The $20M Protection Story:  Listen. Capture. Act.

I started my career in sales, where the customer was everything. When I moved into corporate roles, I discovered that the same skills that made me effective externally: listening, building trust, removing obstacles, were just as powerful internally. The difference was that internally, there was no commission check. Just the satisfaction of watching problems get solved before they became catastrophes.

For over a decade, across three different roles, I planned, built, and ran Data Center Customer Advisory Boards (CABs.) These were not focus groups or feedback surveys. They were structured relationships that gave our most important customers direct access to our engineers and executives, and gave us something far more valuable than market research: real problems, in real time, from the people whose businesses depended on our technology.

My motto was simple: Listen. Capture. Act. You cannot act correctly if you have not listened. And listening means nothing if you do not act. The capture step, tracking every issue, every concern, every request, was what made action accountable.

When we began moving into a new market with a new product line, our executives asked me to build a new CAB from scratch. There was one significant obstacle: we did not know the right people. Our existing relationships were with data and networking decision makers, but these new products required buy-in from application decision makers, a completely different audience, often skeptical of our company and everything we represented in their organizations.

So I created what I called the buddy system. I asked our trusted data center contacts to nominate an application-focused colleague, someone who had purchased or was seriously considering our new solutions. It was a bridge built on existing trust, extending into unknown territory.

In one of our very first sessions, those new and deeply skeptical application decision makers surfaced something our support organization had completely missed: a bug that was causing serious deployment challenges. Our engineers did not know the full scope of the problem. Our support center had been receiving individual cases, but without the context to connect them or the expertise to identify the root cause. The customers experiencing it had no direct line to the people who could actually fix it.

We fixed it.

When the patch was deployed across all affected customers, not just our CAB members,  something remarkable happened. Trouble tickets for these products dropped by 90%. I went back and traced the impact manually, following up with account managers across our enterprise and service provider base. What I found was sobering: customers who had never said a word to our executives, who had simply opened support cases that went nowhere, had been quietly preparing to leave. Large enterprise accounts. Accounts that had trusted us on the data and networking side for years and had extended that trust to our new products — trust we had nearly broken without even knowing it.

The impact I could conservatively document was $20 million in protected revenue. And that did not include our service provider accounts, who would never have deployed at scale with that bug in place, a much larger number that came later, quietly, because the problem had been solved.

The skeptical application decision makers who had walked into that first session ready to dismiss us became some of our strongest advocates. Not because we were perfect. Because when something went wrong, we fixed it and they had seen it happen in real time.

I ran those CABs for over a decade. When I negotiated moves within my company, our executives consistently made continuity of the CAB program a condition of the transition. Many of those CAB customers remain friends today.  Those relationships were established and strong.

What this means for AI today.

We were moving into one new market. With AI, every organization is moving into something new simultaneously:  new technology, new processes, new cultural territory, often all at once. The people closest to the problems are frequently not the people making the decisions. The issues surfacing in support queues are not always being connected to the engineers who can actually solve them. And the customers, employees, or citizens most affected, often have no direct line to the people who could fix things if they were aware of them.

Listen. Capture. Act.

In a moment defined by economic uncertainty, political division, and rapid technological change, the temptation is to move fast and figure it out later. But the organizations that will get AI right are the ones that build the structures to hear from the people closest to the work and then do something about what they hear.

We solved a $20 million problem not because we were smarter than our competitors. We solved it because we were in the room with the right people, asking the right questions, and had built the trust and the systems to act on what we learned.

That is not a technology story. That is a human story. And it is exactly the story AI adoption needs right now.  And it is exactly the kind of problem I focus on solving.