AI delivers over 50% improvement in customer service KPIs for top-performing firms: EY report

Only 23% of organisations have managed to realise high or very high value from artificial intelligence (AI) in customer service, even as adoption of the technology becomes widespread across businesses. These organisations, classified as ‘Achievers’, have reported improvements of more than 50% across three to four core customer service metrics, including customer experience, revenue uplift, average handling time, and agent productivity, according to a new report by EY India.

The report, titled Demystifying AI in customer service: Navigating AI-led transformation to value realization, highlights the gap between AI adoption and measurable business outcomes. Based on more than 100 survey responses and executive interviews with customer service leaders across industries and geographies, the study examines how organisations are deploying AI and the factors influencing the returns from these investments.

EY has classified the surveyed organisations into three categories based on their reported value realisation. Achievers account for 23% of respondents and represent companies reporting high or very high value from AI. Challengers, accounting for 33%, have reported moderate value realisation, while Aspirants, at 44%, have reported low or very low value realisation.

While Achievers reported improvements of 25%-50% or higher across multiple customer service key performance indicators (KPIs), Challengers typically recorded improvements of 10%-25% across most metrics. Aspirants, meanwhile, reported improvements of less than 10% across most KPIs or had not measured the outcomes.

The findings indicate that deploying individual AI tools does not necessarily translate into business value. Organisations reporting higher returns are more likely to integrate AI across customer acquisition, service delivery and operational management instead of limiting its use to isolated applications such as chatbots and conversational analytics.

Voice AI and knowledge management emerge as key differentiators

The report found that voice AI adoption is significantly higher among organisations reporting stronger value realisation. Around 91% of Achievers have scaled voice AI deployments. Within this group, 45% use rules-based voice bots, while 37% have implemented advanced voice bots with AI-driven classification and prioritisation capabilities.

In comparison, more than 80% of Challengers and Aspirants continue to depend on human agents or basic menu-driven interactive voice response (IVR) systems for classifying and prioritising customer queries.

AI copilots are also finding applications in customer service operations. Around 45% of Achievers use conversational AI-assisted agents during issue diagnosis and resolution, while 36% have deployed agentic AI copilots. The corresponding figures point to a greater use of AI in assisting service teams rather than relying entirely on human judgement and static information repositories.

Knowledge management has emerged as another important differentiator. According to the report, 73% of Achievers use AI-powered dynamic knowledge capabilities, compared with 20% of Challengers and just 8% of Aspirants. Such systems allow customer service agents and AI applications to access and use relevant information during customer interactions.

The report also examined how organisations use AI to collect customer feedback. Around 36% of Achievers use conversational AI-based surveys, while 18% use unified AI-recommended sampling approaches. In contrast, between 75% and 90% of Challengers and Aspirants continue to rely on traditional link-based surveys.

Customer experience remains the primary objective

Improving customer experience was the leading objective among organisations reporting the highest value from AI-led transformation. Around 45% of Achievers identified customer experience improvement as their primary objective, ahead of revenue growth and cost reduction.

On average, Achievers have scaled AI across approximately 60% of core customer service technology domains. Challengers and Aspirants, however, remain more focused on evaluating or piloting individual use cases rather than integrating AI across the customer service lifecycle.

The report also highlights differences in the challenges faced by organisations at different stages of AI adoption. Among Achievers, 38% identified data quality and availability as their primary challenge, while 31% cited a lack of financial investment. In contrast, more than 40% of Challengers and Aspirants identified governance, compliance and change management issues as key barriers.

The findings also point to a longer investment horizon among organisations reporting higher value realisation. Around 80% of Achievers began adopting AI in customer service more than two years ago, compared with 50% of Challengers and 64% of Aspirants. However, the report notes that early adoption alone does not guarantee better outcomes, with the ability to scale AI and integrate it into business processes also playing a role.

Further, 27% of Achievers expect returns on their AI investments to take more than two years, while none expect returns within six months. This compares with 30% of Challengers and 17% of Aspirants who expect returns within six months.

Commenting on the findings, Preeti Anand, Partner, Business Consulting, EY India, said, “While AI adoption in customer service has become widespread, value realization remains uneven, with many still struggling to move beyond fragmented use cases. Our research shows that organizations realizing the greatest value are building strong foundations, integrating AI across workflows and, importantly, measuring success through business outcomes. The next competitive advantage will come from how effectively firms connect intelligence across the customer journey and turn it into sustained business value.”

Arun Nagarajan, Partner, Technology Consulting, EY India, said, “Many organizations view AI or technology deployment as the key success driver of transformation. Our findings suggest that organizations reporting the strongest outcomes have invested first in data, knowledge and governance foundations, creating the context AI needs to operate effectively at scale. As AI maturity increases, these foundational capabilities become increasingly important for long term value realization.”