The Problem Your Best People Keep Solving

By Joe Curcillo

Before you congratulate someone on finishing a difficult job, ask what they had to do to finish it. Sometimes you’ll hear about a genuinely hard problem that took experience and good judgment. Other times you’ll hear that they had to chase down the same missing information again, or that they caught a mistake because they’ve learned exactly where it usually shows up. Either way, the work came in on time, which is why the second answer is so easy to miss.

Some problems never make the priority list because somebody already handles them. The work goes to that person, they straighten it out, and everyone moves on. After a while nobody thinks of it as a repair anymore. It’s just how things get done, and the organization has found someone willing and able to absorb a problem it never actually solved.

I don’t quarrel with the person who gets the job done. I’d want that person on my team. What concerns me is what leadership concludes from the result. A review that says the work was completed and the customer was happy leaves out the most useful fact in the whole story, which is that someone had to step in. If they have to step in every time, I’d pay more attention to that than to the result.

Challenger shows how dangerous it can be to take reassurance from previous results. The Rogers Commission found that O-ring damage had come to be accepted as a flight risk. Previous flights without catastrophe helped justify flying again while the underlying joint problem went unresolved. That acceptance was not unanimous. Thiokol engineers continued to oppose the launch after management reversed its initial recommendation against it.

Sociologist Diane Vaughan later described the process as normalization of deviance: a condition that once demanded explanation turns into something people know how to live with.

In the situation I’m describing, somebody corrects the immediate problem, usually the same somebody, before the customer ever sees it. The stakes are different, but I’d argue the misreading is similar. Leadership sees a good outcome and decides the system works, when what the outcome shows is one person catching what the system keeps missing.

Look at what makes that employee so valuable. They know a figure means something different once it reaches another department. They know when a promised date depends on work nobody has put on the schedule. They see more than the task in front of them, and they use what they know about the surrounding work to get it finished.

The Integration Gap™ is the distance between what your people already know and what actually reaches the decision. The person doing the repairs often knows exactly why the problem keeps coming back, but when leadership decides what needs fixing, the only fact on the table is that the job got done. So the knowledge sits with one employee and never changes how the organization operates.

Let that arrangement run until the day that person retires or takes a better offer, and the repair may walk out the door with them. If nobody realized it was happening, nobody thought to write it down. You may find out through a deadline that slips or a customer who catches the error, and leadership will be surprised by a problem that sat in plain view for years.

The natural response is to ask the capable person to teach everyone else how to fix it, and sometimes that makes sense. First, though, find out whether you’re teaching a necessary skill or teaching people to live with a preventable condition. You can become very efficient at repeating work you should have stopped needing to do.

It’s also worth asking what you’re spending that person on. Someone who can anticipate consequences across several departments has better things to do than rebuild the same missing context on every assignment, and the finished job won’t show you what they set aside to deliver it. I’d count that as part of the result.

Hold the Horizon™ includes a tool called Signal Sensemaking: Listening Beyond the Loudest Voice, which is about catching early signs that something is shifting beneath the surface. One of its practices is tracking small anomalies like repeated confusion and strange delays. I’d put recurring rescues on that list. A job that needed saving and still came in on time is telling you something.

Start with one recent piece of work and ask the person who fixed it to walk you through it. Question them the way you would a witness whose account matters. Stay on the details until you understand why they had to step in, because “I sorted it out” is a conclusion, and a conclusion tells you almost nothing about the condition you’ll be asking them to sort out again next week. Then write that condition down somewhere it will outlast the assignment. If you don’t, the next review will show another completed task and give nobody a reason to act.

After that, look across the work. If they keep having to get the same missing information from another team, find out why it isn’t there when they need it. If they keep negotiating between commitments that can’t both be met, find out who made those commitments and who has the authority to change them. What they tell you points to where to look, but it doesn’t prove the cause. Adding another approval or form before you understand the cause just gives them one more thing to work around.

Some interventions are exactly what should happen. A qualified review that catches an error is a control doing its job, and changing circumstances will always call for judgment. What you’re looking for is the same avoidable condition forcing the same rescue over and over. Keep the control and go after whatever keeps feeding it errors. And don’t strip out the protection because the results have been good, since the protection is a big part of why they’ve been good.

The person who keeps the work moving has been handing you more than finished assignments. Without meaning to, they’ve been showing you what it actually takes to get work done in your organization. Ask them about it while the work is still fresh, rather than waiting until a missed deadline makes everyone curious. Hold the Horizon has a question worth starting with, and the answer is usually sitting at somebody’s desk: “Is there a pattern we’ve been treating like noise?”