While the term conjures up images of a futuristic society run by robots, artificial intelligence, or AI, is actually a tool of the present. And for many brands, an international pandemic has only accelerated its adoption.
At the WWD Virtual Apparel and Retail Summit on Thursday, brands weighed in on the ways that data-driven technologies have become essential to their businesses.
According to Puma’s vice president of merchandising, Katie Darling, the collection of data has “been a focus for quite a while now,” especially as the athletic apparel and footwear stalwart moves more heavily into the e-commerce space. But once teams develop the capacity to collect significant data and attributes, they must ask themselves how to use the insights to “make the best decisions”—and that’s not always easy.
“That’s when these massive Excel spreadsheets and even data warehouses have to be coupled with artificial intelligence to really bring out the whole benefit,” she said. Being in the athletic gear industry, innovating and challenging the status quo are familiar impulses, she added, and it helps to be a part of a company that encourages adaptation to new tech tools across the board.
Implementing AI for in-store sales has proven especially critical in a year full of unpredictable circumstances, she said. Data helps teams to plan their assortment for each store by region, tapping historical consumer insights. But making educated buys based on what performed well last season, or even last month, doesn’t always help in a situation like the present.
Related Stories
“This year there were so many disrupting factors to react to,” she said. AI can help “brands that may have started with a really solid plan for each store” to “understand the trends and scenarios, and adjust on the fly.”
Rather than treat 2020 like a fluke, Darling believes that some significant retail shifts could emerge from the crisis. Loungewear could become even more important. Back to school could ship later. Holiday could ship earlier. “These changes might actually become trends that come out of Covid,” she said, and Puma’s planning teams won’t simply “wipe this year away” once it comes to its long-awaited close. “We’ll actually utilize what we’re seeing to impact the future, because some of those trends might be here to stay.”
AI has allowed the brand to spot those nuanced developments early, she said, while it might take human eyes “months to capitalize on, or to understand, that this trend is happening and it’s not going away.”
Kate Nadolny, senior vice president of business strategy and innovation for PVH Corp., which owns brands like Tommy Hilfiger and Calvin Klein, said that the company’s strategy for using data and AI at the store level has been steadily evolving.
“We’ve really been focused on utilizing AI for what we think is the here and now, the lowest hanging fruit,” she said, “which is reacting real-time to making pricing decisions on the floor within our retail stores.”
There are extensive learnings to be gleaned from watching trends unfold across multiple stores in different regions, and even across brands, she said. It’s important to understand, “on a more granular basis,” what shopper behavior looks like at each store, along with “what the different levers are, and how we can potentially pull them” to implement effective promotion strategies, she said.
“Previously, just because of the sheer volume of SKUs and stores and product across our brands, we were really focused at a relatively high level,” she said. “It’s hard to execute at the store level, and at the item level, when you’re working on Excel,” she added. “We’re able to now make decisions that are a little bit more real-time, and a little bit closer to some of the detail.”
The company is now starting to think beyond pricing and look at inventory planning as well. Naldony said that PVH Corp. is now starting to “engage a little bit more deeply into assorting” using historical data and third-party analytics.
AI tools also give the company’s brands an enhanced understanding of what the consumer is going to be looking for earlier on in the product development cycle, she said. Those insights can be used to make decisions on product lines, “and really build the box that our merchants and designers can play in to make the right decisions” she added. “Now, it really is more about understanding how to take those data points and turn them into action.”