Ask the Expert: How can AI be used in implementing pricing in the fashion industry?

Michael Feindt

This week’s Ask the Expert is answered by Prof. Dr. Michael Feindt. Dr. Feindt is the brain behind Blue Yonder. His NeuroBayes algorithm was developed during his many years of scientific research at CERN. Michael is also a professor at the Karlsruhe Institute of Technology (KIT), Germany, and a lecturer at the Data Science Academy.

Ask the Expert: How can AI be used in implementing pricing in the fashion industry?

Pricing is one of the most difficult strategic components to get right in the retail industry. As brick and mortar retailers are increasingly undercut by their online competitors, the temptation is to cut prices to try to keep up. However, this can come at the expense of profitability, and there are so many other factors that can drive a consumer’s purchasing decision, such a retailer’s environmental credentials and YouTube influencers, that slashing prices is not always the right answer.

There has to be a balance in a retailer’s pricing strategy between attracting shoppers
with good value purchases, and maintaining margins and profitability. With an increasing focus on ‘fast fashion’, it is ever more important for retailers to ensure they set the right prices to ensure that they maximise sales and reduce levels of leftover stock that have to be carried into next season.

This is where AI can play a critical role. An AI-based price optimisation solution can analyse vast quantities of data, including sales patterns, promotions, weather and events, determine the influence that the price has on customer demand and suggest the price that is most likely to achieve the retailer’s sales targets. The price elasticity can then be determined by the relationship between the price change and its effect on demand, as well as specific sales forecasts at that price according to stock levels in each store. Price reductions can be re-evaluated daily and are applied at the predetermined frequency in order to meet customer expectations as well as improve select KPIs.