How Walmart Is Using AI to Turn Data into Decisions
In retail, every transaction, product movement, and customer interaction generates a vast amount of data. Businesses can now see almost every aspect of their operations, from sales and inventory to performance at the individual store level. But visibility doesn’t always lead to faster decisions.
This points to an important next step for Data & AI in the enterprise: moving from simply helping businesses see data to helping them understand it and turn it into decisions.
The goal is not simply to add more data or build more dashboards. It is to use AI to surface meaningful insights, identify what deserves attention, and connect those insights to actions that can happen within the business. Because in a world where every organization has access to more data than ever, the competitive advantage may no longer be who has more data, but who can turn data into action faster.
From Data Analytics to AI-Assisted Decision-Making
Scintilla evolved from Walmart Luminate and serves as an intelligence ecosystem for suppliers and Walmart merchants, connecting Walmart first-party data with information across customers, products, stores, and business performance.
From 2025 to 2026, Walmart expanded Scintilla beyond a traditional data and analytics platform toward a Commerce Intelligence Platform, with a greater focus on connecting data, intelligence, and how that information can be used in day-to-day business operations.
One example is Scintilla Boards, which brings relevant metrics and signals into a single view, allowing users to understand business performance without having to manually combine information from multiple reports. The New Item Tracker, for example, brings information that was previously distributed across different reports into one view.
This reflects a broader challenge in enterprise analytics: having more data and more reports does not necessarily lead to faster decisions. What matters is how easily users can access the information relevant to a business question and how much time they need to spend interpreting it.
This is where AI begins to play a role.
Walmart is introducing AI capabilities across Scintilla to help analyze performance, identify changes in customer behavior, and summarize customer research. At the same time, Marty for Scintilla serves as a separate AI Assistant designed to help users search for and understand metrics, features, and capabilities within the platform through natural-language interaction.
Another example is AI Insights in Scintilla Customer Perception, which can generate survey summaries and highlight key findings based on defined research questions and objectives. These insights can also be traced back to the relevant survey questions and supporting data.
The shift, therefore, is not simply about making data easier to view. AI is increasingly being used to help users understand the information they already have, allowing them to spend more time on analysis and decision-making rather than searching for and preparing data.
This is an important consideration for organizations exploring AI adoption. Implementing AI does not end with selecting a model or building a system that can answer questions. Organizations also need to consider how that AI will connect with existing data, systems, and workflows in a way that creates measurable business value.
From Insight to Action
Knowing what happened, however, may still not be enough.
If a system identifies a decline in sales for a particular product category, the next question may be whether the decline is related to inventory, customer behavior, pricing, or another factor. More importantly, the business needs to know: What should we do next?
This is where Walmart is taking Scintilla beyond traditional analytics. Walmart has described the development of Scintilla Opportunities, designed to detect anomalies, surface insights, and recommend actions, alongside Custom Alerts that allow users to monitor metrics and changes that matter to the business.
The difference lies in the role of the system. Rather than simply showing what happened, the platform is being designed to help identify what deserves attention and what actions may follow.
Walmart gives the example of important products at risk of going out of stock across multiple stores. Connecting that insight to a replenishment workflow could reduce the steps between identifying the problem and addressing it.
This is an important principle for enterprise AI. The value of intelligence does not come simply from a system being able to provide an answer. It comes when data and insights can connect to the business processes that need to respond to them.
In practice, this means organizations do not always need to begin by asking, “What can AI do?” A more useful starting point may be to identify where decision-making takes too much time, which data is not being used effectively, or which decisions require information from multiple systems.
From this perspective, AI is not simply another technology layer added to an existing system. It can become part of redesigning how a business process works so that decisions can be made and acted upon more efficiently.
When Intelligence Reaches the Store
In February 2026, Walmart launched Scintilla In-Store for supplier field representatives, bringing store, product, inventory, and task information together in a single application so field representatives can use that information while working in stores. Walmart has also described the development of AI-powered recommendations and prioritization to help field representatives identify which issues deserve attention first.
This illustrates how AI in retail is moving beyond back-office analytics and becoming part of real-world operations.
For retailers, data is constantly generated at the store level, from sales and inventory to product availability and operational activity. When this information is fed back into analysis and decision-making, the system does more than report what has happened. It can also support the next round of actions.
What makes Scintilla particularly interesting, therefore, is not simply the addition of AI to an analytics platform. It is Walmart’s effort to reduce the distance between insight and real-world action.
This is particularly important in retail because valuable data does not exist only in back-office systems. It also exists where customers, products, and employees interact in physical stores. Bringing intelligence into those environments is another important step in making AI useful to the business.
From Data to Action in the Real World with Sertis
What Walmart is building with Scintilla reflects challenges faced by many retailers, from managing large volumes of data and understanding customer behavior to managing inventory and helping store teams respond more quickly to changing situations.
For Sertis, these challenges do not begin with technology. They begin with understanding the business objective and business problem, followed by identifying the data and AI capabilities that are appropriate for the organization’s existing processes.
In retail, Sertis applies Data and AI across areas including demand forecasting, customer analytics, product recommendation, personalization, and inventory optimization, while also bringing AI directly into retail operations.
One example is Shelf AI, which uses Computer Vision and Deep Learning to analyze what is happening on the shelf, including product identification, planogram compliance, and out-of-stock detection. It can also connect with e-commerce and purchasing systems to support automated reordering.
This provides another perspective on the Data-to-Action approach. Data does not have to stop at the dashboard. It can be used to alert employees, support replenishment, and connect with the systems responsible for real-world operations.
For retailers, knowing which products are selling well may only be the starting point. Businesses may also need to know where products are located, whether they are actually available on the shelf, whether the display follows the planogram, and what the store team should do when an issue is identified.
This is where AI can create business value beyond analytics. When data, intelligence, and operations are connected, AI does not simply tell a business what is happening. It can become part of the process that helps the organization respond to what is happening.
Sertis approaches AI adoption as a form of business transformation, rather than simply adding new technology to an existing system. Its capabilities span from use-case identification and AI strategy to the design and development of Data & AI solutions, Computer Vision, and AI applications aligned with each organization’s business processes and objectives.
Ultimately, the value of AI in retail is not simply about giving businesses more data or building systems that can answer more questions. It is about enabling data and intelligence to reach the point where decisions need to be made and actions need to happen. When technology can connect these elements, data that once remained on a dashboard can become part of how real business value is created across the retail operation. Source
