Retailers are meeting demand with Twitter data
Retailers have sophisticated demand-forecast models and agile supply chains to ensure that just the right amount of products are placed on the store shelves to meet demand. And yet, for some products, they get this quantity wrong. Too little stock leads to lost revenue and customer dissatisfaction; whereas, too much stock hits profit margins through heavy discounting to clear the shelves. Even worse, overstocking requires disposing of perishable items before they spoil.
How can retailers reduce such big misses in their demand-forecast models? IBM has discovered that outside forces are often at play, which helps explain these misses. A news article, government comment, scientific publication, the weather and the local rock concert are all examples of external forces that can influence local demand. But to what extent do such external forces influence demand? The real-time, public and conversational characteristics of Twitter data provide the barometer as to how people are reacting to such stories and thus help quantify the strength of the relationship of the external force to the impact on demand forecasts. Even small improvements in the demand-forecast model, when amplified at the scale of a large retailer's supply chain, translates into significant financial benefits.
Turn misses into hits. Learn more about moving Twitter data beyond social listening.