When an hour feels like ten minutes

You looked up and the afternoon was gone. Separately, and on a different day, you were certain the journey took twenty-five minutes, as you have been every time, and it has never once taken twenty-five minutes.

Two different things

The first is duration while it is happening. An hour passes as ten minutes when you are inside something; twenty minutes in a waiting room passes as an hour. Almost everyone has both experiences. What varies is how far the internal sense drifts and how often that matters.

The second is estimation in advance, which happens with a calendar rather than a stopwatch. You can be excellent at noticing that two hours have gone and still be reliably wrong about how long the essay will take. Most advice bundles the two together and then offers something that only addresses one of them.

A note on the phrase

You will see this described online as time blindness. People recognise the term, so it is worth naming once. It is informal community vocabulary — a description many people find accurate, not a scientific construct. There is no measure of it, it is not a mechanism, and it is not an indicator of anything about you. What follows is about the experience, not the phrase.

The prediction half has real evidence behind it

People underestimate how long their own tasks will take, and they do it even when they know they have underestimated the same kind of task before (Buehler, Griffin & Ross, Journal of Personality and Social Psychology, 1994). Knowing about the effect does not appear to remove it.

That relocates the problem. If your estimates run short, you are doing what people do. What varies is the size of the gap and how much the setting punishes it — and the second of those is not a property of you at all.

Where it pays and where it costs

The optimistic estimate is not purely a liability. Projects that would never be started if their true cost were visible in advance do get started, and a good proportion of them get finished. The person who thinks the reading will take an evening opens the book; the accurate estimator, looking at four evenings, does something else instead. A weak internal clock also means you are not stopping good work at arbitrary boundaries, and you finish things in one go that other people spread thin.

The cost is concrete and it lands on other people as well as on you. Lateness gets read as a statement about how much you value someone's time, which is usually the opposite of what is happening. Deadlines that looked comfortable in week two arrive as three consecutive nights in week nine. And the repeated experience of being wrong about your own predictions is corrosive in a way that a single missed train is not — it slowly makes your own plans feel untrustworthy, which is a heavier cost than any of the individual overruns.

Putting time outside your head

Stop asking your internal sense to do a job it does badly, and give the job to something visible.

An analogue clock in the room where you work does something a phone screen does not: it shows the remaining space as an amount rather than a number, so an hour looks like a size. Set the alarm for the thing after the current thing. And for estimation, keep one line per task — predicted, then actual. Two weeks of that gives you your own multiplier, the amount by which your predictions tend to run short, and applying it is far more effective than trying to predict better.

Work backwards from the arrival time rather than forwards from the departure time, including the parts that are not the journey: finding keys, the walk to the stop, the queue.

What the setting can do

Deadlines are almost always issued as dates and almost never as durations. A tutor or a manager who says this is about eight hours of work has handed over information they already have and everyone else is guessing at. That is a reasonable thing to request for a course or a team, and it costs one sentence.

Two more, both cheap. Ask that meeting invitations include travel time in the booking rather than assuming movement between rooms is instantaneous. And ask that estimates for group work are given the next day in writing rather than out loud in the room, so nobody is producing a number on the spot in front of an audience.

Small things to try this week

1. Put a real clock with hands where you work.
2. For five tasks, write your prediction, then the actual. Do not try to improve the prediction yet.
3. Set the alarm for leaving rather than for arriving, and set it earlier than you think you need it.

If your estimates do not improve after a few weeks, that is useful information rather than a failure. Most people's do not. The multiplier works anyway.