The Last Ten Per Cent Takes as Long as the Rest

Everyone underestimates the tail of a job, every time, no matter how many times they have been through it.

The shape of a making project is reliably the same. The early stages move quickly and feel excellent. The material transforms visibly, each session produces obvious progress, and you begin to believe you will finish sooner than expected. Then the transformation stops being visible. The thing is essentially the right shape and everything remaining is refinement: surfaces, edges, fit, the parts nobody would notice individually but everybody notices collectively. This phase absorbs an amount of time that is genuinely hard to believe until you have lived through it several times.

The reason is that the early work is subtractive at a coarse scale, where a single decisive action changes a lot, and the late work operates at a scale where each action changes almost nothing. You are making tiny corrections and then checking, and the checking takes as long as the correcting. There is also a shift in what you are attending to. Early on you are looking at the whole. Late on you are looking at a small area, then another small area, and the whole only reassembles itself once you step back.

This is where most unfinished projects die, and they die in a specific way: not abandoned in frustration but quietly deferred, because the remaining work no longer produces the feeling that the earlier work produced. Experienced makers manage it mostly by expecting it. If you know from the outset that the tail is half the job, the tail stops feeling like evidence that something has gone wrong, and you can settle into it with something closer to patience than resentment. It also helps to accept that this part of the work is quieter by nature. The excitement has moved out and something steadier has moved in, and steadiness is a perfectly respectable way to spend an afternoon.