What’s the next big use case for AI now that many retailers have optimized their back-office workflows? In 2025, more than 40% of retail and CPG companies made operational efficiency their top AI budget priority. By 2027, the top AI spend priority will be service and product innovation to drive “revenue growth and margin improvements,” according to an IBM Institute for Business Value research brief. That shift in spending priorities shows how the competitive landscape is changing again.
by Manish Sharma
The retail leaders setting the pace of transformation are using AI to draw on their massive customer datasets to create individualized real-time customer experiences. This shifts AI from a reporting and automation tool to an instrument for converting intent signals into profitable actions.
The intent-signal edge for real-time experiences
Shoppers will switch tabs in seconds to find what they need. The competitive edge comes from knowing what each customer is trying to accomplish and helping them do it. This is the 360-degree customer view that marketers and CX teams have pursued for years. AI data cloud platforms make it possible to connect internal customer data with external competitive intelligence and act on that information in real time.
Now, retailers can identify behavioral patterns. Instead of relying on demographics, retailers can use behavioral insights to see why customers return items, abandon carts, make repeat purchases, or churn. Those insights allow for targeted actions. For example, when a customer shows churn signals, an AI agent can offer a personalized incentive to stay and buy. If an item in a cart shows return potential signals, an agent can offer support such as virtual try-on or “see it in your home” tools before the order is finalized.
These agents can be part of an ecosystem of tools that align website experiences with each customer’s intent. Models can predict visitor intent based on signals from trillions of data points, including hover time, scroll depth, and search history, and then adjust the site’s layout, offers, messaging, and search results in real time. This gives each customer a unique digital storefront that matches their intent and gives them more reasons to stay and buy.
Move the customer from “Want” to “Buy” faster
The goal of a mature e-commerce operation is to reduce the distance between a customer’s want or need and their purchase. Agentic ecosystems are the key because each agent can focus on one step of the journey while communicating context across the system.
When a customer arrives at a site:
An autonomous product discovery agent can present a storefront featuring products tailored to the shopper’s budget, style, size, and feedback.
A pricing agent can use real-time competitive data, inventory levels, and consumer intent to offer the customer a price that leads to purchase without sacrificing margin.
A promotion agent can decide who needs an incentive to convert and generate targeted offers rather than one-size-fits-all deals.
A comparison agent can display live competitive price checks to keep price-sensitive customers on the site.
A journey optimization agent can surface the right products as the customer moves around the site and remove micro-frictions in checkout to drive purchase completion.
Together, this agent ecosystem gives customers what they want faster, with less effort, and makes it easier for them to buy.
Plug leaks to protect revenue wins
The revenue growth that AI tools can provide also depends on keeping that revenue from evaporating. Operational leakage cost retailers $12 billion in 2025, and preventable fraud and policy abuse siphoned away $100 billion. Now, leading-edge retailers are using AI to detect and reduce these revenue leaks.
AI can track the order-to-cash flow in real time to flag issues with orders, payments, and warehouse updates. Catching and addressing problems as soon as they occur can keep customers from canceling their orders when they realize something’s gone wrong. AI models trained on fraud-related pattern recognition, behavioral signals, and other data can flag potential fraud at the order stage and during returns, with step-up scrutiny for those interactions.
Another anti-fraud use for AI is brand and product monitoring across third-party marketplaces and social media platforms. AI tools that detect impostors and listings that aren’t brand-compliant can protect the brand’s reputation and customer trust. That can help maintain lower average customer acquisition costs and increase average customer lifetime value.
From customer signals to revenue growth with AI
Shifting AI focus from back-office operations to CX for revenue growth requires a strategy that relies on four pillars.
First, as with all AI projects, retailers need to unify and structure their data so it’s easy for AI agents to use.
Second, marketing and CX teams need to consider individual behaviors and intent rather than marketing to more general segments or personas.
Third, marketers can use AI-powered insights not just to drive sales but also to solve customer problems. That can create a more empathetic customer experience that builds loyalty and can raise Net Promoter Scores.
Fourth, retailers need to protect their revenue gains with AI-powered monitoring to reduce order-to-cash problems, prevent fraud, and maintain brand consistency across platforms.
Retailers who embrace AI’s potential for intent signal detection and process orchestration are the ones most likely to deliver the level of real-time personalized shopping that customers will quickly come to expect. AI’s capacity to provide meaningful context to the customer journey enables retailers to create shorter, more rewarding paths to purchase and drive revenue growth.
Manish Sharma is CRO & Industry Head: Retail, Manufacturing & High Tech at eClerx. Manish leads the company’s global revenue strategy and execution. He is responsible for driving sustainable growth across eClerx’s diverse portfolio of services, go-to-market strategy, forging strategic alliances and partnerships, and nurturing long-term client relationships.
Photo by Jakub Żerdzicki on Unsplash
