Insights -- September 2005

In the deployment of RFID systems to date, the retailer has carried RFID momentum and has dictated its direction. However a "paradigm shift" is in progress: As analytical software is developed to benefit suppliers, the relationship will change from a retail push market to a retail-CPG push pull market. The retailer needs to be an equal partner with the CPG supplier in addressing consistency issues across supplier relationships, especially considering retailers often bear the major responsibility when out-of-stock scenarios arise.

CPG manufacturers are now receiving RFID data from trading partners. As the data context clarifies and volume increases, CPG manufacturers can progressively embrace RFID data analysis and scenario analysis to address weak points and pain points.

Micro-Inventory Management Implications

Out-of-stock reports are generally based on sell-through reports and lost sales reports. These help corporate offices to identify items, brands or vendor service levels relating to stock conditions. Market feedback gathered by ABI Research reveals that it is not an issue of whether the point-of-sale-driven systems work, but rather, when the product can be selected and delivered.

Model stock and order points can drive the system to order and deliver before an out-of-stock occurs.This greatly improves in-stock conditions. Model stock examples demonstrate how to minimize out-of-stock conditions and shed additional light on how RFID can address this issue. Retailer vendor lead-times may differ between electronic data interchange (EDI)-driven transactions or drop-ship transactions that are buyer-generated, typically seasonal and one-off events. EDI orders are tied to fast-moving goods that are shipped to retail stores from warehouses. From an out-of-stock perspective, drop-ship deliveries require less coordination: RFID is more helpful in monitoring the movement of fast-selling, EDI-generated orders.

RFID can also be used to analyze store-to-store inconsistencies in retail supply metrics. Trucking services may report high, on-target or on-time deliveries and vendors may articulate high service levels, but this does not necessarily address in-stock levels on the retail floor. These reports don't state whether service levels pertained to EDI orders or drop-ship orders, nor whether services levels were achieved due to in-stock or substitute sales. RFID adds detailed data and increase control.

Macro-Inventory Management Implications

Inventory management affects so much more than out-of-stocks. RFID allows better, more accurate sell-through reports, fewer markdowns and lower costs. RFID can equip and refine these systems to drive improvements in operating performance. It is important to look beyond the simple picture of out-of-stock inventory, and to segment it into clusters.

Retail inventory consists of three tiers. Wal-Mart defines its inventory management across a 7-14-30 day program. Seven-day inventory is known as "warehouse inventory" and represents inventory expected to last no more than seven days in a back room. This category includes fast-moving consumer goods.

"Assembly merchandise" is classified as 14 days' worth of inventory, or lower turn inventory. For example, regular Suave shampoo is warehouse inventory whereas Suave dandruff shampoo would likely be assembly inventory.

"Feature inventory", or 30-day inventory, includes bulk items purchased by the home office that may be new items, seasonal items, items in circular advertisements, or due to an advantageous purchasing deal.

RFID affects each bucket -- warehouse, assembly and feature inventory -- differently. If stores do not have the right product mix, inventory management does not matter all that much. Where out of stocks are primarily focused on fast moving consumer goods -- those tied to top tier CPG manufacturers -- feature inventory opportunities exist in better planning purchase windows for seasonal goods that vary regionally.

For example, the window for spring jacket sales is different in Kentucky from that in Michigan. Using RFID's tracking capability to acquire additional data points across all stores, and adding third-party information (temperature, weather, and traditional business metrics), non-technological planning teams can better manage retail channel relationships.

Data Manipulation

Business intelligence and data warehousing companies are actively seeking to understand how EPC-based data acquisition rates vary, where caching makes sense and where real-time use makes sense. How much intelligence and analytical capability is needed at the reader level? At the network edge? At the enterprise level? How should management of these areas change as business needs evolve and RFID is incorporated into daily operations? Real-time data generation is increasing rapidly and the application software and data management companies are working to understand what data to filter out and what data to keep, not only for today's RFID applications but also for future applications that will find new uses for data considered irrelevant today.