Evidence

Does accountability actually work?

Partly. Deciding in advance exactly when and where you will do something has strong evidence behind it. The narrower claim that reporting your progress to another person makes you dramatically more likely to succeed rests on much thinner ground than the productivity industry admits.

We sell an accountability product, so treat the rest of this page with the suspicion that deserves. We have tried to describe the research accurately rather than flatteringly.

What the evidence clearly supports

The strongest finding in this area is about planning rather than about accountability. If you decide the specific time, place and trigger for an action ahead of time, you are considerably more likely to do it. Psychologists call these implementation intentions.

Peter Gollwitzer and Paschal Sheeran pulled together 94 separate tests covering more than 8,000 participants in a 2006 meta-analysis and found a medium-to-large effect on whether people actually reached their goals. That is a large body of work pointing the same way, which is rare in this field.

The practical version: "I will write the proposal between 9am and 11am on Wednesday, at my desk, with my phone in a drawer" works. "I will get on top of the proposal this week" does not.

Where the 42% figure comes from

If you have read anything about goal setting in the last decade you have met the claim that writing your goals down makes you 42% more likely to achieve them. It traces back to a study by Gail Matthews at Dominican University of California. She recruited 267 people, split them into groups, and found that more than 70% of those who sent weekly progress updates to a friend reported meaningful progress, against around 35% of those who kept their goals to themselves and never wrote them down.

That is a genuinely interesting result. It is also a single study with 267 self-selected participants, outcomes the participants graded themselves, and a write-up presented at a conference rather than published in a peer-reviewed journal. None of that makes it wrong. It does mean the confident percentages you see on coaching websites are resting on one modest piece of work.

What this means for accountability apps

Most accountability tools ask you to tell them how you are doing. The problem is obvious once you say it out loud: your own reporting is the least reliable input available, and it gets less reliable exactly when things are going badly. People stop opening the app in the week they most need it.

A tool that reads your calendar and your task list has a different problem and a different advantage. It can see that you booked no focus time, or that the same task has moved four times, without waiting for you to confess. It cannot see the things that never reach a calendar, so it is not a complete picture either.

What Angry Assistant claims, and what it does not

We claim that a daily email written from your real calendar, tasks and health data will tell you specific things you would rather not think about, and that being asked to name what you are doing today is more useful than a reminder that you have a busy day. That is consistent with the implementation intentions research, which is why every email ends with one concrete action and a question it wants answered.

We do not claim a percentage. We have not run a trial, we have a small number of users, and anyone quoting you a conversion rate for this kind of product is either citing Matthews or making it up.

Common questions