Data processing: financial – business practice – management – or co – For cost/price
Reexamination Certificate
2001-05-04
2003-04-22
Cosimano, Edward R. (Department: 3629)
Data processing: financial, business practice, management, or co
For cost/price
C703S006000, C705S007380
Reexamination Certificate
active
06553352
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates in general to the field of econometrics, and more particularly to an apparatus and method for determining optimum prices for a set of products within a product category, where the optimum prices are determined to maximize a merchandising figure of merit such as revenue, profit, or sales volume.
2. Description of the Related Art
Today, the average net profit generated chains and individual stores within the consumer products retail industry is typically less than two percent of sales. In other words, these stores make less than two dollars profit for every one hundred dollars in revenue. Stores in this industry walk a very fine line between profitability and bankruptcy. Consequently, in more recent years, those skilled within the merchandising arts have studied and developed techniques to increase profits. These techniques are geared toward the manipulation of certain classes of merchandising variables, or “levers.” In broad terms, these merchandising levers fall into five categories: price (i.e., for how much a product is sold), promotion (i.e., special programs, generally limited in time, to incite consumers to purchase particular products), space (i.e., where within a store particular products are displayed), logistics (i.e., how much of and when a product is ordered, distributed, and stocked), and assortment (i.e., the mix of products that are sold within a chain or individual store). It has long been appreciated that manipulating certain attributes within each of these “levers” can result in increased sales for some products, while resulting in decreased sales for other, related products. Therefore, it is no surprise that managers within the consumer products merchandising industry are very disinclined to make any types of changes without a reasonably high confidence that the changes will result in increased profits. The margin for error is so small that the implementation of any wrong decision could mean the difference between a profitable status and an unprofitable status.
Ad hoc methods for manipulating merchandising variables in order to increase profits have been employed for years within the industry. And a whole system of conventional wisdoms regarding how to manipulate certain levers has developed, to the extent that courses of undergraduate and graduate study are offered for the purpose of imparting these conventional wisdoms to future members of the industry. For example, category managers (i.e., those who are responsible for marketing a category of related products within a chain of stores) are inclined to believe that high-volume products possess a high price elasticity. That is, the category managers think that they can significantly increase sales volume for these products by making small price adjustments. But this is not necessarily true. In addition, category managers readily comprehend that products displayed at eye level sell better than those at floor level. Furthermore, it is well known that a store can sell more of a particular product (e.g., dips and salsa) when the particular product is displayed next to a complementary product (e.g., chips). Moreover, ad hoc psychological lever manipulation techniques are employed to increase sales, such as can be observed in some stores that constrain the values of particular price digits (e.g., $1.56 as opposed to $1.99) because conventional insights indicate that demand for some products decreases if those products have prices that end in “9.”
Although experiential lessons like those alluded to above cannot be applied in a deterministic fashion, the effects of manipulating merchandising variables can most definitely be modeled statistically with a high degree of accuracy. Indeed, there is a quantifiable relationship between each of these merchandising levers and consumer demand for a product, or group of products, within a store, or a group of stores in a retail chain. And the relationship between these levers and consumer demand can be accurately modeled, as long as the modeling techniques that are employed take into account a statistically sufficient number of factors and data such that credible and unbiased results are provided. Examples of these factors include price and sales history as a function of time (e.g., day of the week, season, holidays, etc.), promotion (e.g., temporary price reductions and other promotional vehicles), competition (e.g., price and sales history information for directly competitive products that are normally substitutes), and product size variations. Those skilled within the art typically refer to a model as is herein described as a demand model because it models the relationship between one or more merchandising levers and consumer demand for a group of products.
The degree to which demand for a particular product is correlated to a particular lever is called its “lever elasticity.” For example, a product with a low price elasticity can undergo a significant change in price without affecting demand for the product; a high price elasticity indicates that consumer demand for the product is very susceptible to small price variations.
Demand models are used by product category mangers as stand-alone models, or as part of an integrated demand/price model. In the stand-alone application, a category manager inputs potential prices for a product or product group, and the stand-alone model estimates sales for the product or product group. Accordingly, the category manager selects a set of prices to maximize sales of the product or product group based upon outputs of the stand-alone demand model. An integrated demand/price model typically models demand within a set of constraints provided by the category manager for a product or group of products and establishes an optimum price for the product or group of products based partially upon the price elasticity of the product or group of products and the objectives of the model analysis.
Notwithstanding the benefits that category managers are afforded by present day demand/price models, their broad application within the art has been constrained to date because of three primary limitations. First, present day demand/price models do not take into account the costs associated with providing a product for sale. That is, the models can only determine prices as a function of demand to maximize sales, or revenue. But one skilled in the art will appreciate that establishing product prices to maximize revenue in an industry that averages less that two percent net profit may indeed result in decreased profits for a retailer because he could potentially sell less high-margin products and more low-margin products according to the newly established product prices. Hence, determining a set of prices based upon demand alone can only maximize volume or revenue, not profit. And profit is what makes or breaks a business. Secondly, present day demand/price models typically estimate price elasticity for a given product or product group without estimating how changes in price for the product or product group will impact demand for other, related products or product groups. For instance, present day demand/price models can estimate price elasticity for, say, bar soap, but they do not estimate the change in demand for, say, liquid soap, as a result of changing the prices of bar soap. Consequently, a soap category manager may actually decrease profits within his/her category by focusing exclusively on the prices of one subcategory of items without considering how prices changes within that one subcategory will affect demand of items within related subcategories. Finally, it is well appreciated within the art that present day statistical techniques do not necessarily yield optimum results in the presence of sparse and/or anomalous data.
Therefore, what is needed is a technique that enables a user to configure and execute optimization scenarios within a model that determines optimized prices for products within a product category, where the model considers the cost of the products as well as t
Delurgio Phil
Neal Michael
Cosimano Edward R.
Demand Tec Inc.
Huffman Richard K.
Lim Kang
LandOfFree
Interface for merchandise price optimization does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Interface for merchandise price optimization, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Interface for merchandise price optimization will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3031952