GMAT error log: how to record mistakes that become points
Almost every candidate keeps some form of error log, and almost every error log is useless. The reason is always the same: it records what the right answer was instead of why the decision went wrong.
The fields that matter
- Source and question number, so you can redo it later.
- Time spent — above or below the section target.
- Classified root cause: content, interpretation, trap, calculation, timing or carelessness.
- The correct in-test decision: what you should have done at that moment, including 'guess and move on'.
- Date of the next review.
Classifying causes is what creates diagnosis
When every mistake carries a tagged cause, the log stops being a list and becomes a dashboard. After 60 or 80 entries it becomes obvious that, say, 40% of errors come from misreading the prompt — and the study plan shifts from content to active reading.
Spaced review: 1, 3, 7, 21 days
Redoing a question the next day proves nothing: you remember the answer. The useful cycle is revisiting at 1, 3, 7 and 21 days, redoing it from scratch and comparing your decision with the one you logged. If the process repeats correctly twice in a row, the item leaves the cycle.
Strategic guesses belong in the log too
Tagging a deliberate guess as a regular mistake distorts your statistics. Log it separately: a guess made at the right time, on a question already lost, is a correct decision — even when the outcome is wrong.
A good error log is short on text and rich in cause. It exists to change future decisions, not to archive answer keys.
Want the full method and the app?
The GMAT Descomplicado eBook (27 chapters, updated for the Focus Edition) comes with app access: error log, spaced-repetition flashcards and a pacing tracker.