TechStyle Fashion Group — whose portfolio of brands includes ShoeDazzle, FabKids, Fabletics and JustFab — said its data science team has been predicting sales with almost exact accuracy. According to a TechStyle Fashion Group spokesman, its data model can anticipate everything from how long it will take for a customer to complete a purchase to the expected frequency of buys. Sujay Kar, head of analytics at TechStyle Fashion Group, explained how the company’s data science team is influencing each step of the way.
WWD: What role does the data science team play at TechStyle Fashion Group?
Sujay Kar: Our data models refine customer targeting, enabling us to attract the right customers, predict lifetime value and ultimately increase the company’s overall return on investment. By analyzing various customer behaviors across channels as well as purchasing trends, and other feedback, we are able to see gaps in our product offering while ensuring the success of potential roll-outs. We also leverage customer behavior data to customize product recommendations and personalization for members, increasing spend per site visit.
WWD: How does the data science team work to predict sales?
S.K.: We have a forecasting mechanism that is driving inventory prediction almost precisely. When a shopper visits ShoeDazzle, for example, they are asked to take a quiz about their style preferences. Using this information, together with actions taken across touchpoints, we are able to predict how our customers will respond to various trends, making inventory buying much more efficient.
WWD: Why is it important for brands to invest in data science teams?
S.K.: Brands should be making greater investments in the areas of customer analytics. Data science is becoming increasingly crucial as more consumers expect unique, personalized shopping experiences. Companies with great exposure to customer data will have a competitive advantage, allowing them to create a one-to-one relationship with customers and deliver personalized fashion.
For More WWD Business News, See: