Pricing research and pricing optimisation

Pricing is one of the more technical areas of market research. The aim is not to find what customers like, but what they are willing to pay and so what the optimum price point is to maximise profit or revenue or market share. There are four main approaches to pricing research, the Gabor-Granger technique, van Westendorp Price Sensitivity Monitor, Brand Price Trade-off and Conjoint Analysis (also known as Discrete Choice Analysis). Some techniques can be used off-the-shelf and many companies sell branded pricing research packages that are just a variation on one of these techniques, however selecting the right technique ultimately depends on what the problem is you are trying to solve.

Market context, positioning and price strategy are also extremely important in setting prices - what are you trying to do with your prices - eg win share or maximise profits? For instance in technology markets, prices typical fall over time. In business markets "value-in-use" or "total cost" may be more important than absolute price.

Price modelling and market models are a fundamental part of pricing research to estimate demand optimum points and competitor responses. Pricing optimum points can differ for maximising sales versus profits for instance.

"I just wanted to say a big thank you for the pricing study work. The findings will certainly help shape our thinking in this area and we now have a view of pricing that far extends our knowledge in the market place."

Strategy Director, Macromedia

Gabor-Granger (direct or likelihood of purchase pricing)

Gabor-Granger pricing research is named after the economists who invented it in the 1960s and is also know as direct pricing. Customers are asked to complete a survey where they are asked to say if they would buy a product at a particular price. The price is changed and respondents again say if they would buy or not. From the results we can work out what the optimum price is for each individual. By taking a sample of customers we can work out what levels of demand would be expected at each price point across the market as a whole (the demand curve in the following graph). Using this estimate of demand, the price elasticity (or expected revenue) can be calculated and so the optimum price-point in the market established. Note that a revenue optimum may be different from a profit optimum. The ability to model dynamically is extremely valuable in pricing studies to estimate revenue and profit effects.

Gabor-Granger output

A weakness of Gabor Granger is that customers may understate the price they will pay (there are also circumstances in which they will overstate the price). Consequently the phrasing of the "would you buy" question is extremely important as are other contextual questions to place the customer in the buying frame of mind. Typically, Gabor Granger is only used when considering one product in isolation, whereas in real life they would face a choice about which product to buy. Competitive response to different prices cannot be gauged from Gabor Granger and knowing customers what would pay is useful, but not if competitors are offering the same product for less. There are a variety of ways of asking the questions including asking for a price directly, or asking for a rating of likelihood to buy. Likelihood to buy results have to be weighted to try to produce an estimate of take-up as they commonly overestimate potential demand.

Van Westendorp

A more sophisticated variation of direct pricing is called Van Westendorp price sensitivity monitor. Respondents are asked four questions to determine what prices are too cheap, where a price is a bargain, when a price is expensive and where a price is too expensive. By plotting the cumulative curves for each of the four prices, the crossing points are deemed to be optimum points - for instance where the expensive and bargain price curves cross The resultant price "space" helps to determine the range of acceptable prices - and so pricing tactics - available. This is a technique which is more for price positioning type studies than for estimating optimum pricing. As with the Gabor Granger, there is no competitive element and it assumes respondents know the market. It also gives no direct measure of likelihood to buy, so the van Westerndorp is often combined with direct pricing questions, or with a conjoint.

Conjoint analysis

The main major technique for pricing is based on Conjoint Analysis and is more sophisticated and more reliable than other research techniques. Conjoint is excellent at looking at understanding how choices are made and consequently the importance of price. For some, conjoint analysis is the only way of carrying out pricing research, and in particular Discrete Choice Analysis (a subset of conjoint) is often used to estimate price elasticities for brands in supermarket style layouts. However, conjoint analysis is a more technical form of research and requires higher levels of design skills. If pricing is to be conducted it is often advantageous to include it as part of a broad conjoint study into product and service features.

In conjoint analysis, customers trade off price against other product features, or in Discrete Choice Analysis, price against brand alone. By looking at how customers make decisions, economic impact of price changes can be assessed as can 'balanced-value' positions for price positioning

A key output from the conjoint analysis is not just what the measures of price sensitivity are, but also a market model that can be used to investigate both what the optimum is if nothing changes, but can also investigate competitor response and potential profitability by building in fixed and variable costs. In more dynamic pricing models such as transportation or leisure markets, these models can be used by yield managers to help guide ticket 'buckets' for time sensitive pricing

Brand Price Trade-offs (BPTO)

For brand specific studies measures of brand equity and category management Brand Price Trade-off Studies (BPTO) can be used. Here customers evaluate a range of products and prices are adjusted until customers stop purchasing.

Behavioural economics

Most price research approaches assume that pricing is dealt with in a rational manner. Increasingly behavioural economics shows that reactions to prices can be conditioned by other factors and the structure and presentation of pricing options will affect choice. Factors like anchoring (using one price to judge a second price) and framing can be important, particularly in product ranges where there are both price per item, but also relative prices between different products to be considered.

For some markets where prices are very visible, or where there is a large amount of internal pricing data, it is possible to use econometric methods to examine the impact of price and to understand price elasticities. Using pricing tests, discounts and advanced statistical analysis the impact of price can be assessed live in the real world.

The most common approach to pricing research is to rely on market intelligence and follow-my-leader type pricing using a competitor as a benchmark. However a me-too approach leads to high levels of competition, and it is important to consider the strategic impact of pricing as well as the short term sales impact.

Some caution is needed when conducting pricing studies. Statistically speaking, where you are looking to optimise prices where you are looking at relatively small price changes of 5-10%, you will need larger than normal sample sizes to get the statistical accuracy you need. For many companies this can make pricing research expensive, unless combined with a range of other measurement.

It is also important to consider psychological effects like anchoring in price research. The price-range shown and the first values shown can influence perceptions of what is appropriate, or cheap or expensive, particularly in markets where prices are largely unknown such as infrequently purchased or specialist goods.

In general qualitative research is not recommended for pricing studies. Qualitative research can be useful when a price list or price structure has become too complex, but in general when you ask people about prices in a qualitative setting, prices are always too high, or not transparent, and respondents will tend to negotiate with the researcher so it is not possible to produce estimates of demand at different price points.

For help and advice on carrying out pricing research and setting pricing strategies contact


How can we help?

Help with a query Site feedback Contact me