When should my child start SAT prep?

Parents ask this in September and in April, and the honest answer is the same both times: start from the test date, not from today.

Almost every scheduling mistake I see comes from planning forwards — “we should probably start soon” — rather than backwards from a date that is already fixed.

First, check whether school already gives one

Several states require districts to give every junior a college admission test during the school day, paid for by the state. If yours is one of them, your child gets a real, timed, official sitting without booking anything — and that should anchor the calendar rather than being a surprise in March.

Ohio is one. Districts there must give all eleventh graders the ACT or the SAT each spring, and the state covers it — but the district chooses which test, and across Ohio that choice is very often the ACT. So plenty of Ohio students meet the ACT at school and come to the SAT deliberately, on a date they have to book themselves.

Massachusetts has no equivalent. And since MCAS stopped being a graduation requirement, there is no statewide external benchmark at all — which makes the first full practice test do more work there than it used to.

Five minutes with your child’s counsellor answers this, and it is worth doing before you plan anything else. A free official sitting already on the calendar is the cheapest baseline you will ever get.

Pick the test date first

Most students should plan to sit the SAT twice, and treat the first sitting as the real diagnostic. For a typical junior that means a spring test with a second attempt in the autumn of senior year, which leaves room for one bad day without it being a catastrophe.

Colleges take your best section scores across sittings — superscoring — at most schools, though not all, so check the specific list. But the practical effect is that a second attempt is nearly always worth booking in advance rather than deciding on afterwards.

Once the date exists, everything else falls out of it.

Working backwards from the date

Twelve weeks out: a full official practice test, timed, under real conditions. Not to get a score, but to find out which of four problems you have — pacing, content gaps, rules learned wrong, or reading itself. Those need completely different amounts of time, which is why nobody can tell you how long prep takes before this happens.

Ten to four weeks out: the actual work. Two sessions a week, with practice between them aimed at whatever the diagnostic turned up. This is where a score moves, and there is no way to compress it — you cannot cram a pacing habit, and a rule learned wrong takes several attempts to dislodge.

Four to two weeks out: full timed forms only. Stop learning new things. Practise the whole experience: the same start time, the same breaks, no phone. Students who have only ever done twenty-minute drills fall apart at the two-and-a-quarter-hour mark, and that is a fixable problem if you find it now.

The last week: nothing new, and sleep. Light review of your own error log, no new material, no full tests after about four days out. Nothing learned in the final week survives contact with test-day nerves.

That is roughly three months of real work, which is why “when should we start” usually resolves to “about a term before the date you have already picked.”

Sophomore year is usually too early

Parents who start in sophomore year are trying to be responsible, and it tends to backfire for two reasons.

The first is content: a fair amount of the maths simply has not been taught yet. Drilling it in isolation is slower and less durable than waiting until school covers it.

The second is fatigue. SAT prep has a shelf life measured in months, not years. A student who has been preparing since sophomore autumn is exhausted by the time the test matters, and exhaustion looks exactly like a plateau.

The genuinely useful sophomore-year move is the PSAT, taken seriously, and nothing else.

The exception: a diagnostic can happen any time

Sitting one full official practice test is worth doing early, whatever the plan is — sophomore spring is fine. It costs one Saturday morning and nothing else, it is free in Bluebook, and it converts a vague worry into a specific list.

That is different from starting prep. It is finding out what prep would need to cover.

When it is too late

If the test is under three weeks away, a full package will not change the score materially, and anyone who tells you otherwise is selling something. What genuinely helps in that window is narrow: pacing, a handful of high-frequency question types, and calming down.

That is a four-hour crash course, honestly described, not a two-month plan compressed into a fortnight.

The short version

• Check whether school already gives a free sitting, and when.
• Pick the date. Plan to sit twice.
• Twelve weeks out, sit one full official test to find out what is actually wrong.
• Ten weeks of real work, then a fortnight of full timed forms, then rest.
• Sophomore year is early for prep and fine for a diagnostic.
• Under three weeks, buy pacing help and nothing else.

If you want a rough version of that diagnostic before committing anyone’s Saturday, our free eight-question version takes a few minutes and hands back the same four categories a full one would.