News

Rohit Kedia is a software architect turned CEO of Xoriant, a AI-native Applied Intelligence digital engineering company.

gettyFor years, retailers have invested heavily in building a unified view of their customers, bringing together interactions across channels to better understand their behavior. But in engineering over 350 platforms across Fortune 500 and Global 2000 companies, and looking closely at customer data and experience challenges, one distinction became clear: Knowing what a customer has done isn’t the same as understanding what they’re likely to do next.

I call this the intent gap.​

The problem isn’t a lack of data. It’s the inability to connect signals across touchpoints and translate them into meaningful insights about customer intent. For retailers, bridging the intent gap can help them move beyond simply tracking customer interactions to making better-timed, more informed decisions that improve retention.​

Limitations Of The Unified Customer ViewRetail has already moved through two eras of customer intelligence. The first connected point-of-sale systems, loyalty platforms and CRM data into a single record of who the customer is and what they bought, promising that seeing every transaction would be enough to make sound decisions. The second added activity-level insights on interaction data—emails opened, pages browsed and calls logged—on the assumption that more touchpoints meant more understanding.

Both were only partially right.

Per Salesforce’s 2025 State of Sales research, 87% of respondents say data silos and trapped data affect their ability to build a unified customer view,​ with more than one-third of that percentage experiencing a severe impact. Separate research by IDC found that up to 90% of enterprise data was unstructured in 2022, living in emails and transcripts rather than a CRM. That’s an architectural problem.​​​

Identifying The Intent GapCustomers rarely announce their real motivations directly. Some intent is explicit, like a search for a specific product or a request for a quote. But much of what matters is never stated at all. These implicit motivations, revealed only through behavior, show up in price hesitation as an abandoned cart, brand fatigue as declining open rates and competitor shopping as shorter sessions, long before any of it becomes a stated preference.

Closing this “intent gap” means connecting fragmented signals to understand customer intent, uncover unmet needs and use those insights to predict behavior, take action and deliver measureable results, rather than just populate another dashboard.

This is where the context layer plays a key role. The same signal, like a customer’s login frequency dropping, can mean churn, a completed purchase, seasonality or a changed need. The surrounding context tells you which of these may be true.

AI is beginning to both expose the intent gap and hold the power to close it.

Agentic AI-Powered System Of IntentUnlike the previous two eras—identifying who the customer is and what they did—the intent era helps us understand what’s likely to happen next, why and what should be done about it. When paired with agentic AI, intent-driven intelligence runs as a continuous reasoning loop, monitoring changing conditions, investigating anomalies, testing a live hypothesis against new data rather than scoring a snapshot and surfacing a recommendation traceable back to the signals that produced it. This is a departure from most retail AI today, which is predictive—a model trained on historical data that produces a score.

Either way, true intent intelligence only creates value if it’s acted upon.

What AI + Intent Intelligence Can Make PossibleSalesforce’s 2024 Connected Shoppers research found that 89% of retailers expect returns on their AI investments, with employee productivity topping the list followed by better shopping experiences.

An intent-aware retailer can identify churn risk in the erosion of engagement that precedes it— before a complaint surfaces. They can detect buying momentum even before a cart is started. They can catch a launch or promotion drifting off course before it’s too late to regroup.

No matter where intent is spotted, the objective isn’t better customer intelligence for its own sake. It’s earlier intervention, more relevant experiences and measurable business outcomes.

The Role Of Human Judgment In Deriving IntentIntent intelligence doesn’t replace human judgment; it amplifies it. AI can surface patterns and recommend actions at a scale no team could match, but it can’t supply the context of a relationship, the domain expertise to spot a genuine risk or the ethical oversight to act on a sensitive signal. Those remain distinctly human, making it the product of human ingenuity and AI working together.

One lesson I’ve learned is that customer data rarely tells the whole story on its own. Whether it’s a retailer that sees a loyal customer making fewer purchases and assumes they’re losing interest, or an e-commerce business that interprets repeated visits to a product page without a purchase as a lack of buying intent, the real reason may be in signals they haven’t connected.

​For retailers, interpreting these signals and closing the intent gap isn’t just another data or technology initiative. It’s about getting closer to the customer, making more informed decisions and earning the next interaction rather than taking it for granted.

Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?