If you install a handful of coloring apps for a three-year-old, a pattern shows up within about ninety seconds. Twelve colors on the palette, four of them usable and eight wearing a little padlock. A gem counter in the corner. A picture that, on completion, awards something — stars, coins, a spinning chest. Somewhere in the first minute, a button a child can press that opens a price.
None of this is incompetent. It is all quite carefully designed, and it is designed against a different goal than the one a parent has. We decided not to build any of it, and it is worth explaining why, because the reasoning goes further than "ads are bad".
What a locked color actually does
A coloring page is an open-ended object. There is no correct outcome. A child can make the pig blue, color only the sky, scribble over everything, or fill one region with enormous care and abandon the rest. All of those are complete successes, and the value is precisely that the child chose.
Locking eight of the twelve colors converts it into something else. Now there is a set of things you can do and a larger set of things you cannot, and the child's attention moves — reliably, within a few minutes — from what they want to make to what they are allowed to use. The padlock is more interesting than the picture. It is a closed-ended activity wearing an open-ended costume.
So we shipped all of it free. Twelve markers, each opening a shelf of its own shades, fifty-eight colors in total, and not one of them is ever locked, at any tier, for anyone. There is no version of our app where a child discovers that the good brown is for paying customers.
The problem with rewarding a finish
The gem-for-completion mechanic is the one people defend hardest, usually as motivation. The trouble is that it works — it does motivate — and what it motivates is finishing, not making.
Once a picture pays out, the fastest route to the payout is the rational route: fill every region with whatever is nearest, as fast as possible, collect, move on. A child optimising for the reward is doing exactly what the system asked of them. They are simply no longer coloring.
There is a version of celebration we do keep, because we think it is a different thing. When a page is finished ours cheers: a star, some confetti, a voice saying "You did it!", and the finished picture kept forever in the child's own gallery. The distinction we care about is that this is a witness, not a payment. Nothing accumulates. There is no counter that goes up, nothing to spend it on, and no streak to break by not opening the app tomorrow. It says *I saw that*, and then it gets out of the way and opens the next picture.
Nothing on a child's screen sells
This is the rule we have been strictest about, because it is the one where the temptation is largest and the child has the least defence.
Chapters that have not been unlocked do not advertise themselves. Tap one and it yawns — a small sleepy animation, one spoken line asking for a grown-up — and then it goes nowhere. No price appears. No store opens. No "ask a parent to unlock!" modal with a shiny button. The child is told, wordlessly, that this one needs an adult, and the interaction ends there.
Every commercial surface in the app sits behind a gate first: purchases, restores, settings, saving a picture outside the app. The gate is not a puzzle a determined five-year-old solves — three number words to tap in order on a keypad, dealt fresh after any wrong tap.
And there are no ads. Not banners, not interstitials, not a rewarded video offering a color in exchange for thirty seconds of someone else's game. There is no advertising SDK in the binary at all, which is also why there is nothing to track: no account, no analytics, no network calls on a child's path. Pictures never leave the device.
Things we were advised to build and didn't
Each of these is standard in this category and each one was a deliberate no.
- The parent onboarding quiz. Four screens of "how old is your child?", "what are your goals?", "is your child a boy or a girl?", ending on a price. It exists to warm up a purchase, not to configure anything. We ask for nothing and open straight into a picture.
- Age and gender questions. We do not ask. Children self-select by picture, which works better anyway — a two-year-old and a seven-year-old reach for different pages without being sorted first.
- Locked colors as an upgrade prompt. Covered above. All fifty-eight are free forever.
- Streaks and daily goals. A streak is a device for making a child anxious about a day off. In an app for two-year-olds it is indefensible.
- A rewarded-video color unlock. This is advertising to a toddler with a coupon attached.
What we do sell, plainly
We are not a charity and it would be dishonest to write all of the above without saying where the money comes from.
The app is free to download and a great deal of it is free permanently: every color, every tool, all three hundred tracing pages, all three hundred guided drawings, the doodle sheet, and the first chapter of every one of the nine worlds. A family can use it for years without paying, and plenty will.
What costs money is the rest of the book — the remaining chapters across nine worlds, several hundred pages of original art. One family unlock opens all of it for every child on every device, with Family Sharing, sold as a yearly or monthly plan or a one-time purchase. That decision is made by an adult, behind the gate, on a screen a child never sees.
That is the whole business model. If it stops working we will have to change something, and when we do we will write that down too.
Why any of this matters
None of these choices are heroic. They mostly cost us money, and several of them measurably reduce conversion — the quiz funnel in particular is standard because it works.
But the thing a small child is doing when they color is fragile in a specific way. They are choosing, for no reason, to care about something. That is the entire value of the activity and it is exactly what a reward loop displaces, because a reward loop supplies the reason and the child no longer has to.
An app for two-year-olds gets to decide whether it is in the business of holding a child's attention or in the business of deserving it. We would rather be quiet, be finishable, and be closed voluntarily.