Demand forecasting has taken on increased importance now that retailers nearly across the board are carrying less inventory at store level. While the practice protects chains from getting caught with overstocks and helps free up capital tied in non-moving goods, it also carries a danger in that the limited assortments could fail to meet consumer demand.
“It is critical to have the correct styles and sizes available in such a volatile marketplace,” said Dennis Hernreich, EVP, COO and CFO, Casual Male Retail Group, Canton, Mass., which operates more than 500 stores and is the nation’s largest specialty retailer of big and tall men’s apparel. “The only way to accomplish this is to have good practices and solutions centered on demand forecasting at every level.”
To help lower inventory levels and uphold profitability, Casual Male is boosting its demand forecasting efforts with business intelligence. The chain now analyzes past demand as well as customer shopping behavior and sales occurring in all of its business channels. This is a departure from the past, when the chain, like many specialty retailers, had a systemic forecasting process that analyzed demand based on size and style, but not specific SKUs.
“We weren’t validating the lowest level of our inventory plan, and this didn’t properly quantify the fast-paced ‘fashion’ side of our business,” Hernreich explained. “Units within this category are based on specific launch and end dates, and we must accurately forecast what the sales trajectory will be.”
This practice got easier for Casual Male when it applied a BI tool from QuantiSense, Atlanta, to its forecasting practices last year. The chain first loads all inventory information, such as merchandise class, and sub-class information, including retail channel sales, into its JDA Allocation solution from Scottsdale, Ariz.-based JDA Solutions.
Then the BI tool pulls this data and drills down to SKU level information, such as style, color and size, as well as item movement and shopper demand across retail channels. The amount of merchandise that the chain needs, from a SKU level, is validated. The results from the JDA system are then connected to SKU-level demand, and Casual Male can plan its upcoming merchandise buys.
Since applying BI to the mix, the chain has experienced more accurate sales at a SKU level, particularly in its fashion segment, where accuracy is critical, Hernreich reported.
“Instead of buying 8,000 units of a specific style, for example, we may purchase 6,000 units based on more accurate sizes,” he explained. “There are less overstock and out-of-stock conditions, and this has helped the category become more profitable.”
Casual Male plans to augment its BI and forecasting combination with assortment planning functionality by 2012.