Views from the Hills ::

October 12, 2005 - by Robert E. Stevens, GENESIS II(The Second Beginning) E-Mail: views@aol.com

The above was a question I received from one of the readers of the Views. The reader was inquiring about the frequency of using taste test panelists. My response was that it is not a simple question. First, it depends on the purpose for which we are conducting the research. Basically, is the basis for the research Quality Control or Consumer Acceptance? Are we classifying flavors or are we evaluating flavors? Where and how is the research conducted? Is there a learning experience that the participants go through?

The acceptability of the protocol as well as the frequency of testing depends on many factors, including corporate resources, budgeting and most of all, the stage of the research. Is it in the exploratory, screening, brand development or final judgment phase?

All of the above questions have an effect on the validity of the data. It is not a simple question.

Unfortunately, my perspective is from that of a manager and not a practitioner. In my role at P&G, we maintained two primary resources for sensory evaluation. Our first resource was internal, P&G employees. We developed experts in odor, flavor and color evaluation. They were trained and constantly evaluated as to their ability to discriminate. They were utilized for quality control and sensory development.

In the second resource we utilized untrained panelists. Within our corporate resources, we had consumer panels of 600 Female Heads of Households in two or three cities every week. We worked with those consumers for a period of five weeks and then moved to different cities and would not return for at least a year to the same city, let alone the same group of panelists. These panelists were used for consumer acceptability, sensory and use testing.

From my perspective, we use:

Trained panelists to determine degrees of differences and the ability to describe those differences in actionable terms.
Untrained panelists to represent the typical consumer to determine acceptability and preferences.

I can understand the need to spend the time, effort and money to train the sensory experts but I cannot see the need to establish ongoing panels of participants that are to represent the typical consumer. It seems that it is obvious that at some point these participants will no longer be representative of the typical Female Head of the Household. Their experience will take them into an area between the typical consumer and the expert. At this stage, who do they really represent and how do you use their data? But most disturbing is the fact that the availability of the untrained consumer is so easy and cheap to obtain, why would anyone risk the re-utilization of participants that are to represent the typical consumer?

How frequently can I use my panelists?