Interest in trade-promotion forecasting among retailers is growing, says Aspegren. Consumer packaged goods producers are investing heavily in this previously underused technique for improving the success rate of promotions in the retail world.
Until now, there hasn't been a reliable way to calculate the effect of promotions, even though they drive as much as 30 percent of net sales. With the economy in such questionable shape, however, it becomes more important than ever for companies to get a handle on the technique.
Trade promotion embraces all aspects of marketing and advertising, and has widened even more in recent years with the increase in sales channels. Retailers must also look at "all the things that go on within a store," says Aspegren.
A modern-day approach to the problem employs methods based on machine learning. The concept helps companies to understand what actually worked on previous occasions. A retailer can draw on the historical data of promotions, including how much it promoted and discounted the items in question. Based on that data, it can cluster the result and calculate the correlation between the various factors and results of the promotion. "You can actually clean the forecast from the promotion," says Aspegren, then add on relevant variables to calculate and forecast the impact of future efforts.
Aspegren has a fairly optimistic view of the process that companies are making in deploying trade-promotion forecasting - with some caveats. "Based on our market in the Nordic region," she says, "I honestly feel that the cooperation between retailers and CPG [manufacturers] is fairly good. The problem is, we haven't had the methods to calculate the effect."
One of the biggest challenges is dealing with the huge volume of data that is involved in a typical retail environment. "You need to be able to take into account thousands of different promotions," Aspegren says.
.webp?height=100&t=1780891461&width=150 "021_what_is_ai_in_warehousing_and_the_supply_chain- (540p).png")
