The Only Reliable Motivation I Have Found Is Another Person Waiting

Over about two years I tried five approaches to getting myself to do things I did not want to do. I want to describe them in order of how well they worked, which is roughly the reverse of how sophisticated they are.

Rewards did not work at all. The reward was always available anyway, and I am not able to genuinely withhold something from myself. Twice I did the reward without the task and felt nothing.

Habit stacking — attaching the new thing to an existing routine — worked for small things and not at all for large ones. It is excellent for a two-minute action and useless for a two-hour one, because a routine has room for the first and not the second. I still use it and it is not what this piece is about.

Environment design — removing friction, putting the thing where I would trip over it — worked well for about three weeks each time and then stopped, because I stopped seeing the arrangement. This is a real effect and its decay is not much discussed. I now change my arrangements deliberately every so often, purely to make them visible again.

Public commitment — announcing what I would do — worked once, memorably, and then not again. My theory is that the first time carried real social risk and by the fourth time everyone involved had learned that my announcements are aspirational, including me.

Someone waiting for the specific thing works every time and has never failed. Not accountability in the general sense of a partner who asks how it went — a person who cannot proceed until I have done my part. The difference is that the second creates a cost to someone else and the first creates a report.

I have thought about why this one is different and my best account is that it changes what kind of failure is available. Not doing something for myself is a private event with a private cost that I have a great deal of practice absorbing. Not doing something someone is waiting on is a thing that happens to them, and I have no practice at all at being comfortable with that, which is the whole mechanism.

The practical difficulty is that most of what I want to do has nobody waiting. So I have taken to manufacturing it, which feels artificial and works anyway. For writing, I send drafts to one person on a schedule, and they read them, and they notice when nothing arrives. For a piece of long-running work, I agreed to give a colleague a short update on the first Monday of each month — not because they need it, but because an empty update is something I would have to send them.

I would flag one thing honestly: this works because I care what a specific person thinks, and that means it uses up something. I would not attach every goal I have to a person, and I have been careful to keep it to two. Beyond that it stops being motivation and becomes a set of obligations, and obligations produce a different and worse kind of doing.