The decision

Here's the rule. I keep the shelf stocked, I know six plain plates, and I decide dinner before I'm hungry.

That's it. It isn't a rule about takeaways, which you can still have, on purpose, on the night you chose. It isn't a rule about health, and it doesn't ask you to eat anything you don't like. It's a rule about the default. When nothing was decided, the default used to be the app. Now the default is one of the six, and the app is what you choose on Friday.

Three when-then plans, which are a form the research on planning supports, a specific situation paired with a specific act, so the act doesn't need deciding in the moment.

When I get in and the seven o'clock question comes, then I look at the list on the fridge and put the pan on before I sit down.

When I open the fridge and think there's nothing in, then I look at the shelf, not the fridge, and make the one of the six that the shelf can make.

When I find the app open in my hand on a night that wasn't the night, then I close it and set a timer for fifteen minutes and cook, and if I still want the takeaway at the end of the fifteen minutes I can have it tomorrow, on purpose.

What to expect, and this is my expectation and not a finding. The first week is awkward, the plates are plain and one of them goes wrong. The second week is easier because each of the six has been done once. Somewhere in the second month the pan going on at seven stops being a decision and becomes what happens, which is what Phillippa Lally and colleagues found in a study of everyday habits, where a daily behaviour in a fixed setting became automatic over weeks and a missed day didn't undo it. The takeaway on Friday will taste better than any takeaway has in a year, because it'll be the only one. And the box on the side will be gone, replaced by a pan in the rack, which you'll notice one morning and not think much about, which is the point.