Sit the test again with the same preparation and you will score close to the same number. Before planning anything, split your lost points into two piles: content you did not know, and execution under real conditions. They need opposite fixes and most people guess wrong about which one they have. If your practice scores sat far above your real score, suspect the practice was not run honestly. Then spend the weeks between sittings reviewing what you already did rather than taking more tests.
The decision about whether to retake is a separate question, and we have a piece on that. This one assumes you already answered yes. The trap now is momentum: registering for the next date, promising yourself you will study harder, and doing roughly what you did last time with more hours attached. That reliably produces roughly the same score.
"I will study more" is not a plan, it is a mood. Name the thing that will be different or admit nothing will be.
First: which kind of points did you lose?
There are only two kinds, and they are fixed by opposite means.
Content losses are questions you could not do. You did not know the rule, the topic was unfamiliar, you would miss it again tomorrow with unlimited time. These are fixed by learning things.
Execution losses are questions you could do and did not. You ran out of time before reaching them, you misread what was being asked, you solved correctly and answered a different quantity, you changed a right answer to a wrong one. These are fixed by changing how you work, and studying more content does nothing for them.
The reason this matters more than any other decision: a student who spends six weeks learning new material to fix an execution problem has wasted six weeks, and will be genuinely confused about why the score did not move.
I was sure I needed harder questions. Turned out I never reached the last four in either math module. Harder questions would not have helped me reach them faster.
The practice-score audit
If your practice tests were coming out well above what you actually scored, that gap is the first thing to explain, because everything you plan next depends on which of your two numbers is real.
Ask yourself, honestly, about the practice tests you were using as evidence:
- Were they done in one sitting, start to finish, with the real break and nothing else in between?
- Was the timer strict, with no pausing to make tea, answer a message, or look something up?
- Was it morning, at a desk, with your phone somewhere else?
- Was it a test you had never seen before, or one you had already worked through?
A "yes, mostly" to those is usually a no. If any of them slipped, your practice number was inflated and your real score is closer to your true baseline than you want it to be. That is not bad news; it just means you now know where you actually stand, which is a better place to plan from than a fantasy. Our piece on how accurate Bluebook scores are goes further into why the gap appears even when you do everything right.
Two numbers disagree. Before you decide which to believe, ask which one was produced under the conditions that count.
If it was execution
This is the more common answer and the better news, because execution moves faster than knowledge does.
- Set a stopping rule and obey it. Math gives you 22 questions in 35 minutes, so roughly 95 seconds each. If you hit 90 and have no clear path, mark it and move. Every question is worth one point, so trading three easy ones you never reach for one hard one you eventually crack is a bad deal.
- Sweep, then return. Answer everything that goes smoothly, flag whatever does not, and make sure your eyes land on every question before the clock tightens. Running out of time means you never saw items that were easy for you specifically.
- Re-read the question stem before you click. Solving correctly and then answering the wrong quantity is the most common avoidable loss on the entire test.
- Protect Module 1. How you do on the first module decides whether the second is the harder version, and only the harder path reaches the top of the range. A careless slip there costs more than one anywhere else. See how the first module sets your ceiling.
If time was your specific problem, running out of time on the SAT covers why it is usually the careful students who do it.
Practice with the clock set short. Thirty minutes for a math module instead of thirty-five. Then test day feels roomy instead of frantic.
If it was content
Work one skill at a time rather than taking mixed sets. Pick the weakest area, drill it in the College Board Question Bank until you can do it without looking anything up, then move to the next. Filter to the difficulty band where you are getting roughly 60 to 80 percent right, which for most students below 650 in a section means Easy and Medium.
Then, a few days later, mix the skills you have built into a single set. Practising one skill at a time teaches you to solve it; the real test also asks you to recognise which skill a question is, and that step is invisible while every question is the same type. Both stages are necessary and the order matters.
Math tends to respond faster than reading, which is worth knowing when you are dividing limited weeks. Improving your math score has the topic-by-topic version.
Do not spend the gap taking more full practice tests. Testing measures the problem; reviewing fixes it. One test you work through properly, reconstructing the real cause of every miss and every lucky guess, is worth more than three you sit and score. If you take a test and do not review it, you have spent three hours learning nothing. How to review a practice test is the method.
A shape for six weeks
Roughly, and adjust to whichever pile your points came from:
- Weeks 1 to 2. Fix the execution problems first, because they are cheap and they apply to everything after. Drill skill by skill on your weakest content area alongside them.
- Weeks 3 to 4. Mixed sets. Keep drilling new skills, but start combining what you have already built so recognition gets practised too.
- Week 5. One full test under real conditions, then spend as long reviewing it as you spent taking it.
- Week 6. Light. Re-do questions you missed a fortnight ago without your notes. If you can still solve them cold, you learned them. Sleep properly the last few nights.
Deciding on evidence rather than disappointment
Everything above turns on the first split, and a bare score report is not very good at telling you which pile your points fell into. A 1280 that leaked evenly and a 1280 that collapsed on timing in the back half of both sections are different problems wearing the same number.
Forge reads a full diagnostic across five dimensions of reasoning and separates the misses that are gaps from the misses that are slips, which is exactly the split this whole plan depends on. Whichever tool you use to find it, find it before you spend the six weeks.
Name the thing that will be different this time. If you cannot, the second score will look a lot like the first.