Why Monke Reset has no leaderboard

Monke Reset has no leaderboard because a public streak about porn gives people a reason to delete the app. The research behind the streak, a 2016 meta-analysis by Harkin and colleagues, found that reporting progress publicly made monitoring work better. We kept the private part of that finding and left out the public part, and this post explains the trade.

Many habit apps ask for a username on the second screen, keep a friends list in settings, and show your number next to your name. For a number about porn, that design asks people to accept an audience many of them won’t want.

Why track a private habit at all?

Counting is the part with the most research behind it.

Harkin and colleagues, publishing in Psychological Bulletin in 2016, pooled 138 studies and 19,951 people to ask a narrow question: if you get someone to monitor their progress toward a goal, do they reach it more often? They did, at d+ = 0.40, with a 95% confidence interval from 0.32 to 0.48. In a field where many interventions show no effect, that result holds up.

The limit: this is general self-regulation across health behaviours such as eating, exercise and weight. Nobody has run it on porn. Monitoring progress is one of the best-evidenced things a person can do about a goal in general, and that is why the streak and the calendar exist. It is not proven to fix this specifically, because that study hasn’t been done.

Does looking at your progress matter?

Yes. In Harkin’s analysis, the effect was mediated by how often progress was monitored.

Mediated means the checking does the work. Interventions that got people to look at their progress more often produced bigger effects, and without the looking, the effect goes too. A counter you never see is not a small version of this. It is a different thing with the same name. The number matters when you look at it and think about the day ahead, which is why what else the number carries matters too.

The effect was also larger when the information was physically recorded rather than held in memory. A calendar with 365 boxes is a recording surface, and marking today is the physical act the research describes.

Does making progress public help?

In Harkin’s data, yes, and that finding argues against this design.

The moderator analysis found the effect was larger when progress was reported or made public. Telling someone else, or putting the number where others can see it, made monitoring work better than keeping it private.

That result is in the same paper we cite for the streak. For a step counter or a running app, it would be an instruction to build the feed, and building it would be right. For this, we took the recording benefit and declined the publicity benefit, knowing what it costs.

Why is a public porn-free streak a bad idea?

The studies where publicity helped weren’t about porn.

Public reporting works partly because people don’t want to look bad in front of those whose opinion they care about. That pressure helps when the behaviour is neutral to admit. Missing a run is easy to admit, and nobody flinches at a chart showing a bad Thursday of eating.

Attach the same mechanic to porn and every reset becomes an announcement, to whoever is on the list, that you looked at porn last night. For many people that turns the app into something to delete, and an app off the phone has an effect size of zero.

Public numbers also turn into performance. Once others can see the streak, it stops being a record and becomes something to protect, and protecting a number is a different activity from getting through an ordinary day. People describe spending entire days worrying about losing the count, which means a day spent thinking about the thing they are trying not to think about. That pattern is covered in why you keep relapsing on day two or three. An audience makes the pressure worse and makes lying tempting, and a tracker you lie to holds nothing useful.

A leaderboard also ranks people. It turns a private count into a position and invites comparison with someone whose history, phone and past nights aren’t in the column. A number about someone else ends up read as a verdict on yourself.

What Monke Reset has instead

A streak, a best run, and a full-year calendar marking every day free and every relapse. No feed, no leaderboard, no friends list, no shareable card, no username. The number never leaves the phone, so nobody sees it, including us.

That is a trade, not a feature list. The app keeps the parts of Harkin’s findings that work in private, recording and frequent looking, and leaves out the part that needs an audience. If monitoring has a smaller effect when nobody else can see it, the app offers less than the maximum the literature describes. It is still the largest effect available to something people keep installed.

The publicity half doesn’t need a product. The research says reporting to someone helps. It doesn’t say the someone has to be a scoreboard of strangers. Telling one person who already knows is closer to what those studies did than a feed of usernames, and you choose when to stop, which a feed doesn’t allow. The case for and against is in should I tell someone I am trying to quit.

How should you use the calendar?

As a record, not a judgement.

Nothing you mark goes to anyone. With no audience, there’s no reason to leave a day blank to keep the page tidy. A month with five bad days marked is more use than a month abandoned on day six. The first shows when this tends to happen. The second is empty.

Marking the day and looking at it is the mechanism, as far as anyone has measured. Keeping it private is the cost we chose so that people keep doing it.

The source

Harkin, Webb, Chang, Prestwich, Conner, Kellar, Benn and Sheeran, "Does monitoring goal progress promote goal attainment? A meta-analysis of the experimental evidence", Psychological Bulletin, 2016, doi:10.1037/bul0000025. 138 studies, 19,951 participants, d+ = 0.40, 95% CI 0.32 to 0.48. The public-reporting moderator is in the same paper as the number on our home page. If a study is harder to find than the claim it supposedly backs, that is worth noticing, here as anywhere.