The quick version
- Nobody has tested Momentum. Nothing here shows that it works.
- It is not a treatment. It does not replace a clinician, medication, therapy or a coach.
- Its design borrows from nearby research. One choice lacks support in the closest evidence, and I did not find research for two.
Where stronger evidence sits
Randomized trials support structured skills therapy for adults with ADHD. One trial of 86 adults already on medication found that cognitive behavioral therapy beat relaxation with education at the end of treatment, and people who responded to it kept their gains at 12 months. Another, with 88 adults, found that group therapy on time management, organization and planning beat supportive group therapy on inattention symptoms.
If you are struggling, start there, with a clinician. Momentum is a small tool for one moment.
Safren and colleagues, JAMA, 2010. Solanto and colleagues, American Journal of Psychiatry, 2010.
Each design choice
Open one at a time. Each says what the research found and where it falls short.
No streaksIndirect
Across seven studies, people were less likely to repeat a behavior after their log showed a broken streak than after it showed an intact one. In one of those studies, the gap grew when they blamed themselves for the break. A separate habit study enrolled 96 volunteers and had enough data from 82. Missing one opportunity did not materially affect how the habit formed.
The limit: Neither study was about ADHD. Comparisons with no log at all gave mixed evidence on whether a broken streak does harm by itself. Intact streaks increased repeat behavior, so streaks can help while they last. The authors advise companies to avoid highlighting broken streaks, and a chance to repair a streak softened the effect. That is a different answer from having no streaks. Leaving them out is my judgment about what a bad day costs.
Acting in the moment, with the task already chosenIndirect
Adults with ADHD in one small study carried out fewer of their own real-life intentions, which statistically accounted for part of the link between ADHD symptoms and procrastination. Separately, simple if-then plans have helped goal attainment across 94 tests, and a review in children found effects that were stronger in some analyses for children with ADHD.
The limit: In the child review, children wrote their own plans in only 14 studies, and in the other 38 the experimenter supplied a plan or an example. Those studies still used a specific if-then plan, and Momentum uses none, so it may lose the benefit. The ADHD comparison had 29 adults with ADHD and 24 controls.
No guilt, and a kind answer when you stopIndirect
First-year students who reported more self-forgiveness for procrastinating before one exam reported less procrastination before the next one. This link appeared among students who had procrastinated heavily the first time, and not among those who had procrastinated little.
The limit: This was a questionnaire study of 119 first-year students from one course, out of 312 who started, so it cannot show cause. The paper does not report ADHD status. Kind wording in an app is a long way from self-forgiveness.
One card, never a listNot supported
A meta-analysis of 50 experiments found that the average harm from having more options was about zero, with wide differences between studies.
The limit: The popular claim that fewer choices always help is not well supported, and that research is about choosing among options, not starting tasks. One card is my judgment, not a finding.
Asking about your energyNone found
I did not find a study testing whether matching a task to self-rated energy helps people with ADHD start.
The limit: This is my judgment, and I did not run a formal literature search. It may do nothing.
PersonasNone found
I did not find a study testing role-specific prompts for starting tasks.
The limit: This is my judgment, and I did not run a formal literature search. The cards may also miss, because Claude drafted them from rules I set, without people from each role in the room.
Why apps deserve your doubt
A systematic review, with store searches up to 2017 and checks in 2019, included 109 ADHD apps. None of them included information on evidence for its efficacy. Until someone tests Momentum, it belongs in that group too. Saying so does not make it any better tested.
Păsărelu, Andersson and Dobrean, International Journal of Medical Informatics, 2020.
The practices I hold it to
- State the whole ask before you commit. Every card shows its length first.
- Make refusing free. "Not now" and "Stop here" never cost anything.
- Collect nothing. There is no account, no analytics code, and your progress stays on your device.
- Generate nothing live. Every card is written in advance, so the app cannot invent advice.
- Keep cards away from medical, financial and legal advice.
- Support reduced motion, keyboard use and visible focus, and keep the main buttons at least 48 pixels tall.
- Publish the limits, which is this page.
What would settle it
A randomized comparison, with adults with ADHD using Momentum or a plain timer for several weeks, measuring how often they start the tasks they name. I have not run one. A researcher who wants to is welcome to, and I will publish the result either way.
If a source here is wrong, or a card is, tell me and I will fix it.