Which attributes move preference in your category
A preference map crosses two measurements taken on the same samples: the trained panel’s description, attribute by attribute, and how much consumers like each one. The preference drivers come out of that crossing.
With one in front of you, a reformulation stops being argued over opinions: you know which attribute to move, in which direction, up to what level, and which consumer you are optimising for.
How the map is built
The trained panel says what is inside the product. The consumer says how much they like it. Separately, each measurement answers half a question; sensory category mapping puts them in the same table and looks for the relationship between the two.
- The descriptive half
- The panel’s profile, attribute by attribute and on the same scale for every reference in the study. It is done in Sensory profiling.
- The consumer half
- How much those same samples are liked, measured with real consumers of the category in a product test.
One condition settles the whole study before it starts: both measurements are designed at the same time and taken on the same units, served the same way. Kept apart, they stay two sound studies that will never meet in the same table. The two names we give that way of working, SensoConsumer and Dual Approach, are explained in Product test.
The product is described
The trained panel scores every sample attribute by attribute with the Sensory Spectrum™ methodology. This is the objective half of the crossing: what is inside, measured the same way on every reference.
The scale and how it is set →Liking is measured
The same samples go out to real consumers of the category, in a product test. Nobody is asked about technical attributes here: they are asked how much they like it.
The two matrices are crossed
Statistical analysis relates each descriptive attribute to liking. That is where the preference drivers come from — “drivers of liking” in the literature: the attributes that explain liking, and the weight each one carries.
The optimum, not the maximum
An attribute rarely works in a straight line: there is usually a point beyond which more stops helping and starts getting in the way. The map marks the zone where liking is highest, not a direction to keep pushing in without limit.
It also needs range. If every sample is alike there is no variation in which to look for the relationship between attribute and liking, and the map comes out flat: that is why the study almost always includes competitor references alongside your own, and why in many categories part of that range is already measured in the history.
What the map answers
What shows up in the crossing and is in neither measurement on its own.
Which ones carry weight
Of all the attributes the panel describes, a handful explain liking and the rest are only description. The map separates them out and ranks them by weight.
The target range for each
Not just the direction but the level: every driver comes out with the interval where liking is highest. That is the form an R&D manager can take into a concrete formulation.
One market, or several
Average liking hides clusters. When two segments prefer opposite things, the average lands on a midpoint nobody likes. The map picks it up and forces you to decide who you are optimising for.
What arrives, and the order to read it in
- The map
- The category drawn out, with your product and the measured references placed inside it.
- The drivers
- The list of attributes that move liking, ranked by weight, with the target range for each one.
- The clusters
- If segments with different tastes show up, they come with their size and a map of their own, not averaged together.
- The reading
- What to move first, what to leave alone, and which difference the person eating the product will actually notice. This is the part that gets used, and the part that decides whether the study was worth running.
Sorted Napping Test
The drivers map speaks the panel’s language. This other one is drawn by the consumer.
There is no trained panel here and no attributes defined in advance: the consumer groups a large set of products however they see fit and explains in their own words why they put them together. Out comes the category as it exists in the mind of the person buying, which does not always match the company’s own. It tends to be the earlier step: first you understand how the category is ordered, then you measure where your product falls inside it.
- How elastic the category is. How far a product can move without ceasing to belong to its category — according to the consumer, not to an internal definition.
- How a new product fits. Whether a launch is perceived as inside a category that already exists in the market, and which part of it it falls into.
- Whether it opens a category. Or whether, on the contrary, the product is perceived as outside what exists and opens up a newly created category.
With the map done, the question changes
It is no longer what moves preference, but what has to be defended and what can be estimated without measuring again. The whole axis is in Strategic consultancy.
Which category do you want to map?
Tell us the category, how many references are on the table and which decision depends on the result. We tell you what would need measuring and what you will be able to state with the map in front of you.