Using Downstream Data to Improve Forecast Accuracy

Downstream data has been electronically available with minimal latency (1-2 week lag) to companies for over two decades. Sales transaction data are being captured at increasingly granular levels across markets, channels, brands, and product configurations. Faster in-memory processing is making it possible to run simulations in minutes that previously had to be left to run overnight. In fact, companies receive daily and weekly retailer POS data down to the SKU/UPC level electronically, which can be supplemented with Syndicated Scanner data across multiple channels (retail grocery, mass merchandiser, drug, wholesale club, liquor, and others) by demographic market area, channel, key account (retail chain), brand, product group, product, and SKU/UPC. As a result, a new process has emerged called “Multi-Tiered Causal Analysis” (MTCA), which ties downstream data (POS/Syndicated Scanner data) and the performance of a market, channel, brand, product, and/or SKU at retail to internal upstream replenishment shipments (or sales orders) across time (two to three years). Now sales and marketing programs that influence demand can be simulated and evaluated to determine the full impact ...