Knowledge Center

ValueCreator� Pricing

by Bianca Heiszwolf, Dirk Huisman

The specific role of pricing (research)
The right pricing is essential for a successful commercial policy. Because of course a price set too low results in a profit which will be lower than is possible and feasible. A price set too high results in a loss in turnover and cost coverage. The issue of price setting is basically very simple: select the price with which 'he margin times the number of sold units' is at it's maximum (figure 1).

Figure 1: The art of pricing (1)

Although the policy problem is easy to formulate, behind the simple question is a complex of factors, which makes giving a reliable answer difficult. Apart from the underlying complexity there is often also a question of a confusion of tongues. In practically all policy models and theories, price is one of the core variables. However, there is a world of difference between the function of price in the commercial policy models and the function of price in consumer behaviour.

In marketing, price is a special mix-variable, for two reasons. Firstly, price translates directly into results through turnover. Secondly, contrary to other mix-variables, price does not add "value". Even more so, in the mind of the consumer, price is set opposite from other mix-variables. In the most simple reflection of consumer choice behaviour, the consumer compares the sum of the value-adding elements (e.g. product characteristics, brand, charisma, image, availability, etc) to price (a value-extracting element). To select a product, the following conditions must be met: the price must fall within the budget (and the acceptable price range) and the perceived added value � price ratio must be larger than 1 (figure 2). Rationally, the consumer chooses the combination with the highest added value � price ratio. In practice, many purchase decisions are made implicitly. Still, this implicit behaviour or automatism is based on a series of rational choices and the learning effect (the confirmation of the choice).

Figure 2: The art of pricing (2)

Price sensitivity research: an overview
The techniques used to measure price sensitivity can be classified using the table below 1 :

Measure price sensitivity techniques
measured variable	Degree of control over the variables: Uncontrolled - Controlled (experiment)
Actual purchasing behaviour	Aggregated sales data Panel and Audit data Scanner data	Test-market "In store" Experiments
Preference, purchase intention (simulated purchasing behaviour)	Purchase intention questions	Simulated choice behaviour Conjoint measurement

ValueCreator pricing is part of the ValueCreator model as presented in August at the Esomar congress 2. The category 'controlled measurement and simulated choice behaviour' falls in the above table. The central idea behind the ValueCreatormodel is that marketers must have a priori insight into the result of their decisions. The decisions regarding product positioning, product profiling, price setting, segmentation and targeting are closely related and cannot be taken independent from each other. This implies that price sensitivity, product sensitivity etc. is measured per individual. For pricing, this means that for each individual it must be known when and how he/she reacts to price changes of different products in different situations. Apart from insight into the sensitivities per individual, insight into the explanation of the sensitivities is also wanted and possible.

As said before, consumer purchasing behaviour is often implicit behaviour and the consumer makes the final decision (the choice from the alternatives) in a split second. Here, he or she will view the total product in a specific context and situation. People choose based on the image they have of the product (including brand) and sometimes including price. Linked to this image, a number of feelings and emotions are stored in the memory of the individual. If we want to measure consumer choice behaviour, we need to do this as realistic as possible and within a certain context. Since 75% of all communications is non-verbal 3 , and since this percentage will only grow in the future, the transference of information when measuring choice behaviour must reflect reality. Thus, you need to show the consumer the choice alternatives in a certain context / situation. One of the stimuli is then the price tag, and the other stimuli are brand, the shape of the bottle, the type, etc.. N.B.: we took shampoo as an example here, since this is understandable for everyone, but we could have taken the examples of a blood glucose meter, a car, a mobile phone, or an investment fund as well. All elements that add value: brand, bottle design, content, type are shown in total. They are shown in such a way that the consumer can see whether it is a 250ml or a 400ml bottle, whether it is Elseve or Pant�ne, whether it is package A or package B, whether it is for type 'dry hair' or type 'greasy hair'. In that image, the price tag is also included. The consumer is repeatedly presented with a number of images (concepts), depending on the type of product and the context, from which to choose (this corresponds with a traditional Choice-Based Conjoint design). It is essential that the interviewee is not primarily focussed on one variable that changes (as with the traditional BPTO surveys), because then his/her choice behaviour becomes explicit, whereas actual choice behaviour is implicit.

By analysing the consumer's sequential choices we get the added (or extracted) value each separate element included in the survey has (e.g. 250ml, 325ml or 400ml, Elseve, Pant�ne, Schwarzkopf, bottle A, bottle B, type dry hair, type greasy hair, price 1, price 2, price 3, etc.) for each individual consumer.

In ValueCreator pricing the market is constructed from the products included in the model. It is therefore crucial to include all important brands/products and possible new brands/products in the survey. When constructing the market, each product is seen as the sum of its characteristics. Since we know for each consumer what the value is of each element, we also know the value of each product, and we can predict which product he/she will choose. By changing the price in the model we can determine how many and which consumers will adapt their choice.

example graphic of price sensitivity

Figure 3: Price sensitivity

The output is the price elasticity of brand/product A, assuming that the other brands will not change their price. In figure 3 is shown that the market is relatively price sensitive (almost all elasticities are larger than 1), but for some brands a small price increase is possible because the buyers of those brands only react above a certain price increase. In addition, we can also analyse which consumers will react. For instance, in the Netherlands we see that for certain fast moving consumer goods the price elasticity is different per type of outlet. Buyers who buy their products at e.g. Albert Heijn appear to be less price sensitive for those products than buyers who buy at Dirk van den Broek or Aldi (figure 4).

Figure 4: Attribute importance per outlet type

The assumption that other brands will not change their prices is not always as realistic. For instance, if other brands go along with a price decrease, not much changes, except the turnovers will decrease instead of increase. For these situations it is therefore advisable for formulate a number of anticipated reactions from the competition. You then in fact formulate a number of scenarios (figure 5): 1. I lower my price and the competition does not follow; 2. Part of the competition follows, etc. For each scenario is shown how the consumer will react.

Figure 5: Scenarios

When comparing the simulation results from a consumer survey to Nielsen data (one year later), it turned out that the changes in market share on outlet type level (Albert Heijn versus Dirk van den Broek versus Aldi, etc.) were predicted very accurately. Only for one outlet type the predictions deviated from reality. On hindsight, this could be explained by the changes that had occurred in this outlet over that year. The largest advantage of ValueCreatorpricing is that effects can be simulated a priori, where we can also simulate with other mix elements. Apart from that, it is essential that we know the sensitivities per individual. Especially in a time when marketing is becoming more and more about 1on1 actions it is important to know what individuals choose and do. That is why the ValueCreatormodel is also the basis for "benefit segmentation".

Concluding
In primary research, where we try to predict behaviour, it is essential that we reflect reality. And that we determine per individual how he/she will react. Only that way one can get "explicit" insight into the change in behaviour, which is mostly implicit.

1) Nagle Thomas.T., The Strategy and Tactics of Pricing, Englewood Cliffs, Prentice Hall, p. 266, 1987.

2) Huisman. D., Simulate to create a winner. ESOMAR Congress "The Race for Innovation", Paris, 1999.

3) Zaltman G., Rethinking Market Research, "Journal of Marketing Research", XXXIV, November 1997.

Applied methodologies

Technical papers

#03

Research Manager, Pharmaceutical Company

#03 "I greatly appreciate SKIM's flexibility and attentiveness to our needs"