Abstract: In general, this disclosure is directed to presenting eligible promotional items on an offer description page. In some embodiments, the eligible promotional items are personalized based on customer relevancy. One aspect is a method for dynamically presenting items eligible for a promotion on an e-commerce application of a retailer, the method comprising receiving, from a customer user computing device associated with a customer user, a request for an offer description page of the promotion, retrieving the items eligible for the promotion and user interaction data associated with the customer user, ranking the items eligible for the promotion based as least in part on the user interaction data, and generating a user interface for the offer description page presenting the items eligible for the promotion according to the ranking, wherein the offer description page initially displays a subset of the highest ranked items eligible for the promotion.

Abstract: In general, this disclosure is directed to presenting eligible promotional items on an offer description page. In some embodiments, the eligible promotional items are personalized based on customer relevancy. One aspect is a method for dynamically presenting items eligible for a promotion on an e-commerce application of a retailer, the method comprising receiving, from a customer user computing device associated with a customer user, a request for an offer description page of the promotion, retrieving the items eligible for the promotion and user interaction data associated with the customer user, ranking the items eligible for the promotion based as least in part on the user interaction data, and generating a user interface for the offer description page presenting the items eligible for the promotion according to the ranking, wherein the offer description page initially displays a subset of the highest ranked items eligible for the promotion.

Abstract: Product demand forecasting accuracy utilizes partitional clustering of time series data with dynamic time warping. The product demand forecasting disclosed herein is particularly suited to forecasting product demand for products with limited sales data. Time-series sales data of a producs (or group of products) with limited sales data (e.g. a sparse or no time series of sales data) are dynamically time warped with sales data of products, or groups of products, having extensive sales data (e.g., an extensive time series of sales data) to determine a clustering model with an optimal number of clusters and a prototype time series for each cluster in the model. The prototype time series for the cluster in which the product (or group of products) with limited sales data lies is utilized as its product demand forecast.

Abstract: Methods and systems are described for optimizing markdown schedules for clearance items at physical retail stores. For example, price sensitivity of a current inventory item at a physical retail store may be modeled using techniques that account for differences in operating behavior and data availability at physical stores versus online channels for selling items. When the current item is placed on clearance at the physical retail store, a request for a markdown schedule, including goals for the clearance, may be received. The forecasted price sensitivity of the current item may be adjusted based on actual price sensitivity of one or more past clearance items (e.g., based on actual clearance sales data) determined to match the current item. An optimal markdown schedule for the item may then be determined based, at least in part, on the adjusted price sensitivity of the item and clearance goals.

Abstract: Methods and systems for generating an optimal clearance schedule for inventory items in a retail enterprise. A method includes accessing historical sales data for a plurality of inventory items in a department of a retailer. A selection of an item is received and a set of possible discount schedules for the item are generated. Each schedule includes a set of prices and a price duration. A forecasting tool provides a sales performance of the item based on the historical sales data. An optimal clearance schedule is selected from the possible discount schedules.

Abstract: Methods and systems are described for optimizing markdown schedules for clearance items at physical retail stores. For example, price sensitivity of a current inventory item at a physical retail store may be modeled using techniques that account for differences in operating behavior and data availability at physical stores versus online channels for selling items. When the current item is placed on clearance at the physical retail store, a request for a markdown schedule, including goals for the clearance, may be received. The forecasted price sensitivity of the current item may be adjusted based on actual price sensitivity of one or more past clearance items (e.g., based on actual clearance sales data) determined to match the current item. An optimal markdown schedule for the item may then be determined based, at least in part, on the adjusted price sensitivity of the item and clearance goals.

Abstract: Product demand forecasting accuracy utilizes partitional clustering of time series data with dynamic time warping. The product demand forecasting disclosed herein is particularly suited to forecasting product demand for products with limited sales data. Time-series sales data of a producs (or group of products) with limited sales data (e.g. a sparse or no time series of sales data) are dynamically time warped with sales data of products, or groups of products, having extensive sales data (e.g., an extensive time series of sales data) to determine a clustering model with an optimal number of clusters and a prototype time series for each cluster in the model. The prototype time series for the cluster in which the product (or group of products) with limited sales data lies is utilized as its product demand forecast.

Abstract: Methods and systems for managing an item assortment from among a collection of heterogeneous items are disclosed. One method includes receiving item data associated with the collection of heterogeneous items that defines values for a plurality of item attributes, and calculating a score for a degree of substitutability between items. A community detection algorithm is applied to edge weights that are based on the scores between items, to identify substitution groups among the items. Preferred attributes common to items within the substitution groups are found, and an item assortment is updated based on a determination of substitutability among items in at least one of the substitution groups.

Abstract: A system and method for tuning markdown plans is provided. Such a system and method may include configuring initial rule set. Initial Demand models are generated. A first optimization for inventory pricing may be received from the price optimization system. The first optimization uses the initial demand models and cost data. A markdown plan is generated by applying the initial rule set to the first optimization. The plan is implemented. Updated data may be received which mandates a re-optimization of the plan. Demand models are refreshed using the updated data. Initial rule set is updated by cross referencing plan history with the initial rule set and subtracting rule events that have previously occurred. A second optimization is received which uses the refreshed demand models and cost data. Then, the markdown plan is re-optimized by applying the updated rule set to the second optimization. The re-optimized markdown plan is reported, approved and implemented.