The gap that protects your team is the one you have to protect on purpose

Last month something arrived that shouldn't have been a problem. A client moved a date, a supplier fell over, somebody resigned. Unwelcome, entirely ordinary, the kind of thing that turns up every few weeks.

And the whole month went sideways.

Take a customer operations team of twelve. Two years ago they absorbed a full system replacement without anybody losing a weekend. The manager remembers it as hard but fine. Since then: a new product line, two extra reporting lines to service, and one person who left whose work was quietly distributed rather than replaced. Headcount is identical. Nothing was ever formally added.

Then a supplier misses a delivery window and the month falls apart. Escalations pile up, two deadlines slip, and the manager spends three weeks "firefighting".

When she finally sits down and lists everything her team has open, rather than everything that's been formally assigned, she counts thirty-eight items. Eleven of them nobody has touched in a month. Four are things she'd have sworn were finished.

None of that arrived in a single bad decision. It arrived in about forty reasonable ones, each of which committed slightly more of the space than the last, spread over two years, every one of them defensible on the day.

Think about a motorway. At half capacity it flows. Someone brakes hard, a car merges badly, and the disruption absorbs into the space around it. At ninety-five per cent capacity the same bad merge produces a queue that takes an hour to clear, and the car that caused it is long gone by the time the jam is at its worst.

The road didn't get worse. It ran out of room to absorb the ordinary variation that was always happening.

Something similar tends to happen in most systems that move work through people, and the important part is that the curve is steep rather than gradual. Waiting time doesn't creep up evenly as you approach full. It climbs sharply towards the end. Which is why a team can look fine at eighty per cent loaded and come apart at ninety-five, and why the difference feels so out of proportion to the extra work.

Your team is the road, and most hours committed in advance are a lane closed to whatever arrives on Tuesday.

There's a second thing happening underneath the first, and it's usually invisible until somebody leaves.

Rob Cross, Reb Rebele and Adam Grant looked at how collaborative work is distributed inside organisations and found that a small minority of people account for a disproportionate share of the valuable collaborative work. Everybody routes through them. They're helpful, they're good at it, and their name comes up in every conversation about who could take this.

Little of that shows up in a workload report, because most of it isn't assigned. It's the message asking a quick question, the sense-check before something goes out, the person three teams over who knows how the system works. This is a practitioner article drawing on the authors' consulting data rather than a controlled study, and the specific percentages that circulate with it are worth treating carefully. The pattern is the useful part, and it's one you can check in your own team in about ten minutes.

When our operations manager asked three people who they went to when they were stuck, two of them named the same person. That person's formal workload looked lighter than everyone else's, which is why she kept getting asked.

It doesn't create capacity. If the work exceeds what twelve people can do, slack is not the answer and protecting it will just mean the same overload arrives with a block of empty diary in the middle of it. That's a resourcing conversation, and the honest version of this advice is that the list of thirty-eight open items is evidence for it rather than a substitute.

And a gap you cancel is worse than one you never claimed. The first time you fill the protected block yourself, you've taught everyone what it's for. Better to hold a smaller gap you'll defend than a generous one you won't.