What should L.L. Bean do to improve its forecasting process? In order to improve its forecasting process, L.L. Bean should put more emphasis on researching new products. They should conduct more market surveys, host focus groups, etc. to get a better prediction for new items. When gathering information on products, L.L. Bean must also be more specific. Currently, they seem to use general data for items higher up in the hierarchy, such as item sequences and demand sequences. This could explain the consistency in forecast errors across all managers. By using more specific data that relates to specific items such as a certain color sweater, the company will gain unique and insightful information that relates better to the respective products. Although this will add costs to the company, the revenue gain that could arise out of such could be very profitable. Also, Rol Fessenden noted that he wasn't "entirely convinced that [they] go about estimating contribution margin and liquidation costs correctly" (Case, pg. 4). This should be a main target of focus for L.L. Bean. They must be more precise and put better research into figuring out exactly what these margins and costs are. This doesn't just go for the