Doing 20 past papers will not raise your grade. This is what does.

6 min read · updated 10 October 2026

The most common revision mistake

A student tells you they have done twenty past papers and their mock came back the same grade as before. They are not lying and they are not lazy. They have spent twenty times two hours — forty hours — and moved nowhere.

Here is what usually happened: they did the paper, checked a few answers, felt reasonably good about it, and moved on to the next one. The paper was treated as a test. A past paper is not a test. It is a diagnosis.

The loop that actually works

Doing the paper is the smallest part of the work. The sequence is:

• Sit it properly. Timed, no notes, no phone, in one go. A paper done in pieces over an evening tells you nothing about whether you can do it under pressure, which is the thing being tested.
• Mark it against the official scheme. Not generously. Mark schemes award points for specific things, and if you did not write the thing, you did not get the mark. This part is uncomfortable, which is why it gets skipped.
• Write down every mark you lost, and why. Not the question number — the reason. "Did not know the definition." "Knew it, ran out of time." "Misread the command word." Three or four words each.
• Look at the list. After three papers it stops being a list of mistakes and becomes a pattern. That pattern is your revision plan, and it is specific to you.

The fourth step is where grades move. Everything before it is data collection.

The three kinds of lost mark

Nearly every mark you drop falls into one of three groups, and they need completely different responses.

• You did not know it. The only fix is to learn the topic. Go back to your notes or the textbook — doing another past paper will not teach it to you.
• You knew it but answered the wrong question. This is a command word problem: "describe" asked for and "explain" given, or the other way round. Enormously common, and fixable in an afternoon by learning what each command word demands.
• You ran out of time. A pacing problem, not a knowledge problem. The fix is practising with a clock, not learning more content.

Students overwhelmingly respond to all three by revising more content. For two of the three, that is the wrong medicine.

How many papers is enough?

Fewer than most people think, done properly. Five papers that you sat under timed conditions, marked honestly and logged carefully will beat twenty papers skimmed through.

If you have time for more, the right use of it is to redo the papers you did badly on — a month later, having learned the things on your list. Redoing a paper you scored poorly on, and scoring well, is the clearest evidence you have actually improved. A fresh paper only tells you whether your overall level has changed.

When to stop

In the last few days before the exam, stop starting new papers. You will not learn a new topic in that time, and a bad score the day before an exam damages your confidence for no useful return.

Instead: reread your list of lost marks, look over the questions you got wrong, and sleep properly. The list is the most valuable revision document you own by then — it is a record of exactly how you, personally, lose marks.