Why September Feels Like a Good Time to Start Again

The last long weekend of the summer, and then it is the first of September. Something about that date makes people want to begin again. A new notebook, an earlier alarm, the thing you have been meaning to do since June. It happens every year and it is easy to be cynical about it. The odd part is that the urge is real, fairly well documented, and more useful than it looks.

In 2014, Hengchen Dai, Katherine Milkman and Jason Riis published a paper in Management Science called The Fresh Start Effect. Across three field studies they tracked Google searches for the word diet, gym visits, and the commitments people made to pursue a goal. All three rose after what the researchers call temporal landmarks: the start of a new week, a new month, a new year, a new term, and days like a birthday or a holiday. Their explanation is a tidy one. A landmark closes off a mental accounting period. The version of you who did not go for the run belongs to August, and August is finished.

That is worth knowing on a week when the calendar is doing the work for you. It is not that you become a different person on Tuesday. It is that the date gives you permission to stop carrying the last few months around, and permission turns out to matter more than motivation. Most people do not fail to start because they lack a plan. They fail to start because starting means admitting the plan is already a few months old.

There is a limit to this, and it is a big one. What those studies measure is the beginning: the search, the sign up, the first visit. They say very little about the third week of September, when the alarm is still early and nothing about it feels new any more. A landmark is good at getting you moving and has almost nothing to offer once you are moving. If you have ever had a very promising first of January, you already know the shape of it.

The part that carries is duller than the fresh start, and better evidenced. Peter Gollwitzer and Paschal Sheeran's 2006 meta-analysis, published in Advances in Experimental Social Psychology, pooled 94 independent tests of what they call implementation intentions: not the goal itself, but a sentence that fixes when, where and how. If it is nine o'clock and I am at my desk, then I start with the hard thing. The pooled effect on actually reaching the goal was medium to large, d = 0.65. A goal on its own is a wish. The if-then is closer to an appointment.

So take the date, then make the plan smaller than feels satisfying. Attach it to something already fixed in your day rather than to an hour you are hoping to protect: after the kettle, after the school run, when the ten o'clock call ends. Give it a start signal you can repeat. Sound is good at this because it is the same every time and it costs nothing to begin. The same rainfall over a forest, the same stream, the same twenty minutes. After a fortnight the sound stops being pleasant background and starts being the thing that tells you which mode you are in.

And if the first of September comes and goes and nothing changes, the research is kind about that too. There is another landmark on Monday, and another one after that. The date was never the point. It is only a reasonable excuse to put down whatever August looked like and try the small version once, today.