Why "by 10pm" beats "today"

“I’ll go to the gym today” and “I’ll be at the gym by 6pm” feel like the same commitment. They are not, and the difference is measurable.

What the research on plans actually found

The strongest evidence here is not about motivation. It is about specificity.

An implementation intention is a plan in the form if situation Y, then I will do X. Not “exercise more” but “when I get in on Tuesday, I change and run for twenty minutes”.

A review of 94 separate experiments found that writing a plan in that form made a real difference to whether people actually did the thing [1].

Line up one person who wrote the plan against one who did not, and the planner comes out ahead about two times in three. For something that costs a single sentence, two in three is a very good trade.

In another much-quoted experiment, from 2002 [2], students were split into three groups for a fortnight: one just tracked their exercise, one was given material to boost their motivation, and one got that same material plus a sheet asking them to write down the day, time and place they would train. In the third group, 91% exercised that week, roughly double the motivation-only group.

The sample was smaller and self-reported, but the direction is what matters: deciding when, exactly, is most of the work.

Why a vague deadline is not a deadline

Here is the mechanism, and it is almost too simple.

“Today” has no moment at which it breaks. At 2pm you have not failed. At 7pm you have not failed. At 10pm you technically still have two hours. The deadline never arrives, so it never forces the decision. It just quietly expires while you are asleep, and in the morning it is a new “today”.

“By 6pm” has a moment. At 6:01 the answer is no. That is unpleasant, and the unpleasantness is the entire function: a deadline works by converting an open question into a closed one at a known time.

This is also why “I’ll do it later” is such a durable trick. It is not a lie. Later genuinely exists, right up until it doesn’t.

The uncomfortable finding: we set our own deadlines badly

One study asked three questions about deadlines people set for themselves [3]: will they voluntarily accept a costly one, does it improve what they produce, and do they set it well?

The answers were yes, yes, and no.

People do impose deadlines on themselves, knowingly accepting a penalty for missing them, which tells you they understand their own procrastination perfectly well. Those deadlines did improve performance. But the spacing people chose was not the spacing that worked best. Evenly spaced imposed deadlines beat the ones students set for themselves.

That third finding is the one worth sitting with. It means self-knowledge is not the bottleneck. You already know you will put it off. What you are bad at is designing the constraint, because at the moment of designing it you are relaxed, well-rested and optimistic, and the person who has to meet it is none of those things.

What to do with this

Pick a time, not a day. Convert every intention into a clock. “Walk more” becomes “8,000 steps by 9pm”. The number matters less than the fact that there is a moment when it resolves.

Decide the time once, while you are motivated. That is when your judgement about what you want is at its best. It is also when your judgement about what is achievable is at its worst, which is why the target should come from what you actually did last month, not from what you would like to be true. Your recent average is the honest starting point.

Then take away your ability to renegotiate. This is the part the research argues for and the part almost nobody does. A deadline you can move at 9pm is not a deadline; it is a preference. The reason commitment devices work at all is that they remove the option you will want later.

Write down what a miss costs, in advance. Not as punishment, as procedure. The problem with a missed day is never the day. It is that the miss opens a negotiation, and the negotiated kind repeats.

This is the shape Dare is built around: you set a target from your real Apple Health baseline, you name a deadline, and after that the app will not let you move it. Miss it and the apps you chose lock until you make the goal up. The research supports that shape: a specific time, imposed rather than renegotiable. It does not test our app, and we are not going to pretend otherwise.

The short version: a goal without a time is a wish with better grammar. Give it a moment when the answer becomes no, and then make sure that moment is not yours to move.

Sources

1. Gollwitzer, P. M. & Sheeran, P. (2006). Implementation intentions and goal achievement: a meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. Pooled 94 independent tests and measured the effect at d = 0.65.

   That figure is not a percentage, which is the easy mistake to make with it. It means the two groups’ averages sat 0.65 standard deviations apart. The convention is that 0.2 is small, 0.5 medium and 0.8 large, so 0.65 is medium-to-large. Converted into something a person can picture, it puts the odds of the planner beating the non-planner at roughly 68%, which is where the “two times in three” above comes from.

2. Milne, S., Orbell, S. & Sheeran, P. (2002). Combining motivational and volitional interventions to promote exercise participation. British Journal of Health Psychology, 7, 163–184. 248 undergraduates, three conditions, two weeks, self-reported.

3. Ariely, D. & Wertenbroch, K. (2002). Procrastination, deadlines, and performance: self-control by precommitment. Psychological Science, 13(3), 219–224.