How to Use AI to Turn Insights into Strategic Recommendations

Category managers are under more pressure than ever before to move faster and act smarter in a rapidly evolving retail environment. Artificial intelligence promises to revolutionize decision-making but knowing how to apply AI beyond surface-level reporting remains a challenge for many organizations.  

According to a 2024 dataIQ survey, more than 70% of Fortune 1000 companies were starting to experiment with AI. Fast forward a year and nearly half have implemented AI in limited production, with another 23% into at-scale deployment, with the majority seeing high business value. Organizations are quickly realizing that they can achieve measurable business growth with AI.

To stay competitive, category teams must move beyond using AI solely to report on past performance. AI can now be used as a strategic partner and digital workforce capable of identifying opportunities and recommending specific actions.  

In this article, we will explore how retail analytics is evolving from traditional dashboards to AI-powered recommendations for space planning. We’ll also share practical examples of how leading brands are using AI to drive faster, smarter decisions.

The Four Core Tasks of AI

AI may sound complicated, but at the core, most models perform one of these four tasks:

• Regression: Analyzes relationships between variables to predict outcomes (like how price changes could affect sales)  

• Classification: Groups data into predefined categories (like which stores share similar buying patterns)  

• Translation: Converts one format or input into another (like using a text prompt to generate an image)

• Optimization: Recommending the best possible action based on a set of rules and constraints (like identifying optimal product assortment for a specific store)

Many teams are already using AI for regression and classification to process large volumes of data and summarize reports. But optimization is where category management can make an impact on business decisions. With the right models in place, AI can recommend what to change, where and why.

The visual below illustrates how most teams are still operating in the descriptive and predictive phases, not yet unlocking the full value of AI recommendations.  

From Insights to Action: Moving Beyond the Rearview Mirror

Leading category teams are adopting AI optimization tools that layer business constraints like shelf space, pricing, and inventory thresholds, over sales and performance data to generate clear, actionable recommendations.

Here’s some real-world examples:  

These category teams leveraged AI optimization and recommendation models to help reimagine shelf strategies, enabling rapid iteration of scenario testing and opportunity cost modeling with precision. This allowed for more strategic and increased collaboration with retail partners to drive category growth.

• Days of Supply & Expiration: Needed to determine the optimal number of facings for perishable items with very short lifespans by assessing the likelihood of units not selling before reaching their expiration dates

• • Results: $52 million in annual markdown and spoilage losses identified across hundreds of stores

• Marginal Value of a Facing: With limited restocking windows and frequent out of stocks, the category team for a DSD category needed to understand the value of each facing by quantifying sales contribution and space trade-offs between core & secondary SKUs

• • Result: 10% lift in sales by reallocating shelf space and allowing occasional stockouts on slower movers

• Inefficient Product Assortment & Shelf Space: Needed to reduce underutilized space and boost supply for high-performing products while still implementing key business constraints  

• • Results: 2.5% lift on a billion-dollar category & reduced response time from months of analyses to minutes

Augmenting Human Expertise with Artificial Intelligence

With all this talk about AI recommendations, it might seem like there is little left for category managers to do but that is not the case. AI does not replace category expertise; it enhances it by taking on the manual, time-consuming work that often slows teams down. By automating routine analysis and surfacing key opportunities or potential risks, AI gives time back to category managers so they can focus on what they do best: guiding, validating and refining strategies. The most effective approach is augmented intelligence, where human insight and AI work together to accelerate decision-making.

Automated Insight Detection and Root Cause Analysis

An example of augmented intelligence in action is Engine’s AI-powered Auto Insights solution that works like a 24/7 digital analyst. It proactively monitors data for shifts in performance, flags the potential cause (like a stockout or pricing issue) and makes recommendations on how to respond. Instead of digging through dashboards, category managers get clear alerts with helpful context so they can validate the issue, refine their strategy, and act quickly. The result is faster, more informed decision-making that keeps teams competitively proactive instead of reactive.

Here’s another real-world example:

A leading snack manufacturer used Auto Insights Engine to uncover the root causes of a sudden sales decline across over 200,000 points of distribution. It flagged two major shifts: a 21% price increase tied to a 15% drop in sales, and a 45% reduction in shelf space that led to an 11% decline in sales. By automatically analyzing POS, pricing, and shelf data, the AI not only determined where and what changed along with why, but also quantified the business impact, which enabled the team to quickly take action to recover revenue and protect shelf space.

Where do category teams go from here?

AI is already delivering value in automating reporting and surfacing insights. But its true power lies in what comes next: forecasting, detecting experiments, identifying growth opportunities, and making actionable and strategic recommendations.

As category teams continue to explore AI, it will be important to keep these questions in mind to ensure its implementation is delivering true business value:

• Is AI helping the team move faster and make better decisions?

• What role does AI play in day-to-day strategy?

• Are recommendations from AI driving quantifiable results?

Teams that embrace the full potential of AI will be best positioned to lead the next era of category growth.