Most service journeys are designed for the day everything works. The form submits, the system responds, the customer gets what they came for. That is the version everyone signs off, because it is the version everyone pictures when they think about the service at all.
Then something breaks, and a different question arrives, the one nobody prepared for: what happens now?
I ran into this recently trying to book a simple appointment with a well-known retailer. The online form failed. So I went into the store, where the staff were helpful but their own system would not load either. Two channels, one customer, no appointment. Nobody in that store did anything wrong. That is exactly the point.
The failure was not in the design, and not in the technology. It was in the gap between them. Someone had designed the booking journey. Someone else ran the system it depended on. And somewhere between the two, nobody owned the question of what should happen when it broke. There was no fallback, no paper list, no offer to take my number and call me back. The failure state had simply never been designed, because failure states rarely are.
This matters more than it looks, because a failure state is not a minor edge case. It is a moment of maximum consequence. A customer who hits a smooth journey feels nothing in particular. A customer who hits a broken one feels everything, and decides something. The unhappy path is where loyalty is actually won or lost, and it is the part almost nobody owns.
The reason it goes unowned is structural. Design teams are measured on the journey working. Operations teams are measured on the system running. Neither is measured on what happens in the seconds after it stops. So the fallback becomes nobody’s job, and “nobody’s job” is where services quietly bleed.
And the bleed is quiet, which is the dangerous part. Broken journeys do not generate complaints. The customer does not write in to explain that your booking form returned an error so they went to a competitor instead. They just go. The failure produces silence, and silence does not appear on any dashboard. You watch the smooth journeys succeeding and assume the service is healthy, while the failures walk out of the door unrecorded.
So the real work is to design the failure state on purpose. Decide, in advance, what happens when the form breaks, when the system is down, when the information is missing, when the usual person is away. Give every failure an owner and a fallback, however low-tech: a phone number, a paper list, a promise to call back. The goal is not to prevent every failure, which is impossible. It is to make sure that when one happens, the service still has an answer, and someone whose job it is to give it.
If you want to find your own unowned failure states, the test is simple. Try to use your own service on a bad day. Book something on your own system, then try again with the connection throttled or a required field left blank. What breaks, and what happens next, will tell you more about your operation than most of the research you have paid for.
Mapping where a service breaks, and who owns it when it does, is a lot of what a Service Clarity Audit surfaces. → Book a Service Clarity Audit
