Weather Segmentation & Retail Forecasting
It doesn’t take a genius to know that some items sell better in warm weather and others sell better in the cold. Buyers “know it.” Customers “know it.” Store Managers “know it.” But in retail, it is rare for a system to “know it.”
I was recently at a national retailer who was planning a springtime cough and cold event complete with umbrellas, galoshes and allergy medicine. Beginning March 1. Across the US. Even in the Upper Penninsula of Michigan, Vermont, Montana and International Falls, Minnesota. Where shoppers are still refilling their antifreeze and looking for replacement gloves for the ones their kids lose on the playground every week. (I imagine the first thaw on most northern elementary school playgrounds looks a lot like an outerwear flea market.)
When I asked if they were considering that March 1 is still winter in the northern half of the country, I was told that they knew that – but their systems limited them to national promotions and they would manage the inventory balancing with excel spreadsheets – by SKU, by Buyer, by Distribution Center….as always. How inefficient is that?
Retail technology companies will tell you they can solve this with their ERP’s and smart systems and they are technically correct – the thing is that the weather savvy component is almost never part of the core installation and IT cuts get made and the company settles on an excel or (if they are REALLY savvy) access database to manage the nonstandard weather sensitive items. Truth is, the #1 retail tool in America is Excel. Take that, Oracle.
How to handle this is to use a single ONE DIGIT numeric field in the product set up to indicate its weather sensitivity. Define 10 weather seasons (0-9.) DONT expect buyers to use them to set up their SKU’s because they won’t. But once the weather pattern has been created, every 6-12 months run a report to compare the SKU’s sales the the 10 weather patterns. Select the pattern it most resembles and automatically assign one of the 10 seasonal indicators to the item. Use this to improve forecasting by improving the BI (Business Index – see past forecasting post and definition.)
Or, maintain hundreds of spreadsheets. That works too.