Future Paper

Journal of Revenue and Pricing Management (2007) 6, 304–305. doi:10.1057/palgrave.rpm.5160103

The science of selling

E Andrew Boyd1

Correspondence: E Andrew Boyd, Science and Research, PROS, 3100 Main Street, Suite 900, Houston, TX 77002, USA. Tel: +1 713 335 5329; Fax: +1 713 335 8144; E-mail: aboyd@prospricing.com

1E. Andrew Boyd is Chief Scientist and Senior Vice President of Science and Research at PROS. He received his AB with Honors at Oberlin College with majors in Mathematics and Economics in 1981, and his PhD in Operations Research from the Massachusetts Institute of Technology in 1987.

Received 10 July 2007; Revised 10 July 2007.

Top

Abstract

Pricing is often treated separately from the process of selling. When is this a reasonable assumption?

Keywords:

pricing, revenue management, revenue optimisation, sales

When we think of pricing, our minds instinctively jump to the price tag. We imagine a product sitting on a store shelf. People pass by and look at the price. They either buy the product or they do not.

In many cases, this take-it-or-leave-it, or posted, model of pricing is appropriate; for example, when selling boxes of cookies or airline tickets. But in many more cases, the posted price model is wholly inappropriate, since price is only one part of the overall activity of selling. When selling and pricing are intertwined, the models we commonly use for posted pricing are no longer applicable. The entire sales process, of which price is only one factor, needs to be modelled. The issue is not a matter of developing more realistic models, but a matter of logical necessity.

To illustrate this point, consider the posted price application of selling boxes of cookies. If during the course of a year the cookies are offered at different price points, then we can estimate the demand per unit time at various prices and develop some approximation of the demand curve. In practice, factors other than price must be accounted for (advertising campaigns, location on the shelf, seasonality, etc), but at least in theory there is nothing that stops us from estimating a demand curve.

This is not the case when a posted price does not exist. Consider, for example, a car dealership where the actual sales price is determined as a result of negotiation. At the end of any given month, the dealership could easily find that of 100 identical cars that it sold, 25 sold for $30,000, while the other 75 sold for $31,000. The result does not reflect the fact that demand increases with price. It simply shows that negotiations led to more sales at the higher price. The data cannot be used to construct a demand curve and estimate price elasticity in any formal sense.

Many, if not most, business-to-business transactions do not involve posted prices. Buyers may call a potential seller and request a quoted price, and the seller may adjust the price on a call-by-call basis. Sales agents for distributorships may be sent on the road, briefcase in hand, to spend a few hours negotiating prices with potential retailers. Teams of people may work for months hammering out the details of large contracts that establish terms of sale, including price. In each case, price is but one component of the sale, and the conditions necessary to estimate a demand curve and price elasticity simply do not exist.

The pricing and revenue management literature is filled with research on posted price problems. Yet, the literature on actual day-to-day price-related problems where there is no posted price is limited at best. Relevant modelling will require understanding the many different ways goods are sold, recognising that pricing is only part of selling. As we look to the future, we will increasingly find we are not just modelling pricing, but the actual sales process.

Extra navigation

.

Association resources

ADVERTISEMENT