By Indranil Mukherjee

 

Today’s customer can be impatient and highly fickle: in a recent U.S. consumer survey, 70 percent of respondents said they would abandon a brand after two negative experiences. In fact, even a complicated purchasing process can cause many buyers to drop out.

Since traditional methods of engagement are largely unable to keep up with consumers’ expectations of highly personalised services, real-time support, and frictionless interactions, many organisations are adopting artificial intelligence (AI) technologies to bridge the gap.

The good news is that with the emergence of Agentic AI, the possibility of elevating customer experience has gone up manifold. There is an opportunity for organisations to leverage AI agents to create seamless, real-time, AI-first customer experiences that stand out for their adaptiveness and intelligence.

Predict, Anticipate, Act

Agentic AI systems are able to understand, predict, analyse and respond to various triggers, such as a new customer need or change in competitor pricing, with little or no human intervention. Drawing on advanced machine learning capabilities, the systems interpret context or anticipate future events with accuracy; for example, they can analyse customers’ information – purchase history, stated preferences, browsing activity etc. – to understand their immediate situation and tailor their responses accordingly. Based on these insights, they can also predict a customer’s need even before it is fully articulated, and address it proactively: for example, sensing that a customer is browsing through the refund policy on the company website, an AI agent automatically sends them a link to initiate a return.

As opposed to traditional AI, which is rule-based and somewhat static, agentic AI is highly dynamic, adapting its response to changing circumstances in real-time. It can orchestrate various systems – CRM, payments, stock management, logistics – to resolve complex, multi-step problems independently.  AI agents are goal-oriented, modifying interactions based on the situation and making informed decisions to achieve specific outcomes – for example, deciding the best sequence of actions for resolving a customer’s problem expeditiously. What’s more, they are available to customers any time of day or night, ready to solve their problems much faster than any human being can.

Personalise to Please

Personalisation in customer interactions hits a new high with agentic AI, which can enable conversing with customers in their preferred language. Agents can also be quite empathetic – for example, encountering an irate customer, an AI agent may speak soothingly to defuse the tension; when attending to a not very tech-savvy caller, it will use simple, non-jargonised language and multimedia aids to help them with their problem. Agentic AI exceeds earlier AI-based personalisation tools by adapting interactions to customers’ preferences and behaviours to deliver bespoke, engaging experiences.

Trust is Paramount

As a key component of an AI-first strategy, agentic AI helps businesses engage customers in personalised, intelligent and seamless interactions in real-time to drive satisfaction and loyalty. But it also introduces some major concerns that need to be addressed quickly. Declining trust ranks right at the top.  In the seventh edition of the State of the AI Connected Customer report from Salesforce, 61 percent of customers said that the adoption of advanced AI solutions, including agentic AI, underscored the importance of trustworthiness. Worryingly, 64 percent – nearly two in three – believed companies used customer data recklessly, and only 42 percent thought they would use AI in an ethical manner.  Further, 72 percent of respondents felt it was important to know if they were being served by an AI agent.

Hence, organisations embracing agentic AI should prioritise responsible adoption ahead of other considerations. Transparency is critical: businesses need to inform customers if the agent speaking to them is not human; they should also be ready to answer questions about how AI agents make decisions. It is important to align AI strategy with organizational values.

A responsible AI framework can not only guide them in the right direction, but also ensure that they develop and deploy AI solutions as per regulatory and ethical expectations – protection of data privacy, avoidance of bias, fairness and accuracy in AI-driven outcomes, etc. Also, organisations should plan for manual oversight of agentic AI systems, to provide customer experiences that are not only highly impactful but also human-centric.

 

 

(The author is Indranil Mukherjee, EVP and Service Offering Head, Infosys Salesforce Services, and the views expressed in this article are his own)