This week, I found myself in the backseat of a Waymo, cruising silently through San Francisco, hands-free, stress-free, human-free. Initially, I was skeptical, but within minutes, I forgot there wasn’t anyone upfront. It was quiet, efficient, predictable, and everything I wished for during a ride hailing experience. I realized I didn't miss the human driver. At all.

Then it happened: a brief hesitation, a minor delay as the AI navigated around an erratic driver doing a u-turn in front of us. My knee-jerk reaction was disappointment. Here was a machine that had flawlessly navigated complex city streets, and yet my trust wavered at the slightest hint of imperfection.

Paul Adams, Intercom’s Chief Product Officer, shared a similar insight in a recent post:

Welcome to the first edition of The Intelligent Experience, where we'll explore the evolving landscape of customer experience, agent responsibilities, and leadership expectations in a world increasingly powered by artificial intelligence!

Customers: Convenience over humans

In today's customer experience landscape, convenience often beats waiting for a human–until it doesn’t. We're witnessing a fascinating paradox: customers want AI-driven interactions to be quick, seamless, and flawless, yet they're unforgiving when technology falls short. It's not just efficiency they're after; it’s perfection. And the moment AI makes a mistake, trust evaporates faster than you can say "agent please."

AI is expected to be “perfect”

Zendesk’s research captures this tension brilliantly: 68% of consumers expect AI to match or surpass human expertise, yet nearly half can't reliably tell the difference between human and AI interactions. This reveals an intriguing irony: customers demand superior performance from AI but often don’t even recognize when it’s delivering.

Social proof also plays a surprising role. Customers trust AI more when they see evidence of others’ positive experiences. A simple statement like “1,000+ people solved their issue with this solution” reassures users and bolsters AI credibility.

We've grown accustomed to frictionless technology–instant apps, predictable deliveries, and effortless services, which has recalibrated our expectations across all interactions, digital or otherwise. But this seamless digital experience has a downside: even minor hiccups by AI are magnified, leading to swift dissatisfaction and erosion of trust.

Pulling in human heroes

Yet there's a flip side: as much as we crave convenience, we yearn for connection when emotional stakes climb. Sure, we’re comfortable with AI handling simple queries (e.g., tracking packages or resolving minor issues), but when things get complicated or emotionally charged, we instinctively seek human reassurance and assistance. This creates a complex balancing act: companies must provide speed and efficiency alongside empathy and technical expertise.

Take, for example, RB2B, a fintech company experiencing rapid growth. They implemented Intercom’s Fin to streamline their customer support processes. Initially, this automated solution was hugely successful, handling 65% of queries. Customers responded positively, praising the quick and efficient resolution. Yet, as effective as it was at handling routine tasks, the AI stumbled when facing unusual or complicated customer scenarios, causing frustration and eroding trust. (Don’t worry, they improved things in the end!)

The RB2B case highlights a critical reality businesses face today: AI excels at predictable, repetitive tasks but often falls short when nuanced understanding is required. Customers don't want just answers; they want reassurance that their unique concerns are heard, understood, and genuinely cared about. The real challenge for businesses is ensuring seamless transitions from AI-driven efficiency to human-driven expertise.

Ultimately, AI and human agents aren’t competitors; they’re teammates. Successful companies thoughtfully decide when and how each should step forward, continuously refining these decisions based on real-time feedback.

The personalization paradox

To add more complexity, there's the personalization paradox. Customers expect companies to "just know" them (i.e., their preferences, history, and unique circumstances) but simultaneously express heightened concern about data privacy. They're quick to label personalized targeting as "creepy" or invasive if it crosses an undefined boundary. Transparency isn't optional; it’s essential. Customers want control over their data and reassurance that it’s used responsibly. Companies like Spotify and Netflix navigate this brilliantly, using personalization to enhance experiences without triggering privacy anxieties.

The key insight here is deceptively simple: convenience may draw customers in, but trust is what keeps them coming back. Companies must deliver flawless, frictionless AI interactions for routine tasks and compassionate human support when complexity or emotions rise.

Navigating this paradox–high expectations paired with conditional acceptance–means carefully crafting experiences that transparently blend AI efficiency with human empathy. Done right, customers don’t just tolerate these interactions, they embrace them with fierce loyalty.

As my Waymo ride once showed me, the best technology quietly fades into the background. Customers forget they're interacting with technology at all until the moment they need genuine human connection (like a stuck ride), and it's right there, effortlessly available. That’s the future of customer experience: invisible technology, available human expertise.

Agents: From managing tickets to managing systems

Support agents were once considered only frontline troubleshooters, solving customer problems one ticket at a time. Today, the role of the customer support agent is radically transforming. Instead of merely answering calls or emails, agents are becoming sophisticated orchestrators of technology and knowledge engines, fundamentally redefining what it means to work in customer support.

At first glance, AI's growing prominence might seem like a threat to the traditional support role, but the truth is far richer and more nuanced. AI isn’t replacing agents; it's enhancing their importance, reshaping responsibilities, and amplifying their impact. Think of support agents today less as troubleshooters and more as conductors, skillfully guiding a symphony of technology-driven interactions to deliver harmonious customer experiences.

Human handled cases are getting more complex

But here's the catch: as AI takes on the routine, agents are increasingly faced with complex and emotionally charged interactions. Basic inquiries, like account balances or password resets, now comfortably belong to AI chatbots, leaving agents to handle the nuanced, tough, and often stressful conversations. Salesforce's latest State of Service report vividly illustrates this reality, revealing that 77% of agents see their roles as significantly more complicated than just a year ago. This complexity doesn't just challenge agents, it elevates the emotional labor, stress, and burnout they experience.

This creates an intriguing situation: on one hand, AI tools reduce monotony by streamlining straightforward tasks. On the other hand, the problems agents now tackle are increasingly complex, demanding deeper expertise and sharper analytical skills than ever before. The stakes, and the pressure, have never been higher.

Evolving agents’ skills

Yet complexity also brings opportunity. Agents are no longer mere problem-solvers, they’re increasingly becoming proactive, strategic contributors. They serve as critical sources of real-time insights, feeding back valuable data into AI systems to enhance performance. This human layer of insight is indispensable because no algorithm can fully capture human nuance independently.

To meet these evolving demands, agent training programs must radically shift focus. Historically, training emphasized process and procedure; now, programs must cultivate emotional intelligence, advanced empathy, technical proficiency, and strategic problem-solving. The new standard isn't just teaching agents what to do but empowering them with how to think.

This targeted development significantly boosts agent morale and engagement. Agents feel valued when organizations invest in their professional growth, shifting from "call handlers" to integral contributors shaping customer success and organizational strategy.

Forward-thinking organizations understand this necessity. For example, fintech firm KOHO adopted AI-driven analytics to automate quality assurance across all interactions, enabling precise, targeted agent training. The result? Improved skills, heightened effectiveness, happier agents, and better customer outcomes.

This evolution is reshaping career trajectories in customer support. Clearly defined paths from entry-level roles to specialized positions (e.g., Program Managers, Knowledge Managers, Data Analysts) help attract and retain high-caliber talent who recognize the long-term potential within customer support.

Cultivating well-being and security

However, risks remain. Increased complexity and expectations can lead to burnout and turnover if not carefully managed. To mitigate this, organizations must prioritize agent well-being, offering flexible work arrangements, regular morale check-ins, and balanced workloads. Cultivating a culture where seeking support is normalized, and not stigmatized, is essential.

Additionally, there's natural tension between agents and new technology, driven by fears of job security. Companies successfully managing this transition emphasize transparent communication, involve agents early in AI initiatives, and clearly outline how automation will enhance–not diminish–their roles.

The future of customer support doesn't belong solely to AI or to humans, it belongs to organizations adept at harmoniously integrating both. Agents empowered by AI can deliver experiences that delight customers, drive loyalty, and foster meaningful professional growth. As we continue navigating this evolving landscape, businesses must remain committed to equipping their agents with the tools, training, and support they need to thrive.

Leadership Expectations: From Cost Centers to Strategic Storytellers

Customer support used to be a necessary but overlooked cost center. Think of it like the appendix of a business: there but easily forgotten until something goes wrong. Today, savvy leaders realize that customer support isn't just essential; it’s a strategic powerhouse, capable of driving revenue, loyalty, and deep insights.

The shift is clear worldwide: customer support no longer sits quietly on the sidelines but is front and center in strategic discussions. Executives now expect support teams not just to manage expenses but to actively influence revenue growth, retention rates, and brand strength.

What's causing this strategic shift? Rising customer expectations. Exceptional service now decisively distinguishes leading brands. Qualtrics research underscores this: remarkable customer service often outweighs price as the key factor keeping customers loyal or prompting them to jump ship.

Metrics that matter

The metrics we use to gauge customer experience are changing radically. And it's about time.

Historically, customer support metrics have been stuck in the operational weeds: average handle time, call volume, first-call resolution. Useful, sure, but often misleading. Shorter calls don't always mean happier customers; fewer calls don't always signify fewer problems. These metrics were easy to measure but dangerously easy to misinterpret.

Today, smart executives understand that exceptional support isn’t just about efficiency, it’s about effectiveness and meaningful outcomes. They're shifting focus from purely transactional metrics toward a richer, more holistic understanding of customer experiences. But there's a catch: traditional metrics like Customer Satisfaction (CSAT) and Customer Effort Score (CES), while helpful, often fall short in capturing the complex interplay between AI-driven and human interactions.

CSAT, for instance, can be skewed by superficial factors (a friendly agent with a poor solution can score highly). CES is closer to the truth–after all, customers universally hate hassle–but still misses nuances around emotional context and AI-specific friction points. Customers might rate an AI interaction poorly not because it was ineffective, but because they felt uneasy or mistrustful engaging with a machine.

Forward-thinking leaders are now advocating for metrics that blend emotional nuance, interaction complexity, and resolution effectiveness across both human and AI interactions. They're exploring deeper insights: sentiment analysis across interactions, proactive resolution rates, AI-to-human handover effectiveness, and customer trust scores related to AI experiences.

Rather than relying solely on CSAT or CES, they're crafting data-driven narratives: stories that vividly link customer experiences directly to business outcomes like loyalty, retention, and lifetime value. The real power isn't in any one metric, but in how multiple measures combine to tell a meaningful story.

Nick Mehta, CEO of Gainsight, exemplifies this new approach, regularly elevating customer success metrics into strategic discussions with the board. Gainsight ensures that every executive meeting prominently features customer success data, emphasizing how customer-centric metrics directly influence financial performance and strategic goals.

Leveraging predictive analytics

Another shift is toward the proactive management of operational risks. Today's executives no longer accept purely reactive customer support models; they demand forward-thinking, predictive capabilities. Boards expect support teams to anticipate customer needs, forecast potential disruptions, and implement strategies to prevent issues before they become costly problems. This forward-looking approach aligns closely with broader organizational resilience and risk management objectives.

To meet these expectations, support leaders increasingly leverage AI and predictive analytics, using real-time customer interaction data to identify emerging trends, forecast volume spikes, and proactively address potential issues. Companies like fintech firm KOHO exemplify this, using AI to spot recurring customer issues early and proactively collaborating with product teams, reducing customer friction dramatically.

Risks of AI

Yet, AI introduces ethical considerations that have quickly climbed leadership agendas. Leaders understand AI’s vast efficiency potential but also recognize the risks (e.g., biased algorithms, insensitive responses, and data privacy issues) that can damage customer trust overnight.

As a result, many organizations now establish AI ethics committees and demand regular transparency reports, detailed audits, and clear explanations of AI interactions. Executives realize efficiency is crucial, but trust is non-negotiable.

Leaders also recognize the delicate balance between innovation and organizational culture. The integration of AI must carefully consider its impact on team morale and job security. The best leaders communicate transparently about AI’s purpose, involve employees early in the process, and demonstrate how technology empowers workers.

Ultimately, modern leaders envision customer support as a powerful blend of human expertise and technological innovation, creating extraordinary customer experiences and driving tangible business outcomes. Today's support leaders must clearly articulate how operational strategies contribute directly to financial success, customer loyalty, and ethical standards.

The future belongs to organizations that skillfully combine technological advancement, human empathy, proactive risk management, and ethical responsibility. These are the companies that won't just survive—they'll thrive, building enduring competitive advantages rooted firmly in exceptional customer experiences.

Final Thoughts

AI's transformative power isn't about perfect execution; it's about how we respond when things inevitably go awry. Customer loyalty isn't built by avoiding mistakes altogether, but through transparent communication, rapid corrections, and genuine empathy when problems occur.

A personal experience with Waymo crystallized this for me: technology doesn’t have to be flawless; it simply needs thoughtful design, clear oversight, and human readiness to step in precisely when needed.

Interestingly, as AI becomes more widespread, human involvement becomes increasingly critical, not obsolete. The organizations of tomorrow won’t win by deploying flawless AI; they'll triumph by expertly combining technological innovation with authentic human intervention.

Great tech makes human involvement feel optional–until an exceptional, human-powered experience reminds us why it mattered in the first place.

What do you think? What else are you seeing in the evolving ops landscape? Comment and let me know.

Next up

In the next edition, I’ll dive into the tactical side of this conversation: practical, actionable strategies you can implement now to meet–and exceed–these rapidly evolving customer, agent, and leadership expectations. From redefining how you measure success in an AI+human world, to building the ideal Ops Tech team, we’ll outline specific steps that forward-thinking leaders are already taking to turn these shifts into strategic advantages. Stay tuned!