After a decade of experimenting with futuristic, customer-facing in-store technologies, yet experiencing disappointing financial returns, grocery retailers have pulled back from many of their initiatives. But their retreat is premature. Our research shows that store tech is poised to re-emerge as the next frontier of grocery value creation, this time with a focus on the frontline.
The So What
Many grocers have seen uneven results from their attempts to scale innovations such as smart carts, checkout-free shopping, and computer vision robots. Meanwhile, upstream AI use cases in merchandising, marketing, and e-commerce have been competing for their investment dollars. The result? Leading retailers have shifted their attention away from in-store tech.
Yet the conditions that have discouraged investment in grocery store technologies have evolved. (See Exhibit 1.)
The economics have changed. Structural pressures on the grocery P&L have amplified the case for labor- and cost-saving technologies. These pressures include product-cost inflation, real-time price competition, and sustained wage demands.
Tech barriers have fallen. Grocers now have access to lower-cost, more-durable hardware, such as electronic shelf labels, cameras, and Internet of Things infrastructure. Technology deployment and maintenance costs have also declined. Therefore, the barriers to entry for many retailers are lower.
Customer expectations are higher. Customer expectations have risen, with shoppers now demanding near-perfect product availability, seamless order execution, and real-time price transparency. They expect grocery retailers to better leverage technology, using their data correctly and appropriately. And over 80% of consumers expect personalized experiences.
Store associates are more digitally fluent. A new generation of tools, including handheld devices, task apps, and guided workflows, is increasingly embedded into daily work. Meanwhile, the ubiquitous use of smartphones has made staff more tech savvy, reducing barriers to adoption across the board.
Tech has advanced. Improvements in the performance and accuracy of technology has bolstered the quality of data used for tasks such as managing product availability, shrinkage, and operational compliance at scale. And always-on, connected store infrastructure, including segmented Wi-Fi networks, means that every part of the store can be continuously connected, able to send data and trigger actions in real time.
AI is at an inflection point. Traditional systems generated alerts or applied fixed rules using exception flags or thresholds, and humans determined the next steps. In contrast, current AI systems can ingest multiple streams of store data, such as from point of sale and inventory systems, as well as from shelf signals. These systems can then apply machine learning to predict outcomes, such as lost sales. And they can evaluate tradeoffs, such as staff effort versus sales impact, to determine which matters most.
In addition, AI systems can prioritize and sequence actions. For example, they can instruct frontline workers what to do in real time without waiting for direction from management, and they can increasingly trigger actions automatically (within defined guardrails). This shift enables a step change in productivity. Stores can focus scarce labor on the highest-value tasks.
Dive Deeper
We expect that the next wave of value in grocery retailing will come from applying AI to the thousands of small decisions that shape store performance every day, particularly on the frontline. In many cases, the winning formula will come from pairing relatively simple, lower-cost infrastructure with intelligence that helps employees act earlier, faster, and better. In fact, that pairing is already becoming a leading source of value. For example:
- Restocking and On-Shelf Availability. Stores can use shelf-mounted cameras and AI-enabled tools for more than just detecting empty shelves. They can also use it to predict which gaps will matter most, prioritize the highest-impact SKUs to replenish, and guide teams before an out-of-stock item turns into a lost sale. The value is not simply in seeing a problem but also in helping the staff act on the right problem first.
- Equipment Management. Relatively simple connected devices can do more than flag issues. With AI layered on top, they can predict maintenance needs, monitor physical assets (such as refrigeration), and trigger self-regulating actions that reduce downtime, waste, and operating costs.
- Associate and Store Leader Support. AI assistants embedded in employees’ devices can also put expertise at associates’ fingertips in real time, helping junior and senior team members alike answer questions, resolve issues, and execute tasks more effectively. For store leaders, the value comes from knowing what’s driving store performance, what should be prioritized, and where labor should be deployed to have the greatest impact.
By applying AI, stores can spend less time identifying and prioritizing problems and more time fixing the ones that matter most.
Now What
Grocery retailers should consider the following factors when investing in in-store technologies.
Prioritize execution value. Innovation investment has often skewed toward customer-facing experiences. Yet the largest value pools in grocery stores are often invisible to the customer, including the management of labor, shrinkage, price execution, inventory, and working capital management. Grocers should therefore prioritize technology that improves the total cost of doing business. If it doesn’t materially improve in-store productivity or operating costs, it’s unlikely to generate a return on investment.
Design tech for action, not just analysis. Many in-store technologies offer insights; dashboards, alerts, and visibility tools provide useful information. But store teams operate in changing environments and have limited time. Having more data without clear, recommended actions can add complexity. Retailers should therefore focus on technology that can help teams change how they operate. (See Exhibit 2.) This includes technology that:
- Identifies and prioritizes actions, given a specific context and desired impact
- Guides frontline workers through complex, multistep actions in real time or uses agents to execute tasks
Emphasize change. The right technology unlocks value by enabling operating models to evolve. For example, it should help grocers to:
- Redesign workflows to remove or fundamentally change the work being done
- Update employee roles to reflect new task structures
- Clarify decision rights among store teams and central functions, identifying what should be automated, guided by AI, or directed by management
- Embed technology adoption into daily operations through focused change management
- Capture value across functions, given that many technology solutions can help manage employees, shrinkage, product availability, and merchandising
The authors would like to thank Tony Portera and their former colleagues Bennett Morgan and Hallie Greitzer for their contributions to this article.
