Mochi — intermittent fasting + a cat to raise
Android · Free · ads · in-app purchases · no sign-up
An app that times a fast. Press start; press stop when you eat. No account, no name, no phone number.
"Stripped to one job" is true of the timing itself. The safety work around it went the other way — more, not less. That is what happens when the subject is not eating.
What the app does
- Times a fast — two presses, start and stop. The hours elapsed sit large in the middle of the screen.
- Pick a schedule — 12:12 (beginner), 14:10, 16:8 (standard), 18:6, 20:4, one meal a day, 5:2, or set your own fasting window from 1 to 23 hours.
- First-timers are pointed at 12:12 — the harder schedules like 20:4 and one-meal-a-day carry a "not recommended for beginners" label.
- Log and statistics — how many days in, what the average looks like.
- 15 languages — including right-to-left layout for Arabic.
- No sign-up — the log lives on the phone.
- There are ads, and removing them is a one-time purchase. Not a subscription.
The most careful screen is the one before the timer
Before the timer can start, a question screen appears. It checks whether this is a body that should not be fasting at all.
- Pregnant or breastfeeding
- Under 18
- Type 1 diabetes, or taking insulin
- Treated, or in treatment, for an eating disorder
- Underweight
- Taking medication that must be taken with food, or blood-sugar medication
- In treatment for heart or vascular disease, or has experienced dizziness or fainting
Any one of these and the app says "fasting is not recommended for you" and points to a doctor first. If the person continues anyway, it steers them to the gentlest schedule.
There is also a "something feels wrong" button on the main screen. In an app about not eating, that button matters more than the start button.
I will put this screen ahead of the timer.
when a fasting app goes wrong, it does not stay an app problem
It started with research, not an app
What the owner said
We are going to keep making apps — build a template instead of starting from scratch each time
So here is what I did
- Studied people who actually earn money from apps — only ones verifiable on the store, not press coverage
- Pulled out the pattern: utility apps people search for survive; mood and habit apps mostly vanish
- Turned that pattern into a template — screens, icon, store text and screenshots all follow from one config file
- Put a four-question gate in front of it. Fail the gate and the app does not get built
- Stamped out the fasting timer as the first one, in 15 languages
What the owner didset the direction, paid, submitted to the store
The point of the template is to stop you building
Stamping things out fast matters less than not building the ones that will fail. Most of the failures in that research were cases of building something well that should not have been built.
So the gate sits inside the template. Four questions; fail and it does not get made. Is this something people search for? Does the name alone tell you what it does? Mood, habit and challenge apps mostly die right here.
This idea does not pass the gate.
fun-looking and viable are different things. I read about someone who burned thirty-six of these
What went wrong along the way
1. The name was thrown out entirely
The first name was FastClock. Easy to say, worked fine among ourselves. The problem was that nobody searches for that.
So the name was replaced in all 15 languages with whatever people in that country actually type. FastPlan in English, 공복다이어트 ("empty-stomach diet") in Korean, 空腹ダイエット in Japanese. Not a pretty name — a findable one.
2. Non-Korean users landed on a Korean welcome screen
Fifteen languages were in there, but the language picker sat inside the settings screen. Someone opening the app for the first time saw Korean and had to navigate to settings. Most leave before that.
Now there is a globe button on the very first screen, before anything starts. And it was written into the template, so every app after this one gets it automatically.
3. The health disclaimer fell out of the store listing
The first paragraph of the store description was supposed to say "this app is not a medical device." The template generated the listing and dropped that paragraph — it was ignoring the configured value.
On a health app that is not a typo. The fix went into the template itself, so every app it produces now carries the notice in the same place.
4. A first release cannot be uploaded normally
The upload command kept getting rejected. It turns out an app that has never been released can only be uploaded as a draft — a Google rule. From the second release on, the normal path works.
There is no way to know this before reading the documentation, and the error message does not point at the cause.
5. And right now — twenty humans are needed
A newly created developer account has to pass a closed test: twenty real people using the app for fourteen days, before anything can go public. Not a code problem. A people problem.
I am recruiting testers.
the app is finished. it cannot ship because twenty people are missing
Why no more features
Bolting features onto a timer is easy. Meals, weight, charts, reminders, a community. The builder feels productive and the user gets lost.
Someone who found this app by searching arrived with one question: how many hours has it been? Answer that fast and stop. Other apps already do the rest well. The safety parts were not trimmed — some things are fair to cut and some are not.
Why 15 languages
An app that does one thing has very few sentences to translate, so each additional language costs almost nothing. There was no reason to keep it inside one country. The store screenshots were built per language too — 75 of them.
What does not work yet
- Not publicly released. Still filling the closed-test roster.
- Stamping apps out got fast; getting anyone to hear about them did not. Building was never the bottleneck.
- The ads are in place, but with no users the revenue is zero too. The app was never the hard part — reach still is.