EXCLUSIVE Q&A: Bain & Co. analyzes next-gen AI in retail marketing
Leading-edge artificial intelligence models can help retailers transform their marketing efforts to better meet the needs of modern customers.
Chain Store Age recently spoke with Aaron Cheris, global head of Bain & Company’s retail practice; Beth Myers, leader of the personalization capability in Bain & Co’s retail practice; and Yael Mohan, partner at Bain & Co., about how generative AI and agentic AI are enabling retailers to refine and tailor their marketing campaigns for maximum consumer engagement and satisfaction.
How can generative AI help retailers personalize the customer experience?
Beth Myers: Generative AI allows retailers to break two historically limiting constraints in personalization. The first is creative capacity to create different versions of both images and texts. Generative AI enables retailers to create thousands of versions or more of each easily.
The second, and related, constraint that generative AI can lift is on the data side. The technology can easily take unstructured data and make it structured. For example, the image it creates for the customer – was there a family in it, did it picture a steak in the background, etc.
All of this helps retailers analyze which marketing messages work for which customers, and in which circumstances, so they can shift their mix to the variations that work best. This personalized variation can be in the form of emails, app notifications, homepages, view item page texts, etc.
Then there is agentic AI, which can drive personalization through conversational commerce. Agentic AI can understand the customer’s true intent, and their ingestion of images, to help them find the best tie to go with that suit or show how that lipstick would look on them.
But not every solution needs to start with generative or agentic AI. Reinforcement learning can do a great job of learning what individual customers want and driving personalization engines, and then retailers can use agents or generative AI to improve the process surrounding these models.
[READ MORE: Survey: Content marketers see benefits in rise of AI]
How can retailers ensure generative AI personalization satisfies customers?
Aaron Cheris: Generative AI is only as useful as the guidance retailers give it and the success metrics they put around it. Retailers can ensure their generative AI personalization efforts satisfy customers by building in the right variation and experiments to test different versions, and by keeping the objective measures of success in place for customers who go through those experiences, such as conversion, sales uplift and customer satisfaction.
How can retailers integrate generative AI into existing marketing activities?
Yael Mohan: There are several ways retailers can integrate generative AI into marketing activities. On the insights front, they can use it for data collection – for example, AI-moderated customer interviews and focus groups.
On the analysis side, they can use it to read through social media sentiment and customer survey feedback, highlight themes in the feedback, and help identify clear segments of customers with different priorities. On the creative side, generative AI can help retailers create early prototypes or final creative for campaign content, and then meta data tag that content for more effective analysis.
The technology can also help with the administrative aspects of marketing, as it does in other departments – summarizing meeting notes, writing RFIs/RFPs for vendor partners and evaluating responses, etc.
What are the biggest marketing AI trends in retail for the next six to 12 months?
Aaron Cheris: There are two major near-term marketing AI trends. The first is the likely continued proliferation of distinct personalized messages and offers that retailers can put in the field simultaneously.
Second is the speed at which that variation at scale can help speed up the pace of learning, as opposed to traditional A-B testing and champion challenger models. This will significantly impact people and processes as well, accelerating the marketing department shift from production studios to data-driven newsrooms.
