◆ Study method · Method

How to Improve a Math Score

6 min read · Free to read · No sign-up required

The advice that raises a low score is not the advice that raises a high one. Applying the wrong one is the most common reason preparation stalls.

Diagnose first — always

The most common preparation mistake is studying before diagnosing. A student decides they are "bad at math", buys a book, and works through it from the beginning. Weeks later the score has barely moved, because most of that work was spent on material they already knew.

Preparation should start with a full, timed baseline, and every subsequent decision should follow from it. You are looking for three things: which skills you get wrong, how often, and how heavily those skills are tested. The product of those is what should order your study, not the order of chapters in a book.

What works depends on where you are

Score-improvement advice is usually written as though every student has the same problem. They do not. The bottleneck changes as the score rises.

From a low score

The bottleneck is usually genuine content gaps, often in foundations rather than in advanced material — fractions, negatives, basic manipulation, translating words into equations. Progress here comes from rebuilding those foundations properly. It feels slow and pays the largest dividends, because everything else rests on them. Speed work at this stage is wasted; accuracy comes first.

From a middle score

Usually a mix: some real content gaps, plus a substantial share of process errors — inefficient setups, wrong route selection, mistranslated word problems. This is where diagnosis pays most, because the remaining gaps are specific and scattered rather than general. Broad practice is inefficient here; targeted practice is transformative.

From a high score

The content is largely known. What remains is careless error, pacing, and a small number of genuinely hard question types. Additional content study returns almost nothing. Progress comes from process discipline — checking that you answered the right quantity, writing intermediate steps, eliminating the specific recurring slips in your error log — and from sitting enough full timed sections that fatigue and pressure stop producing mistakes.

The practical implication: a high-scoring student who responds to a plateau by studying more content is doing the wrong work, and a low-scoring student who responds by drilling speed is doing the wrong work too.

Why plateaus happen

A plateau almost always means one of four things:

• Your practice no longer matches your weaknesses. You are working through material that no longer contains what you get wrong. This is the most common cause by far.
• You are practising what is comfortable. Marks are lost where you are weak, and weak areas are unpleasant to work on.
• Your remaining errors are not content errors. Careless and pacing errors do not respond to content study, and a student can add many hours without touching them.
• You are not practising under real conditions. Untimed accuracy improves while timed accuracy does not, and only the second is measured on test day.

Note that "not enough hours" is not on that list. Plateaus are usually a targeting problem, not a volume problem — which is fortunate, because targeting is easier to fix than finding more time.

Measure honestly

Study time is not progress. Questions completed is not progress. The only measurement that matters is performance under exam conditions.

• Sit full timed sections regularly and track the results. Fewer, more honest measurements beat frequent casual ones.
• Track by skill, not just by total. A flat total can hide real movement — one area improving while another slips.
• Re-test what you fixed. Understanding a fix is not evidence that it held. A weakness is closed when a later attempt confirms it, not when the explanation made sense.
• Expect noise. Single-session results vary. Look at the trend across several, not at any one.

What does not work

• Doing more questions without diagnosis. Volume reinforces what you already do, including the mistakes.
• Reviewing solutions passively. Feels productive, teaches recognition rather than recall. See study strategies.
• Cramming. Works for a day, does not survive to test day.
• Studying only what is enjoyable. Comfortable and close to useless.
• Waiting for a "system" to fix it. No platform, book or tutor substitutes for the work — they can only make sure the work is aimed at the right target.

A note on expectations

We publish no average improvement figures, and you should be sceptical of anyone who does. How much a score moves depends on the starting point, the time available, the quality of the targeting and — most of all — the work actually done. Any single number offered as a guarantee is either unverifiable or true only of a selected group.

What can be said honestly is this: a student who diagnoses accurately, practises against the diagnosis rather than around it, works consistently, and measures under real conditions is doing everything within their control. That is the method. The outcome follows from it, and no honest source will promise the number.

Frequently asked

How can I improve my SAT Math score quickly?

Diagnose first, then target. Most rapid improvement comes from discovering that a handful of specific skills — or a specific recurring careless error — account for a large share of lost marks. Practising broadly is the slowest route, because most of that time is spent on material you already know.

Why has my score stopped improving?

Usually because your practice no longer matches your remaining weaknesses, or because your remaining errors are careless and pacing errors that content study does not address. Plateaus are almost always a targeting problem rather than a volume problem.

How much can I improve my math score?

We do not publish improvement figures, because any honest answer depends on your starting point, the time you have and the work you do. Be sceptical of anyone offering a guaranteed number — what you can control is the method: diagnose, target, practise consistently, measure under real conditions.

Is it too late to improve before my exam?

Rarely, though what to do changes. Late in preparation, a timed section plus a ranked weakness list is a rational use of remaining time — fixing two or three high-frequency errors is achievable in a short window. Broad content review at that stage is not.