Money tracking that works with ADHD instead of against it

What standard advice assumes

Conventional expense tracking advice quietly assumes several things: that you will remember a spend long enough to log it later, that a task deferred is a task retained, that a broken streak is motivating to rebuild, and that the effort required is stable from day to day.

For anyone whose executive function does not work that way, those assumptions are not minor. They are the entire reason the system fails, and the failure gets misread as a character problem rather than a design mismatch.

None of this is medical advice, and ADHD presents very differently between people. What follows are design principles that address the specific mechanics — friction, memory, and recovery — that most standard advice ignores.

Close the gap between intention and completion

The most important principle: any gap between deciding to log something and actually logging it is where the task disappears.

"I will do it when I get home" is not a deferral. It is, functionally, a deletion. The intention was real and the retrieval will not happen, and no amount of intending harder changes that.

So the capture step has to happen at the moment, and it has to be short enough to survive an impulse-length window of attention. Under ten seconds, thumb only, from unlock to saved. Amount and category typed; account, date, and everything else defaulted.

If a form requires six fields and three dropdowns, it will not be completed at the counter, and therefore it will not be completed at all.

Put memory outside your head

The second principle: never require yourself to hold something in memory that a system could hold instead.

This applies broadly:

• Recurring bills should exist as rules with due dates, not as things you are supposed to remember. In expenie a due item surfaces on its own and waits for you to confirm it.
• Free trials should be recorded at signup with their conversion date, because you will not remember in three weeks and that is not a personal failing.
• What counts as an emergency should be written down once, while calm, so you are not deciding it in a moment where deciding is hardest.
• Loan balances, receivables, and commitments belong in a system that holds them, not in a running mental tally that resets.

The general test: if the system's correct operation depends on you remembering something unprompted, it will fail. Move that thing into the system.

Use whichever capture path has least friction right now

Friction is not constant. Typing is easy some days and genuinely difficult others, and a system with one input method fails on the difficult days.

Having several paths matters more here than it does generally:

1. Typing, when it is quick and you know the fields.
2. Photographing a receipt, when reading and typing feel like too much. The photo captures everything and the details get extracted for you.
3. Describing it in plain language, when structuring the entry into fields is the hard part. "62 at the supermarket and 40 petrol" needs no form.
4. Speaking it, when your hands are busy or typing is not happening.

In expenie all of these feed the same capture flow and produce drafts you confirm. The point is not that any one is superior — it is that on a day when one is impossible, another may not be, and the spend still gets recorded.

No streaks, no shame mechanics

Streak counters and guilt-based nudges are actively harmful here. A broken streak does not motivate a rebuild — it provides a clean reason to stop, and the stopping is the actual risk.

The alternative framing that works better: the books do not need to be perfect, they need to be broadly true. A month with three missed days and one honest estimate is entirely usable. A month abandoned because day four was missed is not.

Concretely, that means deciding the recovery rule before you need it. Something like: if I notice a gap, I reconcile my balances, add one clearly-labelled estimated line, and continue from today. No reconstruction, no restart, no assessment of what it says about me.

Having that decision already made is what makes it available in the moment, which is precisely when making decisions is hardest.

Structure beats vigilance for spending

For impulse spending specifically, sustained vigilance is the wrong tool. Structural friction placed at specific points works far better and requires nothing ongoing.

The highest-yield interventions:

• Remove saved payment details from the specific apps where it happens. One-tap purchasing removes the only pause where reconsideration could occur.
• Delete the app and use the browser. Not a ban — just enough extra steps that an impulse expires before it completes.
• A separate account with a fixed monthly amount for the problem category, so the constraint is real rather than a number you can override.
• Unsubscribe from retailer marketing, which manufactures urgency for purchases you were not making.

Pick one or two. Applying all of them at once is a crash diet and fails the same way. And target the specific place it actually happens rather than adding general friction, which is exhausting and gets dismantled.

Keep the system small

There is a strong pull toward elaborate systems — a comprehensive setup feels like it will finally solve the problem, and building it is genuinely engaging.

Resist it. Elaborate systems require maintenance, and maintenance is exactly the thing that will lapse. What survives is small: capture, a short flat category list, a quick reconcile, and a monthly read.

Every additional component is another thing that can fall out of date, and a system with stale components stops being trusted — at which point it stops being used at all.

If building the system is more appealing than running it, that is worth noticing. The version that helps is the boring one you can still operate in six months.