Category management practices and solutions have evolved over several decades, yet they have remained true to their original process. It is mostly dependent on transactional data, research, and surveys for assortment and space planning, optimizing, forecasting, and so on. These activities are consistently backward-looking. We have experienced incremental solution innovations since its inception, primarily driven by the localization of assortment. This prompted the need for a more granular level of data to deliver a better, localized shopper experience.
Technology advancements provided ways to deliver on localization, but they mostly dealt with the same type of data, just more of it. Over the years, we’ve seen several advancements on the solution side, from space management to analyzing and forecasting, that have consistently relied heavily on historical transactional data.
Enter Social Media
Research relies heavily on panel data and surveys, traditionally delivered by companies like Nielsen. They were lengthy to implement and costly, and the questions were prompted. They did not truly reflect real-world conditions. Results were somewhat effective and reactive, not spontaneous, until now. Enter social media. It is providing a new, more effective means to understand the shopper than what we have traditionally gotten accustomed to with historical data, panels, and surveys.
Listening Where No One Is Looking
I have been interested in computer vision for the past two decades. I have mostly focused on how it can be used to efficiently understand store shelf conditions, which is a perfect and obvious case for reporting on shelf compliance.
It turns out this technology holds even greater potential — offering more relevant insights that, in my view, represent the most significant disruption to category management in years. This breakthrough is poised to fundamentally transform how we conduct category reviews and shape category vision. Setting category strategies will be better founded on real-world insights and will lead to more effective tactics that result in delivering better shopper satisfaction.
Analyzing images and videos from social media posts provided unprecedented insights into consumers in an unprompted and unbiased manner. And for the first time, it is forward-looking.
How and What
I have seen PCSSO, a social visual analytics company, deliver mind-boggling trending, segmentation, and forecasting capabilities by analyzing images and videos. I had my doubts. Originally, the AI was trained in fashion and beauty. I thought it had limitations, but they repeatedly proved me wrong. They demonstrated their capabilities across fast-moving consumer goods, ranging from alcohol to canned beans, as well as home improvements like power tools, paint, and furniture. If it is in pictures or videos, it can be analyzed.
Dr. Brian Harris, father of category management, said after previewing PCSSO’s capabilities, “Amazing stuff! Will transform how we capture predictive consumer and shopper insights into sentiments, segmentation and behavior. Will be a powerful tool in the next generation of AI driven Category Management.”
Best Kept Secret
I thought PCSSO’s capabilities were the best-kept secret until now. Using downtime from the COVID-19 disruption, PCSSO diligently built and hardened its capabilities, working with some key brands and rolling them out to several countries and markets.
The secret is no longer. Earlier this month, at the Blue Yonder ICON conference, I witnessed this capability come alive with some of the world’s leading category management and supply chain solutions. Is Blue Yonder bringing about a differentiating solution of Platform, Solutions, and Data?
Forward-looking insights from PCSSO will now be available directly in leading assortment tools from Blue Yonder. Understanding the category’s potential as part of the review process has just received a major boost, making this activity more precise and less wishy-washy.
Forecast Me Not
The forecasting activity is forever changed. Heavily reliant on transactional data — and yes, AI — it was mostly an exercise in pattern detection. Forecasts were rarely validated; they were adjusted. No more. Social visual analytics have shown us unprecedented forecasting accuracies for products and categories, one to four years in advance. PCSSO is so confident that the accuracies of these forecasts come with a service level agreement. I was stunned when I first heard that.
Future of Market Insights
Awareness is beginning to shift from a crystal ball feel to reality, similar to what we experienced with AI and its derivatives over the past few years. PCSSO has tapped into unprecedented market insights that will forever change how we conduct category reviews. As one retailer recently commented to me, all assortment solution capabilities appear the same, lacking differentiation in innovation. That has now changed forever. Social visual analytics is disrupting market insights and the entire category review process. Who is ready for the future of delivering consumer insights Accurately, Repeatedly at Scale and Speed (ARSS), now?
