AI Reshapes Call Centers: Human Touch Remains Critical

The landscape of customer service is undergoing a profound transformation, largely driven by the integration of artificial intelligence. For call center agents like Armen Kirakosian, who began his career nearly a decade ago, the shift has been monumental. He recalls the bygone era of manually jotting notes, navigating convoluted menus, and dealing with often frustrated customers. Today, AI-powered systems provide him with comprehensive customer profiles and even predictive insights into their potential issues before he even utters a greeting, allowing him to focus more intently on providing actual service rather than administrative tasks.

Kirakosian, a 29-year-old based in Athens, Greece, succinctly describes the impact: “A.I. has taken (the) robot out of us.” This sentiment reflects a broader trend affecting millions of call center employees worldwide. These individuals handle billions of inquiries annually, spanning everything from tech support for electronic devices to order fulfillment for retail goods. Kirakosian works for TTEC, a global company that provides third-party customer service solutions across 22 countries, serving diverse sectors such as automotive and banking that require augmented capacity or have outsourced their customer service operations.

Historically, call center work has been characterized by high stress and monotony, contributing to a significant attrition rate. According to McKinsey, approximately half of all customer service agents depart their roles within a year. A substantial portion of these interactions fall under the category of “break/fix,” where a customer presents a problem and expects an immediate resolution. The pivotal question now is whether these issues will be resolved by a human, an intelligent machine, or a human agent augmented by AI. While AI agents are increasingly handling routine tasks, the future job market for human agents remains a subject of ongoing debate, with predictions ranging from modest single-digit percentage losses to the potential elimination of half of all call center positions within a decade. However, it’s becoming evident that the industry will continue to require human involvement, perhaps even demanding higher levels of training and expertise, as certain customer service challenges remain beyond the current capabilities of AI.

AI’s Evolving Role and Human Indispensability

The integration of artificial intelligence into customer service operations marks a significant evolution, moving beyond simple automation to more sophisticated problem-solving capabilities. While AI promises increased efficiency and cost savings, its limitations, particularly in handling complex or emotionally charged interactions, are becoming increasingly apparent. Some financial institutions have experimented with fully AI-driven customer service, only to discover that the human element remains critical for overall customer satisfaction.

One notable example is Klarna, the Swedish “buy now, pay later” company. In 2023, Klarna replaced its 700-person customer service department with a combination of chatbots and AI. The initial results were a mixed bag. Although the company did realize financial savings, overall customer satisfaction metrics experienced a decline. Consequently, earlier this year, Klarna opted to rehire a number of customer service employees. This decision underscored a crucial realization: certain issues, such as identity theft or highly sensitive personal matters, are handled more effectively and empathetically by a human agent than by an artificial intelligence. This experience highlights a critical balance that companies must strike between technological advancement and maintaining a high standard of customer experience.

Gadi Shamia, an executive at Replicant, an AI software company specializing in training chatbots to emulate human speech patterns, articulated a forward-looking vision in an interview with McKinsey consultants. He suggested that the concept of an “AI-first contact center” is rapidly materializing. In this model, AI agents manage the vast majority of customer conversations, while a smaller, more highly trained, and better-compensated human workforce is reserved for tackling the most intricate and challenging tasks. This hybrid approach aims to leverage the strengths of both AI and human intelligence, optimizing efficiency for routine inquiries while preserving the human touch for situations that demand nuanced understanding, empathy, or advanced problem-solving skills.

Despite these advancements, the customer experience is still far from perfect. Initial customer service contact has long been managed by interactive voice response (IVR) systems. These systems guide callers with prompts like “press one for sales, press two for support,” evolving in the 2010s to allow simple voice commands such as “sales” or “support” rather than numerical selections. However, customers frequently express impatience with these structured menus, often resorting to “zeroing out,” an industry term for pressing the zero button repeatedly in hopes of reaching a live agent. Even after bypassing the automated system, callers often face further transfers and hold times if they don’t land in the correct department, contributing to ongoing frustration.

Streamlining Customer Journeys and Legislative Initiatives

The persistent frustrations associated with traditional IVR systems have prompted both technological innovation and legislative action aimed at improving the customer experience. Companies are actively developing and implementing more sophisticated telephone systems designed to comprehend customer service requests more broadly and accurately predict where to route a customer, ideally bypassing the need for cumbersome menu navigation altogether. This evolution is driven by a desire to make customer interactions smoother and more efficient, reducing the common pain points that lead to customer dissatisfaction.

OpenAI, the creator of the popular ChatGPT, is at the forefront of this shift with its upcoming “ChatGPT Agent” service. This advanced AI is designed to understand complex, multi-part requests, such as “I need to find a hotel for a wedding next year, please give me options for clothing and gifts.” Such capabilities signify a move towards more natural language processing and a more intuitive interaction model, where customers can articulate their needs in a conversational manner rather than conforming to predefined menu options. The goal is to make the initial point of contact as intelligent and responsive as a human agent, capable of understanding context and executing multi-step tasks.

Bank of America has also demonstrated increasing success in integrating similar features into its AI chatbot, “Erica,” which first debuted in 2018. When Erica encounters a request it cannot handle autonomously, it seamlessly transfers the customer directly to the appropriate human department, minimizing delays and frustration. Furthermore, Erica has evolved to become more predictive and analytical. For instance, it can recognize patterns such as a customer repeatedly having a low balance, prompting it to proactively suggest budgeting assistance, or identifying multiple subscriptions to the same service. This predictive capability allows for more personalized and proactive customer service, addressing potential issues before they escalate.

This month, Bank of America reported that Erica has been utilized over 3 billion times since its inception, handling an increasing volume of customer service requests. The chatbot’s name is a clever derivation from the last five letters of the company’s own name. James Bednar, vice president of product and innovation at TTEC, a company dedicated to enhancing customer service experiences, believes these advanced AI tools could eventually render traditional IVR systems obsolete. He envisions a future where the need for customers to “zero out” will cease to exist. Bednar states, “We’re getting to the point where AI will get you to the right person for your problem without you having to route through those menus.” This perspective underscores a future where AI acts as an intelligent conduit, ensuring customers reach the most qualified human agent or automated solution for their specific needs, thereby transforming the initial customer service interaction into a far more streamlined and satisfactory experience.

Policy and the Future of Call Center Employment

The rapid advancements in artificial intelligence and its widespread adoption in customer service have also drawn the attention of policymakers, particularly concerning the impact on domestic employment and the accessibility of human assistance. As companies increasingly rely on AI to handle routine inquiries, concerns have emerged about the potential displacement of human workers and the maintenance of quality customer service. Legislative efforts are now being introduced to address these challenges, aiming to strike a balance between technological innovation and consumer protection.

Aware of a collective impatience among American consumers with current IVR systems and the desire for human interaction, Democratic Senator Ruben Gallego of Arizona and Republican Senator Jim Justice of West Virginia have jointly introduced the “Keep Call Centers in America Act.” This proposed legislation seeks to mandate clear and accessible pathways for customers to reach a human agent when desired, preventing situations where callers are endlessly trapped in automated loops. Beyond ensuring human access, the act also aims to provide incentives to companies that choose to maintain their call center operations within the United States. This dual approach addresses both consumer frustration and the broader economic impact of call center outsourcing and automation on domestic job markets.

The intent behind such legislation is to protect the interests of both consumers and the American workforce. By requiring companies to offer clear options for human assistance, the act aims to preserve the quality of customer service, particularly for complex or sensitive issues that AI may not adequately handle. Simultaneously, by encouraging companies to retain call center jobs domestically, it seeks to mitigate potential job losses due to the rise of AI and offshore outsourcing. This legislative push highlights a growing recognition that while AI offers undeniable benefits in efficiency and cost reduction, a strategic approach is needed to ensure that these technological advancements do not inadvertently compromise the quality of service or negatively impact the labor market. The future of call centers will likely involve a dynamic interplay between advanced AI systems and a skilled human workforce, guided by policies designed to optimize both technological progress and societal well-being.