Why habit streaks break, and what I changed about it
I kept running into the same complaint while researching habit trackers. It shows up in App Store reviews, in Hacker News threads, and in the comments under every "best habit app" roundup. People do not quit these apps because they stopped caring. They quit because they lost a streak on a day they actually did the thing.
That is a strange way for software to fail. The behaviour happened. The app simply did not hear about it, and then punished the person for its own blind spot.
The gap between doing something and recording it
Every habit tracker asks you to do two separate things. First the habit, then the paperwork. The second one is the part that quietly breaks.
One comment I found put it better than I could: if I am too lazy to do the habit, I am also too lazy to open an app and say so. The logging is not free. It is a small tax you pay on top of the thing you were already trying to make automatic, and it gets skipped on exactly the days when life is busiest.
So the streak counter stops measuring your behaviour and starts measuring your diligence about paperwork. Those are not the same thing, and only one of them is worth building a habit around.
What actually happens to people who pay for a habit
There is a well known economics paper that gets at this from a different angle. In 2006, Stefano DellaVigna and Ulrike Malmendier published Paying Not to Go to the Gym in the American Economic Review. They had three years of real billing and turnstile records for 7,752 gym members, so this is not a survey about what people think they do.
Members on flat monthly contracts above 70 dollars attended about 4.3 times a month. That works out to more than 17 dollars per visit, at gyms that also sold a ten visit pass for 10 dollars a visit. The average member gave up around 600 dollars over the life of the membership by choosing the wrong plan for the behaviour they actually had.
The detail I find most interesting is the lag. On average, members kept paying for 2.31 months after their final visit. They had already stopped, and the paperwork had not caught up.
That gap between what you did and what the record says is the same gap, just measured in money instead of streaks.
Punishment is a bad default
Plenty of habit apps respond to a missed day by taking something away. A streak resets to zero. An avatar loses health. A grid of squares gets a hole in it that will still be there in a year.
One review I read has stuck with me. Someone wrote that they could not use a popular habit app because being too sick to do a task was bad enough without being punished for their health on top of it. Another asked, politely, for an option to turn off the shaming notifications, because they did not need shame as a motivator.
I think that is right. The apps that lean hardest on guilt are the ones people delete, and they delete them at the exact moment the habit was hardest to keep, which is the worst possible time to lose your tracking.
What I changed in Multi
Two things, both small, both aimed at the same gap.
The first is called Catch Up. Every habit now carries its last seven days on its own screen. If you did the thing on Tuesday and forgot to say so, you tap Tuesday. If you tapped it by mistake, you tap it again and it goes back. Your streak reflects what you did, not how promptly you filed it.
This sounds obvious. It is not common. Most trackers let you log today and nothing else, which means a forgotten evening is unrecoverable no matter how honest you are about it.
The second is quieter. Reminders now follow the habit they belong to. If a habit repeats on weekdays, its reminder does not go off on Saturday. If it repeats every three days, the reminder shows up every three days. And if you already kept the habit at seven in the morning, the nine o'clock nudge does not arrive at all, because you are done and the app knows it.
That last one mattered more than I expected while building it. A reminder for something you have already finished is not a neutral event. It teaches you that the app's notifications are not worth reading, and once someone learns that, every future reminder is wasted too.
The wider point
A habit tracker is a record of your behaviour. When the record and the behaviour disagree, the record is the thing that is wrong, and the software should be built to admit that.
I do not think streaks are a bad idea. Seeing a run of days you kept something up is genuinely motivating. But a streak is only meaningful if it is true, and it can only be true if you are allowed to correct it.
Multi is a single app for tasks, habits and expenses, built on the idea that they are the same kind of thing wearing different attributes. Add a repeat rule to a task and it becomes a habit. Attach an amount and it becomes an expense. It runs on your iPhone, works without an account, and does its language processing on the device rather than on a server.
Multi is on the App Store, and Catch Up is in the latest version.
Sources
- Stefano DellaVigna and Ulrike Malmendier, "Paying Not to Go to the Gym", American Economic Review 96(3), June 2006, pages 694 to 719. AEA listing.