Ask why a particular manual step has not been fixed, and the answer is almost always some version of “it’s really not a big deal — it’s just one file”. Taken individually, that is usually true. Taken as a pattern, it is the reason a lot of expert time disappears without ever looking like a problem worth solving.

01. Why “just one” defeats scrutiny

A task that takes fifteen minutes does not clear the bar for “worth fixing” when it is considered on its own. Fixing it would take longer than doing it, this once. The trouble is that “this once” repeats — weekly, sometimes daily — and the fifteen minutes never gets multiplied out because nobody is tracking it as a recurring cost. It is re-evaluated fresh, and dismissed fresh, every single time.

02. The arithmetic nobody does

Fifteen minutes, once a week, is roughly thirteen hours a year — from one person, for one task. Most senior roles carry several such tasks simultaneously: one weekly report, one routine sign-off, one recurring data pull. Add them up honestly, for one person, and the total is rarely small. Add it up across a team, and it starts to resemble the billable-time leakage most firms have never actually calculated.

03. The cost that is not financial

There is a second cost that does not show up in any spreadsheet: attention. Small, low-value tasks scattered through a week fragment the time available for the work that actually needed a senior person’s judgement. The cognitive cost of context-switching — stopping deep work to handle a five-minute administrative task, then restarting — is well documented in research on task-switching and productivity, and it is disproportionate to the task’s apparent size.

This is why “just one file” is a more accurate description of the effort involved than of the actual cost. The effort is small. The interruption is not.

04. Why it survives every efficiency review

Efficiency reviews tend to look for big, visible inefficiencies — the process that is obviously broken, the project that is obviously late. A task that is individually fifteen minutes and individually reasonable rarely surfaces in that kind of review, because reviewed one at a time, none of them look like the problem. They only look like a problem in aggregate, and aggregating them requires actually asking where an expert’s billable hours go, task by task, rather than assuming the answer is obvious.

05. What actually resolves it

Not a policy asking people to batch their admin or be more efficient — that puts the burden back on the same person already absorbing the cost. What resolves it is removing the task from the workflow’s manual path entirely: structured intake that does not need re-keying, evidence attached automatically instead of hunted down, a draft assembled and ready for review instead of built from scratch each time. That is the specific job the four layers of a connected delivery system are built to do — not to make each small task faster, but to make it stop being a task a person has to remember to do at all.

One small recurring task, multiplied by 52 weeks, is usually worth a conversation — discuss one workflow →