← Desk

Stopwatch

One task, timed once, on a ruled sheet
Pick one task you do over and over. Write its steps down the left side, time one pass, and date the sheet. Then change one step and lay a fresh sheet beside the first. This page times you and nobody else, and it never sets a rate.

Rule the sheet

A stopwatch and a properly ruled notebook is all the method ever needed. You could rule one on the back of an envelope.

The pair

Two passes of the same task, and the two sheets sit here side by side.
The value is in the pair, not in either page. There is no score on this sheet and no target time, because a rate set from a timing is the part of this story that should not be kept.

Sheets kept

What the record says. No one has tested this practice. What is known is narrower: people are bad at guessing how long their own work takes. In 1994 Roger Buehler, Dale Griffin and Michael Ross asked thirty-seven honors-thesis students at the University of Waterloo for their best estimate of when they would finish. The average estimate was 33.9 days; the average actual was 55.5, and only 29.7 percent finished by their own best guess. In a later study in the same paper, the optimism went away when people were told to connect their predictions to relevant past experiences. A dated sheet is a record of past experience. It will not make the work better on its own. It will make it clearer, and that is what the sheet is for. The rate set at Bethlehem in 1899 was the fastest burst of ten men, cut by forty percent for reasons nobody wrote down, and ten men were paid off for refusing it. There is no one best way in the record, only a best guess wearing a watch.
Why this works, in one picture →
Card 1 of the Desk Playbook: what the founding afternoon says against the report in his own papers, and the one finding about guesses the sheet can lean on.