Why Mood Tracking Apps Get Abandoned (and What the Retention Data Says)
Updated: 1 day ago
Mood tracking only pays off after a couple of weeks, because a pattern needs enough days to show up. That makes it unusually sensitive to one number: how many people are still tracking on day 14. Two large public datasets put that number lower than most people would guess.
Dataset one: 93 popular mental health apps
In 2019, Amit Baumel and colleagues at the University of Haifa pulled objective usage data from an analytics panel for 93 Android apps targeting depression, anxiety, or emotional well-being, each with at least 10,000 installs (median 100,000). The panel reported the share of installers who opened the app on each of the 30 days after installing it. The paper is open access in the Journal of Medical Internet Research.
The headline medians:
Daily open rate, meaning the share of people with the app installed who open it on a given day: 4.0%.
15-day retention: 3.9%.
30-day retention: 3.3%.
The authors describe a decline of more than 80% in open rates between day 1 and day 10.
By primary technique, 30-day retention was 6.1% for tracker and diary apps, 4.7% for mindfulness and meditation apps, 8.9% for the two peer-support apps, and 0.0% for breathing exercise apps.
We turned those medians into counts to make them concrete. Out of 1,000 people who install a typical app in this set:
About 39 are still opening it on day 15.
About 33 are still opening it on day 30, so 967 have stopped.
For a tracker app specifically, about 61 of 1,000 make it to day 30, the best of any technique apart from peer support, and still fewer than one in 16.
The mood tracker's two-week payoff lands right where the curve has already flattened near zero. Most installers never see a single pattern.
Dataset two: 109,914 people in remote health studies
Apps built by researchers do not fare much better. In 2020, Abhishek Pratap and colleagues combined usage logs from eight smartphone-based health studies run between 2014 and 2019, covering 109,914 participants and about 3.5 million completed tasks. The paper is open access in npj Digital Medicine. Their findings:
Median time in a study across all eight: 5.5 days. Individual studies ranged from 2 to 26 days.
More than half of participants stopped within the first week.
The one outlier, a depression study called Brighten, kept its median participant for 26 days.
Participants referred by a clinician stayed a median of 44 days versus 4 days for people who found the study on their own.
Adults 60 and over stayed a median of 7 days, longer than younger groups.
The two datasets agree. Left to an icon on a home screen, most people stop within a week or two, and the ones who keep going are the ones with an outside push, whether a clinician, a condition that makes the data matter, or a study that comes to them.
What the numbers say about why
Three things are consistent across both sets.
The app has to be remembered. An open rate of 4% means that on any given day, 96% of people who have the app did not think of it. Nothing about the app changed; the user's attention did.
Prompting works. A small randomized trial from 2012 that prompted young people on their phones to log mood several times a day saw an average of 3.3 entries per day and an average of 17.7 days in the program, with 56% of the monitoring group completing at least two entries a day for 14 days. That trial is in the Journal of Medical Internet Research too. It was a supervised study rather than a store download, so the comparison is not exact, but 56% reaching day 14 against 3.9% is a different order of magnitude.
The reply has to be tiny. Every extra screen between the prompt and the entry is a place to quit.
Designing for day 14
Take the retention curve as the design constraint instead of a disappointment. The question becomes: what is the smallest thing a person can do every day for two weeks without having to decide to do it?
A text message that arrives on its own and takes a few words to answer is about as small as it gets. It does not need to be remembered, found, or updated. That is the bet behind Mood Textapp: the check-in comes to you at 9pm, the reply is whatever is true, and the patterns summary is waiting at the two-week mark that most app users never reach.
A tracker, by text or otherwise, is a way to see your own patterns. It is not treatment. If low days have become the norm, talk to a doctor or therapist, and if you are in crisis, call or text 988 to reach the 988 Suicide & Crisis Lifeline.
Try it by text
Text Hi to (572) 223-3345 to start Mood Textapp. There is no app to install, nothing to remember to open, and it works on any phone that can text. Reply to the evening and morning check-ins in a few words, and ask what patterns it has noticed after two weeks to see what your low days have in common.
Your first 20 messages are free. After that, one $9.99 a month subscription covers every Textapp. Details are on the Mood Textapp page. Text STOP at any time to end it.
Comments