I've developed a habit of consulting Siri about the weather before I head out for my early morning pre-work walk around the neighborhood. A few days ago she told me it was 57 degrees and cloudy, but to expect rain later in the morning. That sounded OK until I took two steps outside the front door and promptly got soaked. As my wife later commented I could have looked out the window instead of relying on technology.

Was she right? Had I become too reliant on the machine and not enough on the human (me) to decide what my experience was going to be?

So how does that apply to the broader context of the customer experience? How do you decide when and where to use each type of communication? When is it OK to trust the machine, and when do you need the human in the loop?

I’ve long held the notion that the customer experience is founded on a trinity of the customer themselves, the content we provide for them, and the context in which they consume, or interact with it. And it is in this last area, where we should think about the machine/human balance.

Related Article: The IoT May Change the Customer Experience, But Not in the Ways We'd Expect

The Automated Runaround

My wife recently called our life insurance company with a question and got the automated menu runaround. It was almost impossible to speak to a human. In fact, she gave up and, thanks to her excellent googling skills, found a way to get our issue resolved by setting up an online account and working  through that (which could perhaps be a whole other article).

As we were talking over her telephone interaction she said at one point she had become so frustrated she almost picked the “Death Claims” choice on one of the phone menus just to reach a real person. At least I hope that particular function (which could have been much more sympathetically labeled) did connect you with a trained, empathetic individual human and not another automated data capture process.

So how do we achieve that balance, when do we use the machine, and when do we need the human? I think the following sets up some basic guidelines:

Learning Opportunities


  • Basic data acquisition (account set-ups and simple updates).
  • Simple transactions (such as activating a credit card).
  • Online purchases (that new book I really want to read).
  • Repeat transactions (my Friday night pizza delivery order).
  • Keeping track of customer activity (What episode am I on in my latest streaming service binge watch).
  • Answering basic questions (Who, what, where, when?).

Human Interaction:

  • Problem resolution.
  • Answering complex questions (why?).
  • When empathy is required.
  • When changes in context impact the experience.
  • Complex account changes.

Related Article: A Good Chatbot Is Hard to Find

Human vs. Machine Isn't a Binary Decision

Your customers interact with you for two basic reasons: to find a solution to a problem they have, and to conduct transactions. Humans are great for helping with the first of those, while machines and technology allow us to effectively do the second at speed and scale.

But it isn’t a binary relationship. The machine/human divide is a sliding scale across the customer experience continuum and the right solutions need to be bought into play at the right time of the customer journey.

fa-solid fa-hand-paper Learn how you can join our contributor community.