Contact Center AI Hits a Measurement Crisis as Old Metrics Fail to Capture Real Value

Customer service operations are transitioning away from experimental AI deployments toward production systems, yet a fundamental problem has surfaced: the metrics used to evaluate success for decades no longer apply. Knowledge management capabilities are now the deciding factor in whether AI investments deliver tangible business results. Contact centers present an ideal testing ground because they handle enormous call volumes, carry substantial payroll expenses, and create direct customer interactions where failures become immediately visible. These conditions make them the most transparent arena for measuring whether AI technology produces genuine returns.

The shift is driving how technology vendors market their solutions, ranging from Cisco Systems Inc.'s focus on AI agents embedded throughout collaboration and customer engagement workflows to multiple announcements of voice agents designed for cloud-based contact center platforms. Gartner forecasts that conversational AI will reduce contact center labor expenses by $80 billion in the current year. Bob Laliberte, principal analyst for networking and observability at theCUBE Research, emphasized why these environments deserve close examination. "Contact centers have large volumes of interactions. There [are] significant labor costs and direct moments of truth with customers," Laliberte explained. "When something fails, the consequences are also pretty highly visible, and a poor AI interaction can increase customer effort. Damage trust and ultimately hurt the brand."

Laliberte and Zeus Kerravala, principal analyst and founder of ZK Research, a division of Kerravala Consulting, discussed these dynamics during theCUBE's coverage of "The AI ROI in Contact Center Summit." TheCUBE, SiliconANGLE Media's livestreaming studio, examined how AI is reshaping the economics of customer experience.

Knowledge management and new metrics for AI value

Conventional measurement frameworks that have guided the industry for generations are beginning to fail. Traditional benchmarks emphasize speed and resolution containment, sometimes at the expense of actually solving customer problems. This misalignment is pushing organizations toward metrics centered on outcomes rather than process efficiency, according to Kerravala. "Historically, we've measured success in the context of things like average handle time, first call resolution," he noted. "And those metrics don't … matter as much anymore. We've had such a focus on average handle time in this industry, but is that shorter call actually a good thing if the issue remains largely unsolved?"

Transitioning from AI assistants to fully autonomous agents demands more rigorous testing, oversight mechanisms, and data infrastructure. Vendors such as Five9 Inc. have deployed implementation guides and voice AI agents to accelerate deployment timelines, yet the responsibility for foundational work remains with the customer organization. "Organizations should be listening for some practical answers on the importance of high data quality, the integrations that need to be done … being open to redesigning their process," Laliberte stated. "There [are] some issues around knowledge management that need to be addressed as well."

https://www.youtube.com/embed/95HmL7YVi4I?feature=oembed

The path forward depends on three interconnected elements: knowledge management infrastructure, workflow redesign, and the coordination between automated systems and human agents. "It's important to understand that the brands that lead will not be the ones that simply automate the most," Kerravala emphasized. "They will be those that turn AI into a better, more consistent set of outcomes, but will also be able to earn employee and customer trust."