How Much Tutoring Does a Child Need?

This is the question providers are least keen to answer precisely, for an obvious reason: the honest answer usually involves fewer hours than the one being sold. It is also the question parents most need answered, because tutoring is bought in recurring weekly units and a wrong decision compounds every month.

There is no single correct number. There are, however, reasonable ranges, a few findings worth knowing, and a set of signals that tell you when you have bought too much.

Why "it depends" is a cop-out

It does depend. But the things it depends on are knowable, and a provider who stops at "it depends" is usually avoiding the follow-up.

The variables that actually matter:

The size of the gap. Repairing two years of missing foundations is a different job from polishing a B into an A, and they should not be priced or scheduled the same way.
The deadline. Eight months before a board exam allows a pace that eight weeks does not.
Whether the child works between sessions. This is the largest single variable, and it is barely ever discussed at the point of sale.
The age of the child. Attention spans and consolidation differ enormously between a nine-year-old and a sixteen-year-old.
The subject. Sequential, procedural subjects behave differently from content-heavy ones.

Any provider who cannot walk through those five with you before proposing a schedule is proposing a schedule based on their availability rather than your child's need.

What research suggests about frequency

The best-studied version of this is high-dosage tutoring, the model that has been trialled at scale in school systems. In that literature the term is usually defined fairly precisely: small groups or one-to-one, delivered at least three times a week, embedded in the school day, over a sustained period. Programmes built that way have produced some of the more encouraging results in education research.

Two cautions before anyone reads that as "three times a week is the answer for my child".

First, that model was designed for students substantially behind grade level, delivered inside school hours, with the tutor working from the same material as the classroom teacher. Most private tutoring is none of those things.

Second, and more importantly for a family paying by the hour: the research keeps finding that consistency of attendance matters more than the nominal frequency on the timetable. A randomised evaluation of online tutoring published in the Journal of Public Economics found meaningful effects, and the Hechinger Report's coverage of the wider online tutoring evidence notes that attendance and group size are where these programmes tend to come unstuck. Sessions booked and not attended do nothing, and a programme nominally running three times a week that actually runs once is a once-a-week programme with a larger invoice.

The practical translation: a schedule your family will actually keep for six months beats an ambitious one that collapses in week five. Start at the sustainable end.

Reasonable starting points

Not prescriptions. Starting points, to be adjusted after a month of evidence.

Maintaining a subject the child is broadly coping with: one session a week, or one a fortnight. More than that is usually buying reassurance rather than learning.

Closing a specific, identified gap: two sessions a week for four to six weeks, then back down to one. Gaps close faster with frequency and then stop needing it. The step down is the part families forget to do.

Serious catch-up, more than a year behind: two to three sessions a week, sustained, with realistic expectations about timescale. This is the case that most resembles the high-dosage research.

Exam preparation, final term: one to two a week, with the balance shifting from teaching to supervised practice under exam conditions as the date approaches.

A subject that has become emotionally difficult: start with one a week regardless of the size of the gap. Adding intensity to a child who already dreads the subject tends to produce a child who dreads it more.

The single most common mistake we see is buying catch-up frequency for a maintenance problem, then continuing it for two years.

Session length by age

Length matters as much as frequency and gets less attention. Concentration, not the clock, sets the useful limit.

Ages 6 to 9: 30 to 45 minutes. Beyond that you are paying for a child to be tired in the room.
Ages 10 to 13: 45 to 60 minutes.
Ages 14 to 18: 60 to 90 minutes, with 90 reserved for exam-period practice sessions rather than new teaching.

Two shorter sessions generally beat one long one at the same total cost, for the same reason spacing beats cramming. If the choice is one ninety-minute session or two forty-five-minute ones, take the two.

Be wary of the two-hour session for a younger child. It is convenient for everyone except the learner.

How long before you should expect change

Setting this expectation in advance is the most useful thing in this article, because without it families either quit too early or continue for years on faith.

Two to three weeks: the child should be able to tell you what they are working on and why. Not improved results -- orientation. If after three weeks they cannot say what the tutor is doing, something is wrong with the arrangement rather than with the pace.

Four to six weeks: observable change in behaviour. Homework taking less time, less resistance to starting, an ability to redo last week's work unaided. This is the first real checkpoint and the one you should have agreed in advance.

One term: movement in school assessments. Not a transformation. Movement.

Two terms: if there has been no change in marks, in confidence, or in willingness to engage after two full terms, the arrangement is not working. Change the tutor, change the format, or stop. Continuing past this point is the most expensive error available.

Marks are a lagging and noisy indicator, which is exactly why the four-to-six-week behavioural checkpoint matters. By the time a report card moves you have spent a term finding out.

When to reduce or stop

Signals that it is time to step down, which almost nobody volunteers:

The child can do the homework before the session. Congratulations; you have reached the point of diminishing returns. Move to fortnightly.
Sessions have turned into homework supervision. If the hour is spent doing tonight's assignment with an adult present, you are paying tutor rates for a study hall. Sometimes that is what a family wants -- but be clear that is what it is.
The tutor is running out of things to teach. Usually visible as the session drifting toward general revision or chat.
The child asks to stop and can explain why coherently. Distinguish "I'm bored, it's too easy now" -- a legitimate signal -- from "I hate it", which needs investigating rather than acting on.
You cannot name what this term's tutoring is for. If there is no current objective, there is nothing to measure and no reason to continue at this frequency.

Stepping down is not abandoning the support. Weekly to fortnightly to monthly to nothing, with a standing offer to come back before exams, is a sensible glide path and a good tutor will suggest it before you do.

Signs you're buying too much

Tutoring on more than two or three evenings a week, alongside a full school day and homework. At some point the marginal hour is displacing sleep, and an under-slept student consolidates less of what they studied regardless of how many hours went in.
Tutoring in a subject the child is already doing well in "to be safe".
Sessions that exist because they are booked rather than because they are needed.
A child with no unstructured time left. The American Psychological Association's summary of the homework research makes the broader point that additional academic time is not reliably better, and the returns flatten.
You are unable to say what would have to happen for tutoring to end.

That last one is the real test. Tutoring should have an exit condition defined at the start. If it does not, it becomes a subscription.

The thing that decides most of it

Whatever the schedule, the work between sessions determines the outcome more than the schedule does.

A child having one session a week who genuinely practises for twenty minutes on three other days will outperform a child having three sessions a week who does nothing in between, at a third of the cost. The sessions are where the diagnosis and the teaching happen; the consolidation happens elsewhere, and it is not optional. The single highest-return change most families can make is not another hour of tutoring but a better revision method between the hours they already pay for.

This is also why the division of labour matters more than the quantity. A human deciding what this particular child should work on next, with structured practice filling in the volume between sessions, is a more efficient use of money than simply buying more of the expensive part -- which is the arrangement we set out in how AI and a human teacher fit together.

And if you have not yet decided whether to start at all, the frequency question is premature. Whether a tutor is the right response comes first, and for a meaningful share of families the honest answer is not yet.

We would rather sell you the right number of hours than the maximum number. If you are unsure what your child needs, ask us and we will tell you plainly.