Sit on your hands
The whole method, and where it came from.
The thing I keep coming back to
Think of the most capable person you know at work. The one who says the brief is wrong, who spots the cost nobody accounted for, who works out that everyone is solving the wrong problem.
They did not learn that at school.
They learned it in their twenties, doing junior work. Reading the documents, checking the numbers, sitting in on meetings they did not follow yet, watching someone senior make a decision and slowly working out why. The tasks were dull and repetitive. Doing them for years is how the judgment got built.
That first job is thinning out. Graduate vacancies in Hong Kong have more than halved in three years. In the United States, employment among 22 to 25 year olds in the occupations most exposed to AI has fallen while their older colleagues' has grown. The causes are argued about, and I am not qualified to settle that argument.
But every explanation ends in the same place. The apprenticeship that produced everyone senior is being removed, and school was never built to replace it.
So it has to start earlier, and it has to start at home.
How I know
I went to good schools. I passed the tests and forgot the material the next day. I pulled the all-nighters, I got the degree.
Then I started at a design firm and was handed this: design a lightweight, stackable dining chair that embodies understated Italian elegance, everyday comfort, and efficient manufacturing.
I had done briefs at school. But a school brief has no client, no budget, no factory, and nobody who is disappointed on Friday. This one had all four, and I remember thinking, where do I even start?
What taught me was everything that came after. Years of trial and error. Learning to trust my gut when there was nothing else to go on. Running five accounts at once. Being shouted at by a line manager because my proposal went over budget. Working out whether to fix the brake problem or keep saving for the trip to Greece.
None of that was on a syllabus, and all of it is what I actually use.
I could once explain the symbolism of a blue curtain in a Renaissance painting well enough to pass. So can a phone, instantly, for nothing. What a phone cannot do is stand in front of a brief like that chair and start anyway.
Photo: Anaïs Bourcier
What school is very good at
Twelve years of formal education are an extraordinarily effective machine for producing one capability: take a well-defined task and complete it to somebody else's standard, by somebody else's deadline, without complaining.
That is a real skill and I am not against it. My children need to read, write, calculate and finish things. Grades still matter, credentials still open doors, and any parenting philosophy that pretends otherwise is selling you something.
But notice what that capability is. It is execution against a specification. And execution against a specification is now available instantly, for nothing, to anyone with a phone.
What is left is the part nobody can specify in advance. Working out which problem is worth solving. Noticing what was not in the brief. Carrying something through when there is no instruction and no guarantee it works.
That cannot be practised inside a task somebody else has set, because the moment somebody sets the task, the part you were after is gone. Every hour of school, every tutorial, every graded assignment, however excellent, is practice at the other thing.
Which leaves the hours nobody has planned. Not because empty time is magical, but because it is the only condition in which the skill exists at all.
The method, in one sentence
I set children real challenges, built on something they already care about, with the hard parts left in.
Three parts, and each one is doing work.
Part one: it has to be theirs
Nobody pushes through a hard problem they are not curious about.
This is the part most parents get wrong, including me. We choose the subject because we can see what would be good for them, and then we cannot understand why they will not persist.
But look at what actually happens. A child grinding through something they chose is learning that the bad bit passes. A child grinding through something you assigned is learning that you do not listen. Same effort, opposite lesson.
So the challenge starts from what is already there. Football, video editing, streetwear, Minecraft, Greek myths, whatever it actually is. Not a subject you wish they liked.
The common objection: my child isn't interested in anything.
Usually they are, and it does not look like an interest to an adult because it has no obvious future attached. But there is a real version of this problem, which is that children cannot want things they have never encountered.
That is where a parent is genuinely useful, and it is the one job I would not hand to anyone else. Not choosing for them. Supplying material.
Last month I told my thirteen-year-old about Delia's, the mail-order catalogue I read to pieces when I was fifteen. No suggestion, no task, just something I had loved. A week later she spent three hours working out what a tech pack is, most of it frustrated, and got up early the next morning to keep going.
She had come to me excited about a clothing brand and I had asked her about brand identity, demographics and what set it apart. Every question was a good one. The fire went out.
Same parent, same child, same subject. One supplied a task. The other supplied material.
She is still going. She has changed the logo five times. She asked me what a tech pack was, and I told her they are two-dimensional technical drawings, which is the full extent of what I know about it. She asked whether I could get her fabric swatches and I said I had genuinely no idea where you would find those.
She went and found out. Not because I was useful, but because it is hers and she thinks it is fun.
That is worth saying plainly, because parents assume this requires them to know things. It requires almost the opposite.
And here is what I got wrong for a long time.
I used to hand out general briefs. A list of twenty things to try when you are bored. Or simply: how about this, or this, or this?
None of it landed. It did not land even when the suggestion was something I had watched them lose an entire hour to. Beading. The trampoline. Lego.
If it does not come from them, the curiosity to go past the bare minimum never appears. Knowing what your child likes turns out not to be the same as it being theirs that day, and a list of good options is still a list somebody else wrote.
Part two: it has to be real
Real means there is an outcome outside the family. Somebody who is not you sees it, uses it, buys it, plays it, or says no to it.
Not a project with a grade at the end. Not a certificate.
A real challenge has three properties.
No answer at the back. If it can be resolved by a search or a chatbot query, it is homework in a costume. The question has to be one nobody has answered yet, which usually means it is about a specific place, a specific person, or a specific thing they want to exist.
An external check. Somebody outside the house decides whether it worked. A supplier quotes a real minimum order. A friend uses the thing and finds it confusing. A grandmother looks at the design and does not recognise her own life in it.
A finish line they can see. Not "get better at" anything. Something that either exists at the end or does not.
Part three: the hard parts stay in
My mother is a Montessori teacher. I learned most of what I know about small children sitting bored in her classroom at the end of the day while she tidied up and did her admin.
Years later, with my first child, I went to an open day at a Montessori school. The children were allowed to pick a tray and try a station on their own. I was at a table with another mother and her nearly three-year-old, matching wooden shapes to the right holes. Circle into the circle. Pentagon into the pentagon.
The child was struggling. The mother, watching, reached over and did it for her.
The headmistress came past and told her to sit on her hands. However long it takes.
I have thought about that sentence for over a decade. It is hard to watch a child struggle. It is also the entire thing, because resilience is not a quality you have, it is a callus, and it forms one attempt at a time. We want children who can take a knock, get up, try again, fall over, get up again, because that is what the rest of life consists of.
It starts with getting the triangle into the triangle hole at two, even if it takes twenty minutes.
This is the part people find uncomfortable, and it is the whole mechanism.
A challenge with the difficulty removed teaches nothing, because the difficulty was the lesson. The sleeve that will not sit right. The first total that is far too expensive. The three evenings building an app before discovering somebody has already made a better one.
Every instinct you have will tell you to smooth those out. Doing so is the single most common way this fails.
What that looks like in practice:
Buy the parts. Then say nothing about how to do it, including when you can see the mistake coming.
When it goes wrong, and it will, the instinct is to solve it. Say this instead: "I don't know either. Come and tell me if you work it out." Then do not step in for twenty minutes. Not five. Twenty.
If they ask whether they are doing it right, resist grading it, even positively. A mark from you turns it back into an assignment.
And when it works, do not say so out loud. "Well done, you found something to do" tells them the point was you watching, and next time they will show you.
The line I will not cross
Step in for danger. Step in for real distress. Step in for something genuinely beyond them.
Frustration is none of those, and telling the difference is most of the skill.
Two questions that do more than any challenge
The challenges matter. The questions matter more, and they cost nothing.
After they make something, ask about the hardest part. One genuine compliment fused to one real question. "That bit at the end is really good. How did you get it to do that?" Then say nothing else. No lecture, no grading, no returning to it at dinner.
Then, routinely, three more. Walk me through how you got to that. What would happen if this part failed? If you could design it better, how would you do it?
None of those is a trap and none requires you to know the subject. A child asked them every week since they were nine finds them unremarkable, because that is simply what happens after you make something. A child never asked finds them frightening.
They are also, as it happens, the questions every interviewer has ever asked.
What working looks like
Not a finished app. Not a garden practice session repeated fifty times until the target falls, Beckham style.
Early on it looks like a child who complains for thirty minutes instead of an hour. Then ten instead of thirty. Then the same child who announces they are bored and wanders off anyway, mid-sentence.
Then eventually one who complains for a moment, stops themselves, and picks up a neglected Playmobil set or a piece of paper and some markers, with their favourite song on. And does not look over at you for approval.
The not looking over is the whole thing. That is the moment something is being built, and you will almost certainly miss it happening.
Eight signs it is working, none of which can be scheduled:
They start before anyone asks. They will do the boring part. They write down what happened, because they want to remember which one worked. They copy something hard on purpose. They finish it, then care how it looks. They do the talking. They go again after it fails. And they do not need you to say it is good.
And the slowest version of this is the most reliable. My daughters have been responsible for our pets since they were small. Feeding, grooming, baths. Started young enough that it never registered as a chore, because it was never framed as one. Another creature's wellbeing depends entirely on them, so it became something they stopped questioning years ago.
That is what a decade looks like. Not a project. A thing they no longer notice themselves doing.
But we don't have a spare hour
Almost nobody does, on paper.
It usually comes out of the screen time that is already scheduled as rest, or out of the twenty minutes spent arguing with a sibling about what to watch. If there is time to scroll, there is an hour in the week somewhere.
It just isn't sitting in the calendar with a label on it, which is why it feels like it doesn't exist.
And once you have found it, the worst thing you can do is fill it for them.
Does this work at 4? At 15?
Differently, and the difference is most of the point.
A four-year-old is malleable and easily delighted. Put the Hot Wheels or the beading set away for a fortnight, bring them out again, and they are brand new. You barely have to do anything except stop tidying them into view.
A fifteen-year-old wants independence and does not want a parent hovering over their time, managing it. Which is precisely what this offers them: agency, and the assumption that they can be trusted with something real. It is the rare thing you can hand a teenager that is not a rule.
What this is not
It is not anti-school. School builds a baseline that matters. If your child needs a qualification, they need it, and nothing here argues otherwise.
It is not unschooling. This runs alongside whatever education they are already getting, in hours that are currently unplanned.
It is not a course. There is no certificate, no ceremony and no gold star, on purpose. Those are the compliance system in a nicer jumper.
It is not quick. It is years of small repetitions, most of which look like nothing is happening. You will not be able to point at the moment it worked.
Where to start this week
One hour nobody has planned.
One thing they want to exist that does not exist yet. Not a hobby, an object. A shelf that fits the gap. A better version of a game they already play. Something they could sell.
Then let it go badly, and say nothing.
It will look like far too little. There is nothing to buy, book or supervise, and it will feel like you are doing nothing.
You are. That is the work, and it is the only place left where any of this gets learned.
How I can help
If you would rather not work out the first challenge yourself, that is what I do. I design a specific one for a specific child, built on whatever they are already interested in, so they have something to start on their own.
The child's part is usually straightforward. The difficult half is the parent's, particularly if you are the sort who likes to be involved, because these are meant to be worked on entirely independently. What a child needs from you is the time and the space, and then very little else.