New Internet Pricing Models Bring Pain, and Fortune, to RetailersPublished: February 22, 2001 in Knowledge@Wharton
On the runaway train that is the Internet, companies soon found themselves sharing space with an unexpected – and unwanted – passenger. It was called price transparency, and while it would prove to be bad news for business, it was great news for consumers.
Price transparency refers to the ability of consumers (at both the wholesale and retail level), to know what it will cost to buy a given good or service at a variety of outlets. While prices will vary in any marketplace because of such factors as distribution network, consumer preferences and differing rates of sales tax, the transparency to consumers has been historically limited because constraints like distance and time have hampered their ability to engage in widespread comparison-shopping.
The Internet changed all that and gave consumers the ability to price-shop with the click of a mouse. Businesses, in turn, found that their old, static pricing models were becoming obsolete; they had to formulate a new way to respond to changes in the digital marketplace.
"There are various pricing models," notes Wharton operations and information management professor Eric Clemons. "One is dynamic pricing, a mechanism for matching supply and demand through the pricing structure. This is the model used by the stock exchanges and commodities markets."
Think of the way stock prices rise and fall, sometimes in a matter of seconds. Or consider a lower-tech example, like that of Secaucus, N.J.-based Syms Corp., which operates a chain of off-price apparel stores throughout the U.S. "After 10 days on the sales floor, the price of our women’s daytime dresses (work-casual) decreases," says Douglas C. Meyer, the company’s vice president of marketing. "After 20 days, the price drops again."
Clemons points out that another pricing model, variable pricing, is often used in the consumer segment. "Variable pricing aims to generate incremental sales and revenues by varying the price of an item," he explains. "In this model, the customer has already demonstrated that he or she is not willing to pay the current price for the product, but a manufacturer is betting that after a trial use (often sparked by a steep price discount) the customer’s willingness to pay will be altered."
There are several other ways to implement variable pricing based, for example, on different versions of the same goods or services. Think of day-old bread or other baked goods. Will a consumer buy a stale item? He or she might if it’s offered at half the price of a similar, fresh-baked one.
Properly implemented, variable pricing can result in additional sales without cannibalizing existing ones. But as at least one big-league player, Priceline.com, found out, even the skills of the Starship Enterprise captain aren’t always sufficient to navigate this pathway to profits.
Priceline gained fame by offering online "name your own price" bidding for a variety of goods. In fact, it was often hailed as the poster child of the new economy. But Clemons says the company (which recently lost Star Trek’s William Shatner as its celebrity spokesman) stumbled when it tried to expand its variable pricing model from items like airline tickets to another category, groceries.
He notes that variable pricing will work in some perishables environments, like the bakery that offers "day old" goods at a discount. But those items are "damaged" and do not directly compete with the sale of "premium" (or undamaged) goods. So if someone was going to pass on your bakery’s bread because it was too expensive, he might buy a less-expensive day-old loaf.
It’s worth it for the bakery owner to sell the day-old goods at a discount (perhaps even below the original cost) because otherwise the goods would just be discarded, contributing no revenue at all.
But how about airline tickets? Short of ripping up the upholstery or placing tacks on the cushions, how does a class of seats become "damaged goods" which don’t compete with other seats that could sell at full or near-full price?
In fact, says Clemons, railroad companies in 19th century France did something just like that. "First class passengers paid a premium and were feted with champagne and delicacies," he notes. "Second class passengers received standard seating."
Attracting additional passengers through further discounts would undercut the people who paid full second-class fare, says Clemons (Late last year, Amazon.com got into hot water for doing something similar—offering the same Digital Versatile Discs, or DVDs, to different customers at different prices. Although the company denied it was testing prices based on customer demographics, Amazon ended up refunding 6,896 customers an average of $3 apiece, according to published reports).
In the case of the French railroads, the companies created "damaged goods" by actually ripping off the roof of a rail car and billing it as "third class" seating, which was offered at a discount to the second class price.
Airlines had already addressed some of the seating issues through dynamic pricing "yield management," which involves reducing prices on empty seats proportionately as the day, or even the hour, of the flight draws closer.
But Priceline went a step further by letting consumers name their own price for a flight (subject, of course to the airline’s acceptance), but with a catch—the consumer couldn’t name the exact day, time or origination of the flight. In fact, the passenger might not even get his choice of airline.
"The beauty of this variable pricing model lay in the fact that it created ‘damaged goods’ in the form of uncertainty," according to Clemons. "If you were a student on a tight budget but with time to spare, it might not bother you to drive two extra hours to a distant airport, leave a day earlier than planned and fly back a day later, with layovers each way. On the other hand, a business traveler on a tight schedule would never pass up a regularly scheduled flight for a Priceline bid. So instead of losing a full-pay customer to a discount seat, the airlines could fill an otherwise-empty seat with a passenger who otherwise may not have even flown."
Priceline had a hit on its hands, and saw its stock price hit a record $162 per share in April 1999, shortly after its IPO. Then, near the end of that year, the company decided to tap into a wider consumer market by launching the Priceline WebHouse Club, a service that let online consumers name their own price for groceries.
Perhaps the company should have spoken to Clemons or some of his colleagues before launching the expansion. "Variable pricing doesn’t work for every situation," he says. "While it may attract incremental grocery sales in specific situations, this was the wrong application."
One reason is because unlike financial instruments such as commodities futures, where supply and demand can fluctuate on a second-by-second basis, consumer demand for goods like groceries tends to be stable, enabling retail outlets to predict demand with relative accuracy and place their orders accordingly. So the volume of "damaged goods" like day-old bread may not be great enough to support variable pricing in the grocery segment—at least not on a widespread basis. And as Clemons’ marketing colleagues could have pointed out to Priceline, no consumer packaged goods manufacturer would want to participate in a system that "damaged" their products, their brands or their brand equity.
Priceline found this out the hard way, and in October 2000 announced that WebHouse would shut its doors. Priceline’s market value, along with that of many other Internet-based companies, has since plunged and company shares traded recently at $2.84.
"Companies should do their research and determine where variable pricing will work," comments Clemons. "Supermarkets, for instance, can use their vast database of consumer history to encourage product adoption through variable pricing by launching selective deep price discounts."
He offers a hypothetical shopper as an example. "If a fellow buys laundry detergent on a regular basis, he is probably a good candidate for fabric softener as well," says Clemons. "So as he picks up his purchases at the cash register, it can generate a discount ticket for a fabric softener. If it’s inexpensive enough, the consumer may try the product. Even if the initial sale is a loss leader, the manufacturer wants the non-using customer to try the product."
He adds that this is not news to marketing faculty or to brand managers; but what is new is the ease with which these consumers can be identified and courted.
The Internet has brought challenges, as well as opportunities, to businesses. While it enables a local company to hawk its wares worldwide at an affordable price, it has also forced firms to ramp up their response to shifts in consumer demand. But as the example of variable pricing shows, taking the time to research an approach¾ instead of jumping the gun to implement it¾ will still yield value.