The ADHD App Graveyard: Why Productivity Tools Don't Stick

Everyone with ADHD has the graveyard: the planner bought in January, the app downloaded with real hope, the beautiful system that lasted nine days. The standard explanation is that you didn't stick with it. We think the standard explanation has it backwards — and you can see why by looking at exactly where a calendar app asks you to start typing.

The graveyard is real, and it's specific

Read any ADHD forum and the same shape recurs. Someone downloads a tool with genuine intent, then spends hours configuring categories and tags and recurring rules — and by the time the setup is finished, whatever energy was going to power the habit has already been spent on the setup. Someone else describes keeping dozens of to-do lists in various states of neglect, with the punchline that at least they know where all of them are. Someone else remembers, every evening on the drive home, the whole list of things that need doing when they arrive — and has no way to write any of it down at the moment it exists, so it's gone by the door.

Those are three different failures, and none of them is "didn't try hard enough." One is that setup consumes the novelty that was supposed to fuel the habit. One is that capture systems multiply faster than they resolve. One is that the moment of remembering and the moment of recording are not the same moment.

What a calendar app actually takes off your plate

To answer that properly you need to know what the load is made of. Sociologist Allison Daminger's 2019 study in the American Sociological Review established that the non-physical side of running a household is a distinct kind of work — cognitive labor — with four components:

Now hold that next to the moment you add an event to a calendar. You are typing "Dentist, Thursday 4pm."

By then you have already anticipated (something in a school email or a phone call registered as a commitment), identified (which slot, which parent drives), and decided (Thursday it is). All three happened in your head, unassisted, before the app was involved at all.

The input box sits after the expensive thinking is finished. A conventional calendar offloads the last and cheapest component of the load — and, in exchange, adds a fifth step that wasn't there before. That is not a small design quibble. It's the arithmetic of the whole category.

The hinge nobody mentions: offloading is a trade, not a gift

Here is where it stops being an argument about tidiness and starts being about measured cognition.

The evidence for external reminders is genuinely striking. In laboratory studies, people relying on their own memory forgot intended actions around 45% of the time; with an external reminder, about 5%. A 2022 review by Sam Gilbert and colleagues — the field's leading work on intention offloading — puts the effect at nearly an order of magnitude, and a 2026 meta-analysis finds the benefit is largest precisely for remembering to do future things.

But read the same literature one paragraph further and you find the condition everyone drops. Offloading is described there as a cost–benefit strategy: people weigh the effort of setting the reminder against the likelihood of remembering without it, and they offload when it's cheap. Holding an unfulfilled intention in your head carries a real cost — but so does the act of putting it somewhere else.

So the graveyard isn't a discipline failure that happens despite the tool. It's the outcome the offloading literature would predict from a tool with an expensive front door. The app didn't fail because people are inconsistent. People were inconsistent because the app charged too much at the moment of capture — the one moment when the intention is fragile, the hands are on the steering wheel, and the alternative (just remember it) is free right up until it isn't.

Why "just use a simpler app" doesn't fix it

If capture cost is the problem, the obvious answer is a lower-friction capture: quick-add, a widget, voice. Those help at the margin — but they're still on the wrong side of the line, because they still require the user to be the one who noticed.

Look again at the table. Even a perfect one-tap capture only shortens the fifth step. It does nothing about the first three. The school email still had to be read by a human who registered that paragraph four contained a date that mattered. That noticing is the anticipate component — the first, most invisible, least creditable piece of the work — and every tool in the category takes it as an input rather than doing it.

This is also why "the ADHD partner won't use it" is such a stubborn pattern. A tool that requires capture doesn't just require effort; it requires the person with the least reliable capture machinery to perform capture reliably, forever, as the price of relief. That's not a motivation problem either — and it compounds with the difficulty with time that made the appointment hard to hold in the first place. If you want the head-to-head on the tools themselves rather than the argument about the category, we compared ADHD calendar apps separately.

Where every tool actually sits on this line

Once you stop comparing feature lists and start asking who does the noticing, the category sorts itself into three groups — and the sort is unflattering to almost everyone, us included.

Look at the fourth row carefully, because it's the one people mistake for the fifth. Forwarding is a genuine improvement — it deletes the typing, which is real work — and Jam and Sense both do it well. But read what it still asks: find an email with event details and forward it. You have to open it, register that paragraph four contains a date that matters, and act. That's anticipate, plus a capture action. It's the expensive part, still yours.

What would actually change the arithmetic — and what we can't claim

The interesting design question isn't "how do we make typing faster." It's how far upstream can the boundary move — can a system do the noticing, so that what reaches you is already an option rather than a chore?

That's the premise we build on: if a school email arrives, the event should already exist by the time you look, and your job should be to confirm or ignore it, not to transcribe it. We think that's the meaningful line in this category — not which app has nicer chore charts. And in the one place a real user put it plainly, the value they named wasn't the reminders; it was that it didn't all have to live in their brain any more.

We build a family organizer, so we have an obvious stake in the conclusion that capture cost is the thing that matters. Read this as an argument to check rather than a finding to accept — the two sources it rests on are linked, and both say less than we'd like them to.

Frequently asked questions

Why do I keep abandoning productivity apps?

Probably not because you lack discipline. Research on intention offloading describes it as a cost–benefit trade: people put a reminder into an external system when doing so is cheap, and skip it when it isn't. A tool with expensive capture — setup, categories, repeat rules — raises that cost at exactly the moment the intention is most fragile, so the intention stays in your head instead. Note this work was done on general-population adults, not ADHD samples.

What part of the mental load does a calendar app actually take?

Mostly the last part. Household cognitive labor has four components — anticipating a need, identifying options, deciding, and monitoring the outcome. By the time you're typing an event in, you've already done the first three unaided. The app takes over monitoring, and adds data entry, a step that didn't exist before.

Doesn't a quick-add or voice shortcut solve this?

It shortens the typing, which helps at the margin. It doesn't change who did the noticing. Every tool that starts at the input box takes anticipating, identifying and deciding as its inputs rather than doing any of them.

Is automatic capture proven to reduce mental load?

No, and we won't pretend otherwise. The mechanism is well supported — external reminders dramatically improve follow-through, and cheap offloading is more likely to happen than expensive offloading. But we could not locate an evaluated intervention showing that any tool, including ours, reduces measured cognitive load. The argument here is about mechanism, not outcome.

About this piece. Written by the Kinmory team. We build an AI family organizer whose whole premise is the argument above, which is a reason to check our sources rather than trust our framing. The user accounts are paraphrased from public ADHD communities and are illustrative, not evidence. Where the research is general-population, qualitative, or absent, we've said so in the text rather than in a footnote.

• Daminger, A. (2019). The Cognitive Dimension of Household Labor. American Sociological Review 84(4):609–633. doi:10.1177/0003122419859007
• Gilbert, S. J. et al. (2020/2022). Intention offloading — cost–benefit account and the ~45%→~5% effect. pmc.ncbi.nlm.nih.gov/articles/PMC9971128
• Risko, E. F. & Gilbert, S. J. (2016). Cognitive Offloading. Trends in Cognitive Sciences 20(9):676–688.
• Burnett, L. K. & Richmond, L. L. (2026). Meta-analytic investigations of the effect of cognitive offloading on memory-based task performance. Memory & Cognition 54(1). pubmed.ncbi.nlm.nih.gov/40500483