize a product line, the total number of consumers who will purchase at least one variant will need to be maximized, and at the same time, the consumer overlap between variants needs to be minimised’’

To help with this question, Huguette Nicod proposes an analysis method known as TURF ( = Total Unduplicated Reach and Frequency)  analysis, a method originally used by media planners. The main objective of TURF analysis, which is a sequential analysis, is to find at each step the product that reaches the greatest proportion of the market not already met by the product line. In essence, consumers are asked to assess a set of potential variants belonging to the product line in a sequential, monadic way through a balanced presentation in one or more session. All consumers evaluate all products and give their ‘purchase intention’ by responding to the question, “After tasting this product, would you ‘definitely buy it’, ‘probably buy it’, might or might not buy it’ ‘probably not buy it’, definitely not buy it?” Only the responses ‘definitely buy it’ are taken into account in the analysis.

Huguette Nicod continues: “The first step of the analysis is to look at the products with the greatest potential, that is, the variant that obtained the best level of ‘definitely would buy it’. This variant would be the first variant in the product line. In the second step, the consumers who were attracted to this product are eliminated for the analysis and the percentage of consumers who will ‘definitely buy’ the other variants is recalculated. These percentages are based on the original total population of consumers, and not the new recalculated sample of consumers. The second variant of the product line is the variant that obtains the best score in this second step. The next steps are analysed in the same way.”

A concrete example of how this operates can be given in the case of yoghurt, where a manufacturer has the possibility for 10 new variants but the potential to market only 5 of these. The TURF algorithm identifies the optimum product line by maximizing the number of consumers who will purchase at least one variant, at the same time minimizing consumer overlap between flavours. The consumers to be tested should be regular consumers of the product and have no aversions against  the product flavours, aromas, or other product varieties to be presented in the test.  A number of 100-150 test participants is recommended.

Regarding the objective of the potential marketing, the best combination of product variants is H,M,O and J, but this combination will only reach a 72% of the potential market. Two further variants will increase this to 85%.Huguette Nicod comments further: “This method is based on the hypothesis that once consumers are satisfied with a specific variant, they will no longer buy other variants of this product. This is not the case for many product categories. To compensate for this limitation, another method consists of calculating, the number of people who will ‘definitely buy’ all variants of each particular combination. For a given number of variants, the best combination is the ones that obtain the best scores. Furthermore, this analysis does not distinguish between consumers who will frequently buy the product from consumers who will only occasionally buy the product. An additional question on frequency of use can be added to address this question.”


For further information on consumer test methods to optimise the best combination of variants in a product line, please contact

Huguette Nicod

Atlanpôle, rue Pierre Adolphe Bobierre
BP 62303
44323 Nantes Cedex 03