When working in the catalog industry and a customer calls in and wants to order a red sweater and you are out of red sweaters, the company might have just lost the sale if the customer does not want a substitute colored sweater. This is the part of the continuous problem that L.L. Bean, Inc. has with item forecasting and inventory management. Working in a catalog business really helps companies to capture demand, but the problem most companies have is matching demand with supply. Every sale that is generated for L.L. Bean is by customers that want a particular item and if that item is not available, they lose the sale. Customer behavior is hard to predict which affects the demand level of all the products. The double whammy for L.L. Bean is that annual costs associated with lost sales and backorders are about $11 million and costs associated with having the wrong inventory is an additional $10 million.
L.L. Bean, Inc was established in 1912 by Leon Leonwood Bean when he invented the Maine Hunting Shoe, which was a combination of lightweight leather uppers and rubber bottoms. He was able to obtain a list of non-resident Maine hunting license holders and he proceeded to set up a mail order circular for the license holders. When he passed in 1967, the company had 200 employees, a distributed catalog to over 600,000 people, and sales of $4.75 million. L.L. Bean’s golden rule is “Sell good merchandise at a reasonable profit, treat your customers like human beings, and they’ll always come back for more.” By 1990 catalog sales had jumped to $528 million and L.L. Bean had a retail store in Freeport that brought in an additional $71 million. The major competitors are Land’s End, Eddie Bauer, Talbot’s, and Orvis but L.L. Bean was still the highest in overall satisfaction in 1991. They have decided to stay away from the retail-based operations that some of their competitors have gotten into because L.L. Bean management thought it would be too tough to assemble one management team for catalog and retail sales because they were two different entities that required specific skills. The problem that L.L. Bean has is forecasting demand to create inventory of their items without creating backorders and lost sales or having the wrong inventory.
Using Past Demand to Forecast
L.L. Bean, Inc. uses a couple different ways to determine how many units of an item to stock. The first thing that L.L. Bean does is to develop a preliminary item forecast by book and past demand. When a new item is created, they have to make a judgment on whether or not it will justify incremental demand or steal demand from other products. These forecasts are then utilized with the A/F ratios (the ratio of actual demand to forecast demand) for each item in the previous year which determines the range of inventory needed. For example, if 50% of the forecast errors for new items in the past year have been between .7 and 1.6, it would then be assumed that the new products would fall in the same range. If the preliminary forecast was for 1,000 items, it would then be assumed, with a probability of .5, that the stock needed would fall between 700 to 1,600 units. Finally, L.L. Bean estimates the commitment quantity by determining the contribution margin against the liquidity cost. This is where the probability of items sold are calculated versus not sold, and had to be liquidated. L.L. Bean uses this equation to calculate a fractile of the items probability distribution of demand. This also is important to the number of items stocked as to avoid overstocking or under stocking.
Relevant Item Costs and Revenues in Order to Stock
There are many costs and revenues that are relevant to the decision of how many items to stock. The first and most important part of the puzzle for L.L. Bean is what it will cost them to buy/make the item and they also have to know what each item will sell for....