This article was based on the interview with NiCE CMO Michelle Cooper on the most common mistake brands make with AI and CX by Greg Kihlström, AI and MarTech keynote speaker for The Agile Brand with Greg Kihlström podcast. Listen to the original episode here:
In the executive suites and virtual boardrooms of every major enterprise, the conversation around AI in customer experience (CX) has reached a fever pitch. The pressure is immense. We’ve all seen the presentations promising unprecedented efficiency, automated resolutions, and dramatic cost reductions. The narrative, often pushed by technology vendors and eager consultants, is one of replacement—a future where human agents are a costly relic and algorithms handle the messy business of customer interaction. This technology-first approach is seductive, offering a seemingly straightforward path to modernizing the contact center and boosting the bottom line.
Yet, many of us who have been around the block a few times feel a nagging sense of unease. We’ve seen this movie before with other technology waves. A tool is adopted, implemented at great expense, and bolted onto existing processes, only to fall short of its transformative promise. The risk with AI in CX is even greater, because the stakes are not just operational efficiency but the very nature of the relationship a brand has with its customers. The alternative, as forward-thinking leaders are beginning to realize, is not to reject AI, but to fundamentally reframe its purpose. It requires shifting the starting point of the conversation from the capabilities of the technology to the nuance of human interaction—what Michelle Cooper, Chief Marketing Officer at Nice, calls the “customer moments that matter.”
The Foundational Flaw: Starting with the Tool, Not the Outcome
The most common and costly mistake in enterprise AI adoption is putting the cart before the horse. A new generative AI tool or automation platform becomes available, and the immediate mandate is to find a use for it. The project becomes a technology implementation, measured by deployment speed and adoption rates, rather than a business transformation measured by customer outcomes. This approach almost inevitably leads to automating existing, often inefficient, processes, rather than reimagining the experience from the ground up.
Cooper cuts through the noise with a simple, yet profound, observation on where the strategy should begin.
“A lot of companies are kind of approaching this AI era from a technology, right, from a tool perspective. And yes, the tool and the technology that you you pick and select is critical in your journey, but where you really need to start is in thinking through like what are the customer moments that matter. What are the business processes that you’re trying to evolve? What are the outcomes that you’re ultimately trying to get to? And a lot of times we see, you know, organizations start with the technology first versus the the outcomes.”
This is a call to return to first principles. Before a single line of code is written or a vendor is selected, leaders must ask foundational questions. Which interactions define our brand promise? Where do our customers experience the most friction? What moments, if handled exceptionally well, can turn a frustrated customer into a lifelong advocate? A customer-moment-first strategy means mapping these critical junctures and then, and only then, asking how AI can elevate them. It might mean using AI to seamlessly resolve a simple query without human intervention, but it could also mean using AI to arm a human agent with the perfect information to handle a complex, emotionally charged situation with empathy and precision. The tool serves the moment, not the other way around.
Empowering Judgment, Not Just Replacing Agents
The siren song of efficiency often leads to a singular focus: reducing human headcount. While there are certainly routine, repetitive tasks that are prime candidates for automation, viewing frontline agents as a cost center to be minimized is a strategic miscalculation. These individuals are the voice of your brand. In an increasingly digital world, a live conversation with an agent might be the only real human interaction a customer ever has with your company. Devaluing that touchpoint in the name of automation is a dangerous game.
The more sophisticated approach is to view AI as a force multiplier for human talent. The job of a customer service agent is, as Cooper rightly notes, “a heroic, heroic job.” They are expected to be product experts, problem solvers, and brand ambassadors, often while navigating multiple clunky systems and dealing with frustrated customers. AI’s greatest immediate value may lie in making that heroic job more manageable and more effective.
“It’s not about just automation for automation’s sake, but it is really focused on how and when you’re able to deliver a better customer outcome. And a lot of times, yes, there’s cost savings, yes, there’s efficiency, but it’s really more about what are those interactions and those special moments that you can really create for your customers…that at the end of the day makes a difference between them continuing to value and select your brand or to possibly make another decision.”
In practice, this looks like an agent co-pilot. Instead of toggling between seven different screens, the agent is presented with a unified view of the customer, including their history, sentiment, and the likely intent of their call. As the conversation unfolds, AI can provide real-time coaching, surface relevant knowledge base articles, and recommend next-best actions. It can automate post-call work like summarizing notes and scheduling follow-ups, freeing the agent to focus entirely on the customer in front of them. This is not about replacing human judgment; it is about augmenting it with perfect information and context, allowing agents to operate at the top of their capabilities.
The Litmus Test: Transformation vs. Automation
If there is one key distinction that separates successful AI initiatives from the ones that fizzle out, it’s the difference between simple automation and true transformation. Automating a broken or inefficient process just helps you do the wrong thing faster. The arrival of powerful AI tools should be seen as a catalyst to question everything about how customer service is delivered.
This is arguably the most challenging part of the journey because it requires leaders to move beyond the comfort of their existing operational models. It’s not an IT project; it’s a fundamental rethinking of how work gets done, who (or what) does it, and how value is measured.
“Often times we see the biggest challenges is just trying to take AI and automate their existing business processes, right? And and almost treating it as a technology project versus a transformation. And I think one of the things that is so powerful about AI today is you can completely reimagine the way that work gets done. What’s handled by AI agents, um, versus human, how to make those, how to balance, you know, the the handoff, um, and the orchestration between those two worlds.”
Consider the example Cooper shares of Lufthansa, which now automates 70% of its tier-one and tier-two interactions with agentic AI. This move doesn’t just create efficiency. It strategically shifts their human agents away from transactional tasks toward handling more complex, higher-value interactions where their expertise truly matters. This reimagining of the work elevates the role of the agent, improves their job satisfaction, and ensures that the most critical customer moments are handled by well-equipped, focused human experts. This is the essence of transformation—not just doing the same things more cheaply, but creating a more intelligent, resilient, and human-centric operating model.
The journey to AI-powered CX is not a technology race; it is a strategic marathon. The leaders who win will not be the ones who adopt the most tools the fastest, but those who exhibit the discipline to start with the customer, the vision to empower their people, and the courage to fundamentally reimagine their processes. This approach is undoubtedly more complex than simply plugging in a new piece of software. It requires cross-functional collaboration, a willingness to challenge long-held assumptions, and a commitment to measuring what truly matters—resolution quality, customer trust, and long-term loyalty.
Looking ahead, the conversation will evolve even further. As Cooper astutely predicts, we will soon move past discussing AI as a feature and begin grappling with it as a new operational reality. The challenge will become managing a hybrid workforce of human and AI agents, seamlessly orchestrating the handoffs between them to create a single, coherent customer journey. The organizations that have already done the foundational work of building a customer-moment-first strategy will be positioned not just to compete, but to define the next era of customer experience. For them, AI will not be a source of anxiety, but a powerful engine for building more meaningful and durable connections with the people they serve.
