NPS Burnout Pt. 4 — “We're on it, first findings tomorrow.”

By then, we had already spent weeks working through the data. The machinery manufacturer’s customer journey was long and complex, running from the initial inquiry through the quotation process, technical clarification loops, and order processing to after-sales and support. Dozens of touchpoints, several departments, and a fragmented picture. We were looking for the moments that truly made a difference, the high-impact points where experiences took shape and influenced customer behavior.

Then it appeared. And we almost overlooked it.

The data had been right in front of us all along. That was the first thing that stood out. The second was that cause and effect could be connected. But the data was scattered: the ticketing system here, the CRM there, customer feedback in a third tool, operational metrics in a fourth. With hybrid journeys and companies organized in silos, that is unfortunately the reality. Each department collects, manages, and uses its data on its own schedule and according to its own logic. Often with a certain claim to ownership and interpretive authority (which is why so many organizations struggle with CDPs, data lakes, and building the MarTech stack, but that is another story).

The CMO knew that too. She had just put it differently: “I’ve got enough numbers.” In other words:
We don’t have an availability problem, we have an analysis problem.

Only when we linked sentiment data with operational data—customer feedback, tone and phrasing in inquiries on one side, and first resolution time, escalation history and processing cycles on the other—did something readable emerge. A pattern. A correlation. A finding.

Leading and lagging indicators, CPIs: that is what the previous episodes were about. On their own, they would not have been enough. They show what customers value. Only by connecting them to the operational context can you explain why something happens and where the lever lies.

The finding was unspectacular: technical customer inquiries took too long on average. That was known. What the data now showed was something else: the absolute handling time was not the problem. It was the silence in between. Customers who received no interim update followed up. Then again. Then someone escalated. The cost: frayed nerves on one side, euros on the other. What was missing was the feeling that someone was taking care of it. So what to do: hire more account managers or change the process? What can, and what must, the machinery manufacturer invest to heal this customer pain point?

Thanks to our contextualized analysis, we were able to weigh the value contributions of the measures under consideration against one another with considerable precision. What mattered, then, was the experience behind the measurably slow response time. Perceived attentiveness (“Is anyone taking care of this?!”) was the actual driver of loyalty. The finding seemed so obvious to us that we also conducted interviews with selected customers to validate it.

In the end, the measure was small. Almost embarrassingly small, considering the turmoil we had gone through beforehand.

Short, empathetic interim updates. “We have your request. We’re on it, first findings tomorrow.” No promise that couldn’t be kept. No elaborate process overhaul. A response that signaled: your request is a priority.

The impact was measurable on three levels at once. The number of follow-up inquiries dropped. The escalation rate went down. Cost-to-Serve fell, because there were fewer loops, less escalation effort, and less rework. And customer satisfaction rose, because customers felt their concern was taken seriously. Even when things take longer.

A single process moment. Three impact levels. That was the type of intervention the CMO had asked for.

What has been on my mind ever since is a simpler question: Is it reproducible?

Or was it a stroke of luck: the right dataset, the right company, the right moment?

This is what the final episode of this season is about.