Method and apparatus for automatic pricing in electronic commerce

Abstract

An automatic pricing method and apparatus for use in electronic commerce environments is described. Automatic pricing uses live price testing to estimate and measure demand for specific products—taking into account where appropriate, a vendor selected segmentation scheme. The results of live price testing are compared using a vendor selected goal function, e.g. profit maximization, to select a new price. A goal function that balances short term gains versus long term gains based on customer lifetime value is described. The live price testing approach used is designed to minimize losses due to price testing through statistical methods. Additionally, methods for distributing price testing across time so as to avoid problems caused by too many ongoing tests as well as side effects from testing are described. The selected price is a win for both purchasers and vendors as the automatic price will approximate the efficiency of a reverse auction without the inconvenience of the auction format while being goal maximizing for the vendor. For example, a vendor that normally sets prices of items for sale to customers can use embodiments of the invention to great effect.

1 . A method of pricing a product for sale, the method comprising:
 testing each price of a plurality of prices by communicating offers that allow potential customers to purchase said product through the use of devices;   wherein the communicated offers include at least one offer for each price of the plurality of prices;   wherein the devices through which the potential customers may purchase said product are not configured to fulfill orders by providing the product;   gathering statistics generated during said testing about how the potential customers responded to the offers, wherein the statistics include number of sales of the product made at each of the plurality of prices;   using a computerized system to read said statistics to automatically determine, based on said statistics, an estimated outcome of using each of the plurality of prices for the product;   selecting a price at which to sell said product based, at least in part, on the estimated outcome determined by said computerized system; and   after using the estimated outcome to select the price at which to sell the product, communicating offers, to potential customers, of said product at said selected price.   
     
     
         2 . The method of  claim 1  wherein:
 the product includes a group of products, the price of each of which is independent of the prices of the others; 
 prices for products within the group are tested by varying the price for each individual product relative to some product specific reference point by the same proportion at the same time for each product in group; 
 the estimated outcome for each test is determined based on the aggregate results across the entire group; and 
 the step of selecting a price includes selecting the proportional adjustment that produced the best estimated outcome for the entire group, then applying that proportional adjustment to each product specific reference point to set a price for each product that is a member of the group. 
 
     
     
         3 . The method of  claim 2  wherein the product specific reference point for each product in the group is the Manufacturer's Suggested Retail Price (MSRP) of each respective product. 
     
     
         4 . The method of  claim 1  wherein the product is a bundle that includes a plurality of pricing units. 
     
     
         5 . The method of  claim 1  wherein:
 testing each price of a plurality of prices includes:
 sending a first set of electronic messages over a network to the devices; 
 wherein the devices are programmed to communicate offer terms, including the prices contained in the messages received by the devices; 
 wherein said electronic messages include offers, of said product, to be presented to potential customers of said product to allow said potential customers to purchase said product for the prices included in said offers; and 
 wherein the devices are programmed to receive orders for the product based on the offer terms; and 
 
 communicating offers of said product at said second price includes sending a second set of electronic messages over the network, wherein the second set of electronic messages include offers, to be presented to potential customers, of said product at said selected price. 
 
     
     
         6 . The method of  claim 1  wherein the product is a service. 
     
     
         7 . The method of  claim 1  wherein at least one price, of the plurality of prices, is selected for testing based, at least in part, on prices at which one or more competitors are offering the product. 
     
     
         8 . The method of  claim 1  wherein at least one price, of the plurality of prices, is selected for testing based, at least in part, on the Manufacturer's Suggested Retail Price (MSRP) of the product. 
     
     
         9 . The method of  claim 1  wherein the step of testing is initiated by a first party in response to a change in a price at which the product is offered by a second party that is different than the first party. 
     
     
         10 . The method of  claim 1  wherein:
 the plurality of prices include a first price and a second price; and 
 selecting the price at which to sell said product includes:
 determining, based on the statistics generated during said testing, that the first price and the second price produce estimated outcomes that satisfy certain criteria; and 
 selecting to sell said product using the first price, rather than the second price, based at least in part on at least one of:
 a) prices at which one or more competitors are offering the product; and 
 b) a MSRP of the product. 
 
 
 
     
     
         11 . The method of  claim 10  wherein the certain criteria is that the estimated outcomes produced by the first price and the second price are similar. 
     
     
         12 . The method of  claim 10  wherein:
 the certain criteria is that a confidence level falls below a certain threshold; and 
 the confidence level is an estimated likelihood that one of first price and the second price will produce a superior outcome than the other of the first price and the second price. 
 
     
     
         13 . A method for pricing a product for sale comprising:
 a computerized system receiving an indication of a deadline for selling a quantity of the product; and   the computerized system sampling demand for the product at a plurality of distinct prices, and automatically selecting a highest price which, based on results of sampling demand, will result in the quantity of the product being sold by the deadline.   
     
     
         14 . The method of  claim 13  wherein the quantity is all of said product that is in stock in a particular inventory, and the deadline is a time by which the inventory of said product is to be liquidated. 
     
     
         15 . The method of  claim 13  wherein the step of sampling demand comprises:
 testing each price of the plurality of distinct prices by sending a first set of electronic messages over a network to devices programmed to communicate offer terms, including the prices contained in the messages received by the devices; 
 wherein said electronic messages include offers, of said product, to be presented to potential customers of said product to allow said potential customers to purchase said product for the prices included in said offers; 
 wherein the devices are programmed to receive orders for the product based on the offer terms; 
 wherein the devices are not configured to fulfill orders by providing the product; and 
 wherein each price of said plurality of distinct prices is used in the offer associated with at least one electronic message in said first set of electronic messages. 
 
     
     
         16 . A method for pricing a product for sale to a particular customer, comprising:
 a computerized system sampling demand for the product at a plurality of distinct prices;   the computerized system generating an estimated probability of return business, for the particular customer, based on a model of average behavior of customers and historical information about prior purchases of the particular customer; and   selecting a price at which to offer the product to the particular customer based, at least in part, on both (a) results of sampling the demand, and (b) the estimated probability of return business.   
     
     
         17 . The method of  claim 16  wherein the model uses different probabilities of return for different stages of a customer's life cycle. 
     
     
         18 . The method of  claim 16  wherein the estimated probability of return business of the particular customer is generated based, at least in part, on timing of prior purchases of the particular customer and timing of repeat purchases of other users. 
     
     
         19 . The method of  claim 18  wherein the model does not treat as a repeat purchase any purchase by a customer that is made after more than a threshold period of time has elapsed since all prior purchases by the customer. 
     
     
         20 . One or more non-transitory computer-readable media storing instructions which, when executed by one or more processors, cause performance of a method of pricing a product for sale, the method comprising:
 testing each price of a plurality of prices by communicating offers that allow potential customers to purchase said product through the use of devices;   wherein the communicated offers include at least one offer for each price of the plurality of prices;   wherein the devices through which the potential customers may purchase said product are not configured to fulfill orders by providing the product;   gathering statistics generated during said testing about how the potential customers responded to the offers, wherein the statistics include number of sales of the product made at each of the plurality of prices;   using a computerized system to read said statistics to automatically determine, based on said statistics, an estimated outcome of using each of the plurality of prices for the product;   selecting a price at which to sell said product based, at least in part, on the estimated outcome determined by said computerized system; and   after using the estimated outcome to select the price at which to sell the product, communicating offers, to potential customers, of said product at said selected price.   
     
     
         21 . The one or more non-transitory computer-readable media of  claim 20  wherein:
 the product includes a group of products, the price of each of which is independent of the prices of the others; 
 prices for products within the group are tested by varying the price for each individual product relative to some product specific reference point by the same proportion at the same time for each product in group; 
 the estimated outcome for each test is determined based on the aggregate results across the entire group; and 
 the step of selecting a price includes selecting the proportional adjustment that produced the best estimated outcome for the entire group, then applying that proportional adjustment to each product specific reference point to set a price for each product that is a member of the group. 
 
     
     
         22 . The one or more non-transitory computer-readable media of  claim 21  wherein the product specific reference point for each product in the group is the Manufacturer's Suggested Retail Price (MSRP) of each respective product. 
     
     
         23 . The one or more non-transitory computer-readable media of  claim 20  wherein the product is a bundle that includes a plurality of pricing units. 
     
     
         24 . The one or more non-transitory computer-readable media of  claim 20  wherein: testing each price of a plurality of prices includes:
   sending a first set of electronic messages over a network to the devices;   wherein the devices are programmed to communicate offer terms, including the prices contained in the messages received by the devices;   wherein said electronic messages include offers, of said product, to be presented to potential customers of said product to allow said potential customers to purchase said product for the prices included in said offers; and   wherein the devices are programmed to receive orders for the product based on the offer terms; and   
 communicating offers of said product at said second price includes sending a second set of electronic messages over the network, wherein the second set of electronic messages include offers, to be presented to potential customers, of said product at said selected price. 
 
     
     
         25 . The one or more non-transitory computer-readable media of  claim 20  wherein the product is a service. 
     
     
         26 . The one or more non-transitory computer-readable media of  claim 20  wherein at least one price, of the plurality of prices, is selected for testing based, at least in part, on prices at which one or more competitors are offering the product. 
     
     
         27 . The one or more non-transitory computer-readable media of  claim 20  wherein at least one price, of the plurality of prices, is selected for testing based, at least in part, on the Manufacturer's Suggested Retail Price (MSRP) of the product. 
     
     
         28 . The one or more non-transitory computer-readable media of  claim 20  wherein the step of testing is initiated by a first party in response to a change in a price at which the product is offered by a second party that is different than the first party. 
     
     
         29 . The one or more non-transitory computer-readable media of  claim 20  wherein:
 the plurality of prices include a first price and a second price; and 
 selecting the price at which to sell said product includes:
 determining, based on the statistics generated during said testing, that the first price and the second price produce estimated outcomes that satisfy certain criteria; and 
 selecting to sell said product using the first price, rather than the second price, based at least in part on at least one of:
 a) prices at which one or more competitors are offering the product; and 
 b) a MSRP of the product. 
 
 
 
     
     
         30 . The one or more non-transitory computer-readable media of  claim 29  wherein the certain criteria is that the estimated outcomes produced by the first price and the second price are similar. 
     
     
         31 . The one or more non-transitory computer-readable media of  claim 29  wherein:
 the certain criteria is that a confidence level falls below a certain threshold; and 
 the confidence level is an estimated likelihood that one of first price and the second price will produce a superior outcome than the other of the first price and the second price. 
 
     
     
         32 . One or more non-transitory computer-readable media storing instructions for a method of pricing a product for sale, the method comprising:
 a computerized system receiving an indication of a deadline for selling a quantity of the product; and   the computerized system sampling demand for the product at a plurality of distinct prices, and automatically selecting a highest price which, based on results of sampling demand, will result in the quantity of the product being sold by the deadline.   
     
     
         33 . The one or more non-transitory computer-readable media of  claim 32  wherein the quantity is all of said product that is in stock in a particular inventory, and the deadline is a time by which the inventory of said product is to be liquidated. 
     
     
         34 . The one or more non-transitory computer-readable media of  claim 32  wherein the step of sampling demand comprises:
 testing each price of the plurality of distinct prices by sending a first set of electronic messages over a network to devices programmed to communicate offer terms, including the prices contained in the messages received by the devices;
 wherein said electronic messages include offers, of said product, to be presented to potential customers of said product to allow said potential customers to purchase said product for the prices included in said offers; 
 wherein the devices are programmed to receive orders for the product based on the offer terms; 
 wherein the devices are not configured to fulfill orders by providing the product; and 
 wherein each price of said plurality of distinct prices is used in the offer associated with at least one electronic message in said first set of electronic messages. 
 
 
     
     
         35 . One or more non-transitory computer-readable media storing instructions for a method for pricing a product for sale to a particular customer, the method comprising:
 a computerized system sampling demand for the product at a plurality of distinct prices;   the computerized system generating an estimated probability of return business, for the particular customer, based on a model of average behavior of customers and historical information about prior purchases of the particular customer; and   selecting a price at which to offer the product to the particular customer based, at least in part, on both (a) results of sampling the demand, and (b) the estimated probability of return business.   
     
     
         36 . The one or more non-transitory computer-readable media of  claim 35  wherein the model uses different probabilities of return for different stages of a customer's life cycle. 
     
     
         37 . The one or more non-transitory computer-readable media of  claim 35  wherein the estimated probability of return business of the particular customer is generated based, at least in part, on timing of prior purchases of the particular customer and timing of repeat purchases of other users. 
     
     
         38 . The one or more non-transitory computer-readable media of  claim 37  wherein the model does not treat as a repeat purchase any purchase by a customer that is made after more than a threshold period of time has elapsed since all prior purchases by the customer.