The Small Pond

Open any fitness app with a global leaderboard on it.

Somewhere near the top is a person who has logged more hours this month than you have been awake. You look at that number, and what you feel is not motivation. It is a small, quiet click of disengagement. That person is not in your world. Whatever they are doing does not apply to you.

Then you close the app, and you do not train, and you are slightly worse off than before you opened it.

The leaderboard did that. And there is a seventy-year-old theory that explains precisely why, which almost nobody who ships leaderboards appears to have read.

The idea, named

Social comparison theory. Leon Festinger, 1954.

The core of it: people have a drive to evaluate their own abilities, and where objective standards are missing, they do it by comparing themselves with other people.

That much gets quoted a lot. The part that gets skipped is the qualifier that makes the theory useful. The comparison only does its work when the other person is similar to you. Festinger is explicit. Comparison with someone far outside your range on the dimension in question tends to be abandoned rather than acted on.

Not "is less motivating." Abandoned. Your mind declines the comparison and files the person as a different category of thing.

Festinger, L. (1954). A Theory of Social Comparison Processes. Human Relations, 7(2), 117 to 140. doi.org/10.1177/001872675400700202

So the entire industry built it backwards

Look at what a global leaderboard actually is under that theory. It is a machine that guarantees the person above you is dissimilar, because in a population of a hundred thousand the people near the top are outliers by definition. It maximises the number of comparisons that get abandoned.

And the worst part is the abandonment feels fine. It does not read as giving up. It reads as perspective. Well, they're a different kind of person. That is not a defeat you would ever notice, which is why the mechanism can sit there quietly making the product worse and still test well on engagement.

Meanwhile the comparison that would have moved you, the person who is roughly where you are and got started forty minutes earlier this morning, is invisible on page 400.

The lineage, credited

Matthew Larsen's Work Form System is the direct parent of what I built: the 60-minute unit, the written record of every hour, the wall of documents, the small group that knows whether you showed up. Larsen is openly assembled from Hill, Maltz, Cardone, Hormozi, Jobs via O'Leary and Clear.

He has a leaderboard, and he has an accountability group, and crucially they are the same small set of people. He did not build a public ranking. His board is his group, and the group is people who know each other.

That is the correct shape and I do not think it was arrived at by reading Festinger. It was arrived at by it being the only version that works when you try it.

What we add, as always, is that ours is populated by people you have actually sat in a room with rather than people you signed up alongside.

What we built

The leaderboard is you and your buddies. That is the entire population. There is no global board, no percentile, no rank out of anything.

Two columns, and the second is the one that changes behaviour.

Total hours is the obvious one.

Median time-to-first-hour is the one that does the work. How long after your day starts does your first real hour land. It is a much more honest number than total hours, because total hours rewards long days and time-to-first-hour rewards actually beginning. Nobody's is as good as they assume.

And it is a number where the people around you are genuinely comparable. Your buddy started at 6:12 today. You started at 6:51. That gap is small, real, and immediately actionable, which is exactly the condition Festinger says produces effort rather than a shrug.

This is Full Access, not the free plan. The free session gives you the room, the person and the timer.

The uncomfortable part

There is a version of this that is just competition with your friends, and I want to be honest that it can go wrong.

The safeguard is that the board only counts two things, and neither of them is output. It does not know whether you shipped. It does not rank your work. It counts hours you showed up for and how quickly you got going, both of which are entirely within your control on any given morning, and neither of which is a proxy for whether you are any good.

That is deliberate. A small pond only stays useful while nobody in it is being ranked as a person.

That's why I built this. I needed it. My time-to-first-hour was appalling and I did not know that until there were five other numbers next to it.