Field Notes · Time

The Interruption Tax

6 min read · organizational behavior

Your calendar isn't the problem. What happens to your attention between the meetings on it is.

Picture a Tuesday. A CEO closes a laptop after forty minutes of deep work on a term sheet, walks into a status meeting, answers a Slack message about a vendor invoice, sits through twenty minutes of a meeting that didn't need her, then tries to go back to the term sheet. She reads the same paragraph three times. It isn't fatigue, and it isn't distraction in the way we usually mean it. It has a name, and it was measured.

The residue that doesn't clear

In 2009, organizational psychologist Sophie Leroy, then at the University of Minnesota, ran a series of experiments on what happens when people switch from an unfinished task to a new one. She called the effect attention residue: part of your cognitive attention stays stuck on the abandoned task, and performance on the new one suffers measurably, even when you fully intend to focus. The residue was worse, not better, when the first task was important. The more it mattered, the harder it was to put down.

This is the mechanism hiding inside a fragmented executive calendar. It isn't that meetings take time, which is obvious and already priced in. It's that every unfinished thing you were pulled away from keeps drawing on attention you've mentally allocated somewhere else, for the rest of the day. A CEO who takes twelve interruptions before lunch isn't tired from twelve interruptions. She's carrying twelve open loops into every conversation that follows.

The meeting that ran long doesn't just cost the thirty minutes it stole. It costs a piece of whatever came before it, too.

Why the calendar keeps filling anyway

C. Northcote Parkinson made the observation in 1955, as a joke that turned out to be true: work expands to fill the time available for its completion. A meeting scheduled for an hour will use the hour, agenda or no agenda, because nothing in the room is pushing back on the clock. Multiply that across a company where meetings are the default and silence is the exception, and you get what Harvard Business School's Leslie Perlow spent two decades documenting under the name collaborative overload: teams so continuously available to each other that no one individually has room to think.

Perlow's most cited intervention is almost embarrassingly simple. At the Boston Consulting Group, she and her research partners tested "Predictable Time Off," blocks of protected, uninterruptible individual time, chosen and defended in advance. Teams that adopted it didn't just report less burnout. They reported better collaboration, because protecting individual focus forced the team to get explicit about who actually needed to be looped in on what, instead of defaulting to including everyone, always.

The cost of staying the decision-maker of last resort

There's a second, quieter tax. Every decision that gets routed to a CEO because no one else is authorized to make it draws on the same limited daily reserve of self-regulation, a phenomenon Roy Baumeister's lab spent years studying under the name ego depletion. The most famous illustration, a 2011 study of Israeli parole judges whose favorable rulings reportedly dropped through the morning and recovered after a food break, has since faced real replication scrutiny, and it should be cited carefully rather than treated as settled law. But the underlying, better-replicated point survives the debate: decision quality is not constant across a day of decisions, and routing the small, low-stakes approvals to the top means the person making the biggest decisions is making them on a tank that's already been drawn down by the small ones.

Why "someone should handle that" doesn't work

There's a reason unstarted projects stay unstarted even at companies full of capable people. Bibb Latane and John Darley's classic research on the bystander effect found that responsibility diffuses in exact proportion to how many people could plausibly take it. Say "someone should own the CRM cleanup" to a room of eight people, and you have said it, technically, to no one. Each person privately assumes it's more likely someone else will move first. The project isn't blocked. It's just never anyone's specific job, which functions identically to being blocked.

What this means in practice

None of this is solved by working faster or caring more. It's a structural problem, and structural problems get fixed by changing who holds the structure. A part-time Chief of Staff's actual function, underneath the calendar management and the inbox triage, is to be the place attention residue and undecided ownership go instead of the CEO: closing loops before they're carried into the next room, and putting a name on the tasks that would otherwise diffuse into nobody's.