Screen time or creation time: why what your child makes matters more than the minutes

Counting minutes tells you very little about what a screen is doing for a child. The better question is what the child has at the end of the session. A practical way for parents and teachers to judge screen use by what it produces, and why we refuse to build streaks.

Ask a room of parents what worries them about screens and the answer usually arrives as a number. An hour on a school night. Two at the weekend. The family argument is fought in minutes, the parental controls are set in minutes, and the guilt is measured in minutes.

We think the minute is the wrong unit. Not because time does not matter - a child who is on a device until midnight has a sleep problem whatever they were doing - but because the same thirty minutes can contain completely different childhoods. Thirty minutes watching an algorithm serve up the next video is one kind of half hour. Thirty minutes writing a story, arguing with an AI editor about whether the dragon should survive chapter two, is another kind entirely. A clock cannot tell them apart. A parent can, if they know what to look for.

Consumption has a shape, and so does creation

Consumption on a screen has a recognisable posture. The child is leaning back. The content arrives without being asked for, one piece after another, each chosen by software whose only goal is that the next piece gets watched too. Nothing the child does changes what they are becoming; they are the audience, and the session has no natural end because it was designed not to have one.

Creation has a different shape. The child is leaning in. There is friction - a sentence that will not come out right, a quiz question that turns out to be too easy, an AI answer that is not quite what they meant, so they have to ask again, better. Progress depends on effort. And crucially, the session has an end built into it, because the thing being made is either finished or it is not.

The distinction is not screens versus no screens, and it is not even apps versus other apps. The same device, sometimes the same tool, supports both postures. Which is why the useful audit question is not "how long was my child on that?" but "what exists now that did not exist before they started?"

Sometimes the honest answer is: nothing. That is fine occasionally - adults watch television too. It becomes a problem when nothing is the answer every day.

Streaks are the tell

Here is a quick way to judge any app aimed at your child: look for what happens when the child tries to stop.

An app built around creation lets them stop. The half-written story is saved. The finished quiz is done. There is nothing to lose by putting the tablet down, so the decision to come back tomorrow belongs to the child and, at these ages, to you.

An app built around engagement punishes stopping. Streaks that break. Daily rewards that expire. Notifications timed for the exact hour a child gets home from school. Leaderboards that reset. Every one of these mechanics was invented to solve a business problem - retention - and every one of them works by taking the decision to stop away from the child and handing it to the product. When those mechanics appear in something labelled educational, the label deserves scrutiny. A streak does not care whether your child learnt anything on day forty-three. It cares that they opened the app.

We hold a hard line on this in our own product, and we state it publicly so that we can be held to it: no streaks, no daily-reward mechanics, no notifications engineered to pull a child back, no feed. The tutor helps with the lesson and then stops. A learning tool for children should be designed to be put down, and "time in app" should never be the number a children's company celebrates. If a vendor cannot say the same, ask them why not, and listen carefully to the answer.

What creation time looks like with AI

AI tools sit awkwardly in the middle of this divide, because they can serve either posture. Used one way, an AI is the ultimate consumption machine: infinitely patient, always ready with an answer, happy to do the homework and the thinking that goes with it. Used another way, it is the most capable creative collaborator a child has ever had access to.

The difference is in who is doing the making. Some patterns we consider genuinely creative:

• The child as author, the AI as editor. The child writes; the AI suggests, questions, and points at the flabby sentence. The words stay the child's.
• The child as designer. They build a quiz for the family, a treasure hunt, a revision game - and the AI helps them test whether the questions actually work.
• The child as examiner. They ask the AI to explain something, then check it against a book, a teacher, or a parent, hunting for the mistake. Finding one is a triumph, and it teaches more about how these systems work than any lecture could.
• The child as explainer. They have to teach the topic back - to the AI, to a sibling, to you - in their own words. If they cannot, the session is not finished.

Notice what all four have in common: at the end there is an artefact, or a demonstrable new capability, that belongs to the child. That is the test.

The end-of-session question

You do not need a dashboard or a philosophy degree to apply any of this. You need one question, asked kindly and consistently at the end of screen sessions: "show me what you made."

Asked often enough, the question changes behaviour on both sides of it. Children start choosing activities that give them something to show, because showing is satisfying. And parents get a far more accurate picture of screen life than any timer can offer - because a child who spent forty minutes making something will talk about it at dinner, unprompted, while a child who spent forty minutes being fed content usually cannot remember what it was.

Teachers can run the same audit on classroom tools. When the pitch is engagement metrics, ask to see what pupils produce instead. Completion rates tell you the software was open. Work tells you a mind was.

What to try this week

• Run the end-of-session question for seven days. After each screen session, ask "what did you make?" or "show me". No judgement attached, just interest. Notice which apps ever give your child an answer.
• Audit one app together. Pick the app your child uses most and look for the retention machinery: streaks, expiring rewards, notification settings, autoplay. Show your child what you find and explain, in one sentence each, what that mechanic is for. Children are affronted by manipulation once they can see it, and that affront is protective.
• Swap one session, not all of them. Take one habitual consumption slot this week and replace it with a making slot: a story with the child as author, a quiz built for Sunday lunch, a fact-check hunt. Keep it under half an hour. End when the thing is made.
• Say the quiet part at dinner. Tell your child the minutes matter less to you than the making. It changes what they optimise for, and it moves the family argument onto ground where you can both win.