Timing Your Child on Maths: When It Helps and When It Backfires

23 August 2026 · by Larry

A timer is the most tempting tool a parent has. It is easy to set, it produces a number, and the number seems to go in the right direction if you use it enough.

It is also the tool most likely to make a child worse at maths, and the reason is not that timing is bad. It is that timing does two completely different things depending on what is being timed, and most of us apply it to both.

The one distinction that matters

Split maths into two piles.

The first pile is facts that should be automatic. Number bonds to ten. Times tables. Simple addition within twenty. These are meant to be recalled, not worked out. A child who computes 7 × 8 by counting up in eights will get the right answer and will still be at a disadvantage, because the working memory they spent on it was needed for the actual question.

For this pile, speed is the skill. Timing here is not pressure — it is the measurement of the thing you are trying to build.

The second pile is everything that requires reading and thinking. Word problems. Multi-step questions. Anything where the child has to work out what is being asked before they can start.

For this pile, a clock is actively harmful, and here is the specific reason: the correct behaviour on a hard word problem is to read it again. A timer turns rereading into a penalty. So the child stops doing the one thing that would have got them the mark.

What going wrong looks like

The damage does not announce itself. It shows up as habits that look like carelessness:

• Answering before finishing the question. They see two numbers and a familiar shape and commit, because stopping to check feels like losing.
• Accuracy falling as speed rises. If the time improves and the marks drop, the timer is not measuring progress. It is buying speed with correctness.
• Guessing rather than leaving it. A guess ends the discomfort immediately. Thinking does not.
• Not wanting to start. This is the loudest signal and the easiest to misread as laziness. A child who liked maths last term and now stalls at the table has usually learned that the table is where they are found wanting.

If you see the last one, stop timing entirely for a few weeks. Nothing is lost. The facts will still be there.

If you do use a timer, four rules

1. Time the topic, not the child. "Let us see how fast the six times table can go" is a different sentence from "let us see how fast you are". The first is a game against a table. The second is a verdict.

2. Compare them only against themselves. Never a sibling, never a classmate, never a number a friend mentioned at pickup. Their own time last week is the only useful comparison, and it is the only one that is fair.

3. Only count a time when everything was correct. This one matters more than it sounds. If a fast round with three mistakes can beat a slower perfect round, you have told your child, clearly and in a way they will believe, that finishing matters more than being right. When we built the timed mode into StudyLab, we made a best time record only on a fully correct round for exactly this reason.

4. Never time something being learned. New method, new topic, first week — no clock. Time it once it is understood and you are trying to make it quick. Timing during learning teaches a child that not-yet-knowing is a failure of speed.

What we do in the app, and why

Practice mode is the default and there is no timer in it at all. Timed mode is a toggle the child chooses, not the state they arrive in.

Two deliberate details inside timed mode. First, a timed round does not show the worked explanation between questions — because a clock that runs while a child reads why they were wrong punishes them for reading it. Second, anything they got wrong comes back afterwards in Fix My Mistakes, untimed, with the full working shown.

Round length also adjusts to how the child is actually doing on that topic, and it shortens when accuracy is low, so a bad day ends sooner rather than grinding on. Ending on a win is not a soft idea. It is the difference between a child who comes back tomorrow and one who does not.

What to say instead of "faster"

The instinct is to say "quicker" or "come on". Some sentences that do more:

• "Read it to me." Half of all word-problem errors disappear when the child says it out loud, because they hear what it is asking.
• "What are they actually asking for?" Before any calculating.
• "Does that answer make sense?" A sense-check is a real mathematical habit and it is invisible on a clock.
• "Which bit is the sticky bit?" This gets you a diagnosis. "I do not know" usually means one specific step, and children can often point at it.

The thing worth remembering

There is a version of fast that comes from a child finding the work easy, and a version that comes from a child being afraid of the clock. They produce similar times for a while, and then they diverge sharply — usually around the point the questions start needing thought.

Use the timer on the facts that deserve to be automatic. Leave it well away from the thinking. And if the number is improving while the marks are not, believe the marks.

I wrote about doing exactly this with my own daughter in why I stopped timing her — this piece is the practical version of that. If you want the practice side of it, the maths practice has both modes and starts untimed. If your child is at the stage of counting on fingers, when that matters and when it does not is the more useful piece to read first, and careless mistakes and what they usually are covers the errors a timer tends to create.

Related: if the real question in your house is not how to time the work but how long maths homework should actually take, that one is about where the evening went rather than how fast each question was.

Related: if the freezing started when the timing started, when your child freezes on a question they know is the piece that follows this one.

Children read finishing early as being clever, which is one reason speed gets over-valued at home. Comparing your child to their classmates covers why that signal is unreliable.