From quick wins to long-term strategy, retail c-suite leaders discuss their AI journeys and where they plan to go next.
When it comes to artificial intelligence (AI), retailers are at a tipping point. Until now, most have made only modest progress toward incorporating AI into their operations. However, the launches of ChatGPT and other generative AI tools this year have been so impressive that retailers are realizing that they can’t afford a wait-and-see approach. In fact, a 2023 Gartner report predicts that AI for retail will soon become retailers’ “nervous system, supporting continuous adaptation.”
However, most retailers are still getting their heads around AI, including its myriad potential uses, the challenges of making AI insights actionable, and the competitive threat AI poses for those who fall behind the curve.
For all these reasons, we decided to devote the inaugural session of the Logic Retail Leaders Forum to AI for retail. A series of topical, invite-only thought leadership roundtables, the forum brings together Logic experts with some of the industry’s leading practitioners. Logic’s Rupesh Pradhan, COO, and Tony Rost, Global Managing Director for Cloud & Managed Services, hosted the event. They were joined by an impressive roster of retail leaders, including:
- Megan Douglas, CIO, Maurices
- Rick Dreiling, VP & CIO, Bealls
- Mike O’Reilly, Head of Global Data & Analytics, Ralph Lauren
The forum provided the participants a chance to share their real-world journey to adopting AI for retail, and how they expect it to transform their organizations. Here are eight key takeaways from the conversation.
1. First, nail down data governance
The forum participants were unanimous on one point. You have to get your data ready first. “Clean data and data governance are the foundation of AI,” says one participant. And without high-quality data, AI for retail cannot help guide and grow the business. “In fact, it can lead you astray,” they said.
To make their organizations future ready, all four leaders recently completed major data governance initiatives. “In terms of the data, I really stay ahead of the business, to make sure when they come up with an idea, we are already halfway there,” said another leader.
2. Don’t get distracted by all the possibilities
The potential ways retailers can harness AI is endless, especially with the advent of generative AI tools like ChatGPT. The advent of generative AI has significantly increased excitement across retailers’ organizations, but that excitement can actually be distracting. “At one point, I was having 10 conversations a day about AI,” says one leader.
However, as one leader put it, you can’t eat an elephant in one go. Excitement is great, participants agreed, but success requires focus, e.g. committing to a limited number of high-priority use cases at a time, executing on them, and then moving to the next. Endless conversations can actually distract from that goal.
3. If new to AI, start within the IT organization
AI for retail is developing incredibly quickly, but most retail IT organizations have only limited experiences executing on AI projects and lack in-house data science resources. So before unleashing a new and unfamiliar technology across the organization, a number of the forum participants have decided to first learn the ropes by using AI within their own departments.
“Testing AI on IT practices is a good place to start,” says one retail IT leader. “Start by playing in your own sandbox. Eat your own dog food,” adds another. In particular, they recommended trying AI for systems monitoring and debugging.
4. Leverage existing tools from the big cloud providers
Until recently, AI was a daunting field to enter that required scarce and expensive expertise. That is quickly changing. Before trying to build homegrown systems, attendees suggested first exploring the prebuilt AI tools provided by big cloud providers like Amazon AWS, Microsoft Azure and Google Cloud.
Take Amazon AWS. They offer more than 100 easy-to-use APIs for AI and ML. So, for example, an entry-level programmer can be up and running on image recognition within about an hour. And in most cases, the microservices and metered billing delivered by cloud providers make them a more cost-effective and flexible choice than third-party offerings, which are often just a repackaging of those same services.
5. Get quick wins with help desk/tech support
Automating key aspects of internal tech support and management is a great place for retailers to build their AI chops – and one that can both improve service while reducing costs. There are many ways AI for retail can help speed resolutions while consuming fewer resources, including:
- 24/7 chatbot support. AI-powered chatbots, once considered more trouble than they were worth, are improving quickly. Not only can they provide a 24/7 form of triage, they are increasingly able to keep down ticket numbers by providing quick solutions to common issues.
- Ticket prioritization and routing. Increasingly, AI can not only recognize a problem and route it to the right specialist, but also help to define the severity of a problem.
- AI-driven, human-in-the-loop resolutions. While a user-facing chatbot may not be able to solve all problems, specialists empowered with AI-driven tools can speed solutions to more complex issues.
6. Harness AI as a creative partner
As Neuroscience News reports, a 2023 University of Montana study found that AI scored as well as the top 1% of human thinkers on a standard test for creativity. And one forum participant reported significant success using generative AI to boost the creativity of both copywriters and designers.
In fact, they are looking for ways to leverage AI to create and then dynamically serve up content tailored to individual customer preferences rather than one-size-fits-all product descriptions.
7. Automate the generation of product attributes
AI is proving to be an effective partner in developing and managing product attributes, usually a painstaking manual process. By doing so, said one participant, the organization can capture attributes much earlier in the product development process, helping speed time to market.
Better yet, the scalability of AI means retailers can capture many more attributes than manual systems make possible. More attributes means more data for targeting the right products to the right customer—and fuel for richer insights into customer preferences for future campaigns.
8. Keep an open mind to all the possible use cases
Now that their data has been cleansed and AI-ready, all four participants were brimming with ideas for the use cases they want to try next, including:
- Inventory optimization. By unleashing AI on inventory data and demand signals, retailers can expect big improvements to the bottom line. In fact, one retailer is already seeing that happen.
- Associate empowerment. Armed with AI-driven insights, associates have the potential to sell more effectively and deliver more satisfying experiences to customers.
- Training content. AI, especially generative AI, has the potential to speed the creation of more efficient, effective and intuitive training content for a new generation of associates.
- Weekly reporting. AI is already being used to automate the time-consuming painful process of creating weekly reports—and making reports more useful to decision-makers.