Sightline PrepGet started

How many hours of SAT tutoring does a student actually need?

Parents ask me this before they ask anything else, and they usually apologise for asking. They should not. It is the right question, and the answer decides whether the money does anything.

I am paid for my time, so read what follows with that in mind. I would rather tell you the truth and lose the booking than sell you twenty hours you did not need — partly because I teach every session myself, so a family working through hours they do not need is a real cost to me too.

The short answer

For most students I work with, a first package of ten hours is enough to find out whether the approach is working, and enough to move the specific things we find. Many families do a second package after that. A smaller number do a third.

But the number is not really the point, because it depends almost entirely on which of these you are dealing with.

It depends on what is actually wrong

A pacing problem is the cheapest thing on this list. A student who knows the material but runs out of time in the second module is usually working from one or two identifiable habits — rereading the passage before the question, or refusing to move on from a hard item. That is a handful of sessions and a lot of timed practice they can do alone. It is also the most common problem I see, and the one families are most often surprised by.

A few specific content gaps — circles, say, or rhetorical synthesis — are also cheap. A topic that was never taught properly takes about a session to teach and then some homework. If a diagnostic turns up four such topics, you can roughly count on four sessions plus practice.

A rule learned wrong takes longer than either, and is the reason I care so much about which questions a student got wrong while feeling certain. If somebody has been confidently applying a broken rule for two years, we have to find it, prove it is broken in a way they believe, and then rebuild the habit under time pressure. That is several sessions for one rule, and more practice tests than anybody wants.

A reading problem is the expensive one. A student who cannot reliably follow a dense paragraph is not going to be fixed by SAT strategy, because the SAT is not the problem. This takes months and it is not really tutoring for a test. I will tell you if that is what I think is going on.

When more hours stop helping

There is a point where the useful work is all homework, and paying someone to watch you do it is a waste. You have hit it when sessions start being mostly review of work that went fine, or when the error log stops surfacing anything new.

I would rather move a student to every other week at that stage. That is a smaller invoice, and it is the correct thing to do.

Three situations where tutoring is the wrong purchase

The test is in under three weeks. There is no version of this where ten hours changes a score materially in that window. A four-hour crash course to tighten pacing and calm nerves is honest; a full package is not.

Nobody has taken a full official practice test yet. Buying tutoring before a diagnostic is buying a solution before anyone has identified the problem. Sit one full Bluebook form, timed, and then decide. This costs nothing.

The student does not want to be there. I can teach a student who is anxious, a student who is behind, and a student who thinks the whole thing is stupid. I cannot teach one who will not do any work between sessions, and neither can anyone else. If that is where you are, the money is better spent on figuring out why.

What I would actually do first

Sit a full official practice test under real conditions — timed, one sitting, Bluebook, no phone. Then look at the questions rather than the score: which ones were wrong, which were right but slow, and which were wrong despite feeling certain.

Those three categories point at completely different amounts of work, which is why “how many hours” cannot be answered honestly until somebody has looked at them. If you want a rough version of that read before committing to anything, our free diagnostic does it in a few minutes with eight questions — no account, no email.

And if you would rather just ask me, the first two hours are free and I will give you my honest estimate at the end of them, including if the estimate is zero.