Rapidly advancing AI technologies are helping category management practitioners do the unthinkable
Category management is at an epic turning point as innovative technology gives retailers the ability to do things that were unimaginable a few years ago. Predictive and generative AI advances are changing how retailers manage shopper data and identify insights, develop promotions and sales strategies, optimize assortments, automate mundane tasks, create planograms, and ensure compliance.
In each of these areas and others, category management has been a field where change was largely incremental. However, the extension of predictive and generative AI into the category management ecosystem has changed everything, as workers can make decisions faster and have greater certainty about achieving intended outcomes.
This is a dream scenario for anyone connected to category management as retailers discovered at the industry’s largest event focused on category management and shopper insights. Nearly 1,000 attendees—including category managers, space planners, and shopper insights professionals—gathered in Dallas between February 26-28 for the annual conference of the Category Management Association (CMA) and Shopper Insights Management Association (SIMA), which are part of The Association of Retail & Consumer Professionals (ARC).
As a leader in AI-powered category management solutions, SymphonyAI had an expert team on-site to meet with retailers and consumer goods companies. Out of those conversations, some key themes and priority areas surfaced, including the following:
Insights Discovery
Retailers want to uncover actionable insights with a higher degree of granularity faster than ever because shopper behavior is highly dynamic and product preferences can change quickly. Analyzing and identifying insights in historical information is a sweet spot for predictive and generative AI. But the bigger opportunity is to identify reasons why behavior changed. That requires powerful technology to access and distill information from diverse datasets in minutes rather than hours or days. With an understanding of why, retailers can move from insights to action faster than ever. Where things get interesting with predictive and generative AI is the discovery of insights and identification of correlations retailers may not have known to look for, along with recommended actions and estimates of value creation. That’s powerful and a major advancement for category management.
Copilots
CMA/SIMA attendees had a lot of questions about copilots. The SymphonyAI team had answers, as the company was the first to launch a predictive + generative AI category manager copilot in October 2023. The productivity-boosting innovation is why there is so much talk about category management disruption and new ways of working. The AI-powered category manager copilot gives retailers the ability to engage in scenario planning and make optimal decisions more quickly by combining the company’s decades of retail and CPG knowledge, processes, content, and customer segmentation to provide rapid insights into current customer merchandising and sales data. The category manager copilot understands the intent of a user’s questions and rapidly identifies all performance-affecting factors in a product category and recommends the best actions.
Assortment Optimization
This is a topic near and dear to every category manager’s heart, especially as shopper behavior has grown increasingly dynamic and retailers need to create localized, space-aware offerings that will resonate with shoppers. Retailers showed strong interest in the application of predictive and generative AI to assortment optimization because the technology brings simplicity to an inherently complex process that requires accounting for a large number of shifting variables. In many respects, assortment optimization is an optimal use case for AI, but retailers also want to ensure that recommendations are explainable. SymphonyAI’s models provide explainable results, thus eliminating the so-called “black box” effect.
Planogram Compliance
This perennial issue was an area of keen focus at CMA/SIMA because computer vision and optical character recognition (OCR) are giving retailers unprecedented visibility into what is happening on store shelves. While there are some obvious benefits to detecting on-shelf availability issues, the ability to improve planogram compliance is a huge needle-mover for sales. Retailers and their partners invest tremendous energy in creating planograms with the highest probability of sales success, so it is essential to identify when planograms aren’t compliant because that detracts from planned performance. SymphonyAI calls this ability to gather images, detect issues and prioritize corrective actions store intelligence, and it was a hot topic at CMA.
Technology usability
CMA attendees were intrigued and optimistic about predictive and generative AI, but they want to be sure the user experience (UX) and user interface (UI) of AI applications are intuitive and well-designed. SymphonyAI has done a lot of work in this area and employs a design thinking approach to ensure users have a favorable experience to derive maximum value from the underlying AI.
The pace of innovation
Because predictive and generative AI is such a fast-moving space, retailers and CPGs at CMA/SIMA were understandably interested in how solutions accelerate and improve their category management abilities. They also showed great interest in how solution providers are keeping pace with innovation to infuse solutions with next-generation capabilities. SymphonyAI’s approach to this situation is unique thanks to its Eureka Gen AI platform which functions as an innovation core to speed the development of solutions across all of the industry verticals SymphonyAI serves. To better understand this capability, it’s worth watching this short video explanation from SymphonyAI Chief Technology Officer at SymphonyAI Pushpraj (Raj) Shukla.
In short, Eureka Gen AI is a shared foundation that provides the cutting-edge data, machine learning, and generative AI core that allows engineers to build applications on accelerated timelines. SymphonyAI builds industry-specific knowledge into Eureka, and the retail-specific version of Eureka is called CINDE Gen AI.
LLMs
It’s telling how far things have come with predictive and generative AI that the large language model (LLM) acronym is now part of retail nomenclature. LLMs are of keen interest within the category management ecosystem because they are the data repository on which generative AI is trained. Retailers and CPGs were looking for assurances that LLMs are trained on industry-specific data by companies with deep domain expertise in the grocery industry vertical. Retailers have quickly gained an understanding of not just the linkage between generative AI and the underlying LLM, but that the LLM must contain industry-specific knowledge (data) to produce accurate results.
Taking action
One of the top takeaways from the CMA/SIMA conference that was also evident at the NRF Big Show in January was a desire among attendees to demonstrate value with pilot programs. It’s why 2024 is setting up as a year of experimentation and validation as retailers and CPGs execute pilot programs that are easily stood up and prove an ROI in just a few months.
It’s clear from the three-day CMA/SIMA conference that predictive and generative AI are accelerating the disruption of category management and creating a new source of competitive advantage for retailers who embrace the movement. Game-changer may be an overused metaphor, but if ever there were a context where it is appropriate, the impact of predictive and generative AI on category management would be it.
If you weren’t at CMA/SIMA and want to better understand any of the areas mentioned above, please reach out at info@symphonyretailai.com.
Executive Insight: How generative AI accelerates innovation and increases productivity.
