The number

How much should you pay a kid for schoolwork?

Written 10 September 2026 by Daniel, who built this

Between one and five dollars a week per kid, and I would start at the bottom of that range. The way to get there is to pick the weekly number first and divide it down, rather than trying to guess what one session of math is worth. The defaults here are ten cents a session with a seventy five cent daily maximum, which sets a ceiling of five dollars and twenty five cents a week and in practice lands nowhere near it.

Decide the week, then divide

Almost everyone starts by asking what a session should be worth. That is the wrong end of the problem. A per-session price has no ceiling attached, so you cannot tell whether twenty five cents is reasonable or ruinous until you know how many sessions a motivated eight year old will do on a wet Saturday. It is more than you assumed.

The decision you are making is how much of your money moves to your child each week because of this. You already have a rough number for that, the same way you do for what a birthday present costs. Start there. Write down the most you would be content to hand over on a Friday without feeling like you had been outmaneuvered. The rate falls out of it.

A dollar a day, worked all the way through

Say you set the rate at twenty five cents per completed session and the daily maximum at one dollar. That is above the shipped default, and it is the setting I use for a nine year old, so it is the one worth walking through.

Four sessions fill the day's cap. A session is ten questions for a seven year old, fifteen up to age ten, twenty from eleven, so four of them is ten to fifteen minutes of practice. That is a real amount of math in one sitting, and most days they will not do it. Two sessions is a normal day, which is fifty cents.

Two sessions on five school days is two dollars and fifty cents a week. Over a month that is roughly eleven dollars. Over a year, call it a hundred and thirty. For one kid, for something they did most days, that is somewhere between a decent pair of sneakers and a video game.

Now the ceiling, which is the number you are agreeing to. If your kid hit the full dollar every single day, weekends and holidays included, you would owe three hundred and sixty five dollars over a year. Look at that squarely before you decide two dollars a day sounds more generous. Two dollars is seven hundred and thirty. Everyone debates the rate. The cap decides what this costs you.

How age changes it

Less than you would expect, because the session already gets longer on its own. An eleven year old answers twenty questions for the same session that pays a six year old for ten. They are doing twice the work at the same price, which means you do not have to double the rate to keep things fair.

What I would set, offered as one parent's opinion rather than anything I can prove:

• Seven and under: ten cents a session, fifty cent daily cap. At this age the money is a token and the novelty is doing most of the work.
• Eight to ten: twenty five cents, one dollar cap. This is the band where I have the most evidence, which is two children. It runs above the shipped default on purpose: the default is deliberately cautious because it is what a stranger gets before they have decided anything.
• Eleven and up: twenty five to fifty cents, cap at a dollar fifty or two. Older kids have a functioning sense of what money buys and a twenty cent balance will read to them as an insult.

For scale against what kids already get: Greenlight, an allowance app, published 2025 figures from its own customers showing an average weekly allowance of $6.66 for the youngest kids, $8.58 for ages nine to eleven, and $18.11 across the teenage years. That is one company's user base rather than a national survey, so treat it as a rough guide. It is enough to make the point, which is that math practice money sits well underneath allowance money. This is not meant to replace the allowance. It is a smaller, separate thing a kid can move on their own.

The cap is the setting that matters

A rate is a price. A cap is a budget, and only one of those protects you. With a daily maximum in place you can be relaxed about the rate, because the worst case is bounded and you have already looked at it.

The boundary pays out partially rather than refusing, which took me a while to get right. If ninety cents is already banked today against a one dollar cap and the kid finishes a twenty five cent session, that session pays ten cents. Then the money stops for the day while everything else keeps running: mastery still records, the streak still counts, the difficulty range still widens. A kid who wants to keep going is never told the work stopped mattering, only that the budget did.

There is also a minimum accuracy setting, seventy percent by default. A round below it earns nothing, and the kid is told on screen why rather than finding a mysterious zero.

Start low, because you cannot come back down

This is the one piece of advice here I feel completely sure of. Set the number lower than you think you should.

Raising a rate is a good day. You get to say the work has gotten harder and the pay has gone up with it, and a kid hears that as being taken seriously. Lowering a rate is a different conversation entirely. However carefully you explain the budget, what your child experiences is being paid fifty cents and then twenty, and the only available reading of that is punishment. You will spend a week relitigating it and it will never quite recover its shine.

So start at ten or fifteen cents even for an older kid. Watch what happens for two weeks. If they are practicing and you are comfortable, raise it and make a small occasion of it.

Too big becomes a negotiation. Too small gets ignored.

Set it too high and you have handed your child a bargaining position. Money large enough to matter is money large enough to argue about, and now every session comes with a conversation about whether tonight is worth it and whether the deal could be better. You did not want a negotiating partner. You wanted them to do fifteen minutes of multiplication.

Set it too low and nothing happens. A kid who works out that a session pays two cents has correctly concluded it is not worth crossing the room for, and no amount of framing rescues that.

The test I use is simple. Imagine paying it out at the end of the week. If the amount would irritate you, it is too high. If your kid would shrug at it, it is too low. What you are looking for is a boring number that neither of you thinks about very much. The habit is supposed to be the interesting part.

What I cannot tell you

Whether any of these figures are right for your family. The numbers I use at home came from a few months of adjusting, which is a sample of two children in one house in New York. The shipped defaults sit below them deliberately, because a default is what someone gets before they have thought about it and the cheaper mistake is the small one. The age bands above are reasoning, not research, and I would rather say that than dress them up.

There is also a per-correct-answer model as an option, and I have deliberately left per-session as the default. Paying per answer makes the money track accuracy, and I did not want a child sitting down to a hard round and working out that a hard round pays worse. The part they control is finishing the session. Some families will prefer the other version, and it is one setting away.

If you want a single number to leave with: twenty five cents a session, one dollar a day. Run it for two weeks and adjust from what you see rather than what you guessed.