Providing units on hand, units sold, period and regional comparisons and more, point-of-sale (POS) data can be a robust source for insights concerning market size, share, and trends over time. Think of POS data as a census of all items and brands bought in particular time periods, regions, and categories (e.g., men’s running shoes). Capturing spending habits, payment methods, home delivery, online/offline trends and then layering in such data as regional population growth and weather patterns can bolster predictive power even further. Machine learning techniques have become more sophisticated and fine-tuned models enable for quicker and more telling comparisons of critical information. We’ll dive into the advancements and trends in this area, shed light on potential pitfalls, and share advice on data sets and analytic methods that can be used in conjunction with POS data to help render the answers you seek.
Presented by: John Bremer, EVP of Research Science, The NPD Group