Have you seen organizations go too far with automation? What happened?

Yes, more often than you’d think and across all industries. And it usually starts with good intentions.

Many companies turn to automation as a way to cut costs. On paper, that makes sense. But the problem begins when automation becomes primarily a cost-reduction tool instead of a way to improve the customer experience.

A common example I see is in customer service. Some organizations replace large parts of their service teams with automated systems that aren’t mature enough to handle real-world complexity. Customers can’t resolve their issues. They get stuck in loops. They can’t reach a human. Frustration builds quickly.

And it doesn’t stop there. Tickets start piling up because the system can’t properly solve them. Meanwhile, there are fewer humans available to step in. It creates this domino effect. Angry customers, unresolved cases, negative social media sentiment, and eventually brand erosion.

The reality is customers have endless options. If the experience is painful, they will leave. So while the company might save money in the short term, they often lose revenue, and trust, in the long run. 

Are there parts of the customer journey that should never be fully automated?

There are definitely moments where a human presence is critical.

Customers actually like automation, especially when it makes their lives easier. If it speeds things up, reduces effort, or improves quality, they appreciate it. But what they don’t like is feeling blocked or unheard.

We talk a lot about personalization. But true personalization isn’t just remembering someone’s name or purchase history. It’s about feeling understood. And that’s very hard to fully automate.

What works best is automating the back-end processes. The things customers don’t see. That’s where efficiency can really shine. But when it comes to support, advice, reassurance, or emotionally charged situations, people still want access to a real human.

The human doesn’t have to be involved at every step. But they need to be reachable. And they need to be empowered to step in when it matters. 

With AI advancing so quickly, how should companies decide when human judgment must stay involved?

AI is incredibly powerful. It can analyze massive amounts of data, generate insights, make predictions, and even recommend actions. The productivity gains are real.

But the risk is also real. Especially when decisions impact people or financial outcomes.

That’s why the idea of “human in the loop” is so important. AI should support decision-making. It should surface insights and recommend options. But humans should validate and take responsibility for the final call – especially when stakes are high.

If a decision affects someone’s finances or access to services, a human should be accountable at the end of that process.

AI is a fantastic assistant. It should not become an unchecked authority. 

We’ve all experienced chatbots that feel scripted and robotic. How can organizations make automated journeys feel more empathetic?

Empathy in automation isn’t really about sounding polite. It’s about behavior.

First, systems need context. If a customer has interacted with you five times before, the system should know that. It should remember preferences, past issues, and prior resolutions. That creates the feeling of being recognized rather than starting from zero each time.

Second, give customers options and explain them. When automation clearly outlines why one option might be better than another, it builds trust.

Third, be honest about limitations. I personally appreciate when a system says, “I don’t have enough information to answer that.” It’s much better than getting a confident but incorrect response. That transparency really builds credibility.

And maybe most importantly, design for recovery. Automation will fail sometimes. That’s inevitable. The key is making it easy to escalate to a human and ensuring that the full context transfers with the customer.

When empathy is built into how the system behaves (not just how it talks), the experience feels much more human. 

If leaders could remember just one principle when balancing efficiency and empathy, what would it be?

That companies should use automation to improve customer outcomes, not just to reduce costs.

When automation helps customers resolve issues faster, access better products, or reduce effort, they love it. But the moment it becomes a barrier, when it blocks access to help or adds friction, they will start to resent it.

Leading organizations don’t automate everything just because they can. They think carefully about where automation truly enhances the experience and where a human touch makes the difference.

And trust still matters deeply. Customers feel reassured when they know there’s a human validating the decisions behind the scenes.

Any final thoughts for leaders navigating this shift?

We need to use automation responsibly.

AI, data, and automation can absolutely elevate service quality and efficiency. But we can’t lose sight of judgment, accountability, and trust.

I often hear leaders ask, “How do we ensure AI quality? How do we make it secure? How do we know we can rely on it?” These are the right questions. Responsible use is not optional. I’s critical.

The human in the loop isn’t going away. Customers want speed and convenience, yes. But they also want reassurance. They want to know someone is ultimately accountable.

The companies that succeed commercially are the ones that know how to combine automation with humans for optimal results.