Retail AI investment is shifting toward connected operations

Retail and enterprise leaders appear to be facing greater pressure to connect AI spending with financial outcomes. Pigment’s Q3 CFO Index is based on a survey of 2,000 CFOs and finance executives across the U.S., UK, France, and Germany. It found that 83% said consumption-based AI costs had exceeded expectations. Confidence in AI’s business value averaged 8.2 out of 10.

The figures suggest a growing focus on translating AI investment into tangible operational value as organizations gain a clearer view of the costs involved.

That pressure may help explain the proliferation of individual AI applications across organizations. Departments can adopt copilots, chatbots, and task-specific automations to address immediate needs. That creates increasingly varied technology environments.

LayerX Security’s State of AI Usage Report 2026 looked at this, as reported by The Hacker News. It found that AI activity is becoming distributed across personal accounts, browser extensions, embedded copilots, connectors, and other tools.

When AI meets fragmented data

The challenge becomes more consequential when AI encounters fragmented business information. McKinsey’s 2026 research on retail merchandising found that merchants reported spending about 40% of their time on lower-value activities. These include consolidating data, preparing reports, and reconciling information across systems.

The same research found that 71% of surveyed merchants said AI merchandising tools had produced limited or no effect so far. And 61% said their organizations were only slightly prepared or unprepared to scale AI across merchandising. These findings suggest that useful intelligence can depend heavily on how effectively commercial data and workflows connect.

That consideration is encouraging a broader conversation about AI as an orchestration layer. The National Retail Federation reports that 86% of surveyed retailers have AI governance policies. Meanwhile, 39% expect AI to represent more than 10% of technology spending within three years.

For retailers, this may place greater emphasis on connecting existing infrastructure with the people responsible for turning information into decisions and action.

Systems of record, systems of intelligence

Joseph Bullock, co-founder and CEO, and Anton Kulikalov, founder and CTO, lead FastQuery AI, an AI-native operating platform focused on retail operations and insights. They see this transition through the practical experience of employees working across complex technology environments.

Their perspective starts with a simple observation: modern retailers already possess extensive technological capabilities. These run from point-of-sale and inventory systems to ERP platforms, supplier networks, and analytics.

“The real opportunity of AI is to finally break down the silos between where decisions are made and where work gets done. I think there’s also a huge opportunity to finally simplify the technology ecosystem that retailers rely on,” Bullock states.

That philosophy gives FastQuery a particular role within this emerging field. Its work is oriented around connecting information from existing retail systems with the questions and decisions arising in retail execution and back office strategy.

Kulikalov says, “the data to make great decisions has always been there. The challenge has been how to make sense of it all, how to distinguish the signals from the noise, and how to answer the question, ‘what requires my attention today?’”

Where the work happens

The distinction matters because employees often interact with technology through the immediate needs of their role. That could be a product question from a customer, an inventory issue, a merchandising decision, or a task requiring information from several systems.

An intelligent layer brings relevant context into that interaction. It reduces the need for employees to open each underlying application separately.

Kulikalov describes the concept as a progression toward an enterprise where employees increasingly interact with an intelligent layer. The underlying systems continue operating in the background. It reflects a broader change in how people access information digitally.

Search engines reduced the need to go directly through individual websites for every question. Enterprise AI may similarly reduce the number of source applications employees need to interact with directly for routine information and decisions.

“We’ve spent decades building systems of record,” Bullock remarks. “The next opportunity is to build systems of intelligence that help people use all of that information where the work actually happens.”

One layer, many roles

For retail, that distinction can have practical implications. A category manager may need information about assortment, pricing, vendor performance, and demand at the same time. A store associate may need product specifications, availability, and instructional information while standing with a customer. An operations leader may need to understand conditions across locations before assigning priorities.

“Connecting these contexts can allow the same underlying data to become useful to different roles without requiring every employee to become an expert in the technology stack,” Bullock explains.

The workforce implications are similarly important. Kulikalov views AI as a means of expanding employee capability. It has particular relevance for roles that have historically absorbed significant amounts of information gathering and reconciliation.

“AI should make people more capable,” Kulikalov notes. “If we can give an employee the context they need at the moment they need it, we can change the amount of value that person can create.”

FastQuery’s role in this field also reflects the particular circumstances of independent retail. Large retail organizations may have extensive analytical resources dedicated to extracting insights from their technology environments. Smaller operators can have fewer people available to perform that work.

An AI layer that brings information together could potentially make sophisticated analysis more accessible. That applies to teams whose responsibilities already span multiple functions. The significance lies less in adding another destination for employees to visit. It lies more in making existing organizational knowledge and data usable within everyday workflows.

People still decide

That possibility also raises an organizational requirement: AI capability depends on people knowing how to work with it. Bullock notes that AI fluency and change management are important components of adoption. Integrating intelligence into daily operations can affect responsibilities, decision processes, and expectations across teams.

The technology can provide new capabilities. Organizations still need to determine where human judgment remains essential, how decisions should be reviewed, and how employees can develop confidence using new workflows.

The wider implication extends beyond any individual platform. As retailers continue investing in AI, the value of those investments may increasingly depend on how well intelligence travels across the organization. Connecting data to decisions, and decisions to execution, could become an important part of how businesses translate increasingly sophisticated technology into everyday productivity and customer experience.