What Do Predictive Analytics Mean for Inventory Planners?
One of the buzz phrases that’s surfaced with merchandise and inventory planning systems over the last few years is “predictive analytics.” At its simplest, the term reflects the fact that data processing capabilities have advanced to the point where they don’t simply generate operational reports and ask the user to interpret information. They’re now capable of providing targeted recommendations.
Areas where predictive analytics can be deployed include the following:
- recommendation of purchase orders based on sophisticated end-of-quarter algorithms;
- recommendation of product assortments based on historical analysis of prior period sales in profiled clusters;
- optimal allocation of inventory to stores, interpreting location-specific sales and forecasts;
- inventory price optimization, such as advising the optimal timing and degree of price cuts to sell through inventory ownership;
- competitive price optimization (i.e., advising optimal price for merchandise relative to competitors’ pricing); and
- deployment of big data analytics to identify “phantom shelf inventory” based on interpretation of the sales-to-inventory ratio of any store in relation to peer stores.
To focus on one example, tools such as SPI’s Shiloh Analytics Advantage analytics application are now available to interrogate the sales-to-inventory relationship for every SKU, in hundreds or even thousands of stores, to identify those stores where sales are lagging behind peer stores with similar inventory ownership.
Such analysis is based upon a high degree of statistical probability. As a result, the store manager can be alerted that while the allocation system believes there’s still inventory available at the store, the absence of sales relative to peer stores advises that the inventory count may be wrong and, at the very least, they should do a stock count of that SKU.
I’ve often said that inventory planning relies on timeless fundamentals. That remains true, but the deployment of predictive analytics clearly provides gains in staff efficiency and measurable, incremental gains in sales and profits.
No doubt, technology makes this an exciting time for inventory professionals. While predictive analytics isn’t about to replace the noggin on an inventory planner’s shoulders, it has shown the ability to help you make better, more informed decisions. Are you making use of this feature in your day-to-day operations?
Joe is Vice President of Product Solutions at Software Paradigms International (SPI), an award-winning provider of technology solutions, including merchandise planning applications, mobile applications, eCommerce development and hosting and integration services, to retailers for more than 20 years.
Joe is a 34-year veteran of the retail industry with hands-on experience in marketing, merchandising, inventory management and business development at multichannel retail companies including Lands’ End, LifeSketch.com, Nordstrom.com and Duluth Trading Company. At SPI, Joe uses his experience to help customers and prospects understand how to improve sales and profits through applying industry best practices in merchandise planning and inventory management systems and processes.