The Friction I Kept, and the Friction I Killed
Two apps, one year, and a rule about friction that only holds if you know what it’s for.
In the same year, I built two apps that quietly disagree with each other.
One of them, Sunday, exists to make sure nothing you need to remember ever slips. The other, V&O Expenses, exists to make sure you notice what you’re spending. Both started from the same diagnosis: something in daily life was falling through a gap because doing it right took too much effort. Both ended with me redesigning that effort down to almost nothing. Only one of them stopped at “almost.”
A task not caught in three seconds is a task lost
Sunday is a task manager built around a single bet: the moment a thought crosses your mind is the worst possible moment to also open an app, find the right list, and file it correctly. You’re driving, or mid-conversation, or holding a kid. If capturing the task costs you anything at all beyond noticing it, you won’t do it, and the task is gone by the time you’re free to try.
So the whole app is built to remove that cost. Type a task and it gets sorted into the right place without you deciding where. Set a time and a small toast confirms it landed on your calendar; “got it, it’s handled” is the entire emotional job of that beat. Nothing about Sunday asks you to slow down and think about the task before it’s captured. Thinking about it is what you do after it’s safe, not before.
That’s the whole design philosophy in one sentence: friction is the enemy, full stop, because the only thing friction produces here is a dropped task.
The same diagnosis, in a different app
V&O Expenses started the same way. Before it existed, my wife Vered and I tracked spending in a shared Excel sheet, reconciled once a month with our accountant. Adding an expense meant opening a laptop, finding the file, locating the right sheet, finding the row, typing the amount, saving it. Every one of those steps is a small tax, and BJ Fogg’s research on behavior change has a name for what a stack of small taxes does to a habit. Fogg’s B=MAP model, from his 2019 book Tiny Habits, says a behavior only happens when Motivation, Ability, and Prompt line up at the same moment. We had the motivation; we wanted to get better with money. Ability was on the floor, and there was no prompt attached to the actual moment of spending. By the time we sat down to log something, the receipt was gone and the amount was a guess. The habit didn’t survive more than a few days.
Get Omer Gaiero’s stories in your inbox
Join Medium for free to get updates from this writer.
The fix looked identical to Sunday’s fix. One tap opens a text field on the home screen. Describe what you paid for. Done in about three seconds. Ability went from the hardest part of the day to the easiest, exactly the lever Fogg’s model points to. Cut the friction, get the behavior. Same principle, same fix, same designer.
Where I stopped cutting
Here’s the part where the two apps split. V&O Expenses doesn’t only run on typed entries. It also reconciles bank and credit card data automatically, in a separate view built for the end-of-month picture. A fully automatic version of this app already exists inside it: connect the accounts, skip the typing, get a complete, accurate record of every transaction with no effort from either of us at all.
We didn’t make that the default. The manual log stayed the main, daily view, the one we actually open. Not because the automatic version is less accurate. It’s more accurate; it catches cash payments a typed entry might forget entirely. It stayed manual because the three seconds of typing “coffee, 18 shekels” is the only moment either of us actually registers that we spent eighteen shekels. Hand that job to a bank feed and the record survives, but the noticing doesn’t. You get a perfect spreadsheet of money you don’t remember spending, which was the exact feeling the old Excel sheet gave us once a month, at the reconciliation, when it was already too late to do anything but wince.
What friction is actually for
That’s not something I wrote down as a principle while building either app. It surfaced afterward, looking at the two side by side: friction isn’t one thing with one correct amount to have. It’s a lever, and which way you push it depends on what the behavior needs from the person doing it.
Sunday needs you absent from the process. Its whole value is that a task gets captured whether or not you’re paying attention to the capturing, and any friction at all works against that, because attention is exactly what you don’t have in the moment a task occurs to you. Remove it entirely and the app does its one job better.
V&O Expenses needs the opposite: it needs you present, for about three seconds, at the exact moment money leaves your hand. Automate that moment away and you’ve solved the record-keeping problem and thrown out the reason the record-keeping mattered in the first place, which was never really the spreadsheet. It was catching the decision while it was still a decision, not after it had already become a number in a report you’d read three weeks later.
The question I ask now before removing anything
I used to treat “reduce friction” as a design instruction you apply the same way everywhere, the same way I’d apply “reduce load time” or “cut the number of clicks.” Two apps in one year cured me of that. Now the question comes first: is this friction stopping the behavior, or is it the only part of the behavior the person is meant to feel? Sunday’s friction was pure cost, so it came out. V&O Expenses’ friction was doing a job, so three seconds of it stayed, on purpose, sitting right next to a fully automatic version it could have replaced and doesn’t.
Neither app is more disciplined than the other about friction. They just needed it to do two different jobs, and the only way to know which job you’re looking at is to ask what happens to the person if the friction disappears completely. Sometimes the answer is: nothing, they just do the thing more easily. Sometimes the answer is: they stop noticing they’re doing it at all.
