Advanced data science and machine learning can give fashion retailers an accurate view of demand for sizes and styles across each store location
Fast fashion retailers are under an immense amount of pressure to get their in-store product assortments allocated perfectly. While customers expect their desired styles and sizes to be in stock, retailers grapple to minimize excessive inventory and dramatic markdowns.
Part of the newly launched Oracle Retail Science Platform Cloud Service, the Size Profile Science features enable retailers to tackle this problem by helping accurately predict sales to ensure they have the right inventory mix in the right locations.
“Customer expectations are higher than ever and retailers are constantly walking a tightrope, balancing between too much of the wrong inventory and not enough of the right,” said Jeff Warren, vice president, Oracle Retail. “Predicting sales at the size and style level, with assortments changing monthly, creates an exorbitant amount of data. Oracle continues to invest in powerful machine learning and artificial intelligence-based applications that enable retailers to quickly turn that data into usable insights and an action plan when it comes to ordering.”
For example, using the application, a beachwear retailer would be able to analyze proprietary and third-party customer or consumer fit datasets to determine what size flip-flops were top sellers in its specific markets.
Combining that data with machine learning and artificial intelligence, the company would be able to understand what assortment of sizes would be needed to quickly sell at the top price in each location and minimize those sizes that would end up on the sale rack. On a million dollars of inventory, even a five percent reduction in average load would result in $50K in savings, all as a result of putting the right sized products in the right locations at the time of demand.
Business benefits of Size Profile Science include the ability to:
Maintain and even decrease inventory levels necessary to drive additional revenue
Increase customer satisfaction by improving in-stock rates by location
Reduce missed sales by helping ensure optimal sizes are available at time of demand
Decrease workloads through intelligent alerts, exceptions and streamlined parameter setting
Design escalation paths to deliver accurate profiles when useful lower level data is unavailable
“With the amount of choice the customer has today, retailers must be able to effectively understand size and fit preferences to preserve their brand equity and foster loyal shoppers. A targeted assortment plan, as well as an optimized size strategy, will maximize the sell-through of inventory at full price, enabling a significant margin increase,” said Marc Koehler, solution director at Oracle Retail.
