Get a question wrong in the app and you are shown the explanation, the pattern the wrong answer was built on, and four small buttons.

Fell for the trap. Careless slip. Didn't know it. Ran out of time.

One tap, optional, and nothing happens on screen if you skip it. It is the smallest feature we have built and the one we argued about most, so it is worth explaining what it is for.

The explanation is the wrong place to start

Reviewing a wrong answer usually begins by reading why the right answer is right. It is the natural order, and we think it costs you the most useful information you had. What follows is our reasoning for the design, not a finding from the research literature.

Before you read the explanation you know something nobody else knows: what happened in your head. You were down to two and picked the wrong one. You misread the question. You knew it and slipped. You never really got to it.

Once you read the explanation, that knowledge is hard to recover, because the explanation supplies a coherent account of the right answer and it is easy to adopt as your own. Misses start to feel like things you did not know. And "didn't know it" points at revision, which is the most expensive response and often the wrong one.

The button exists to catch that information in the two seconds before the explanation overwrites it. That is the whole design.

Four buckets, because four different things need doing

The categories are not a taxonomy of question types. They are a taxonomy of what you should do next, which is why there are four rather than fifteen.

Fell for the trap points at the answer choices rather than the passage. If you understood the text and still lost, re-reading the text more slowly is not the fix. Naming the pattern is.

Careless slip points at conditions. Slips cluster where you are rushing or tired, and they usually say something about pacing rather than knowledge.

Didn't know it points at content. Learn the rule, re-test on it. Conventional revision is designed for this bucket and only this one.

Ran out of time points at pacing directly, and is the one people most often mislabel as not knowing the material.

Splitting those four apart matters because the responses are mutually exclusive. Two hours of revision is a good answer to "didn't know it" and a poor answer to the other three.

The bucket most likely to be mislabelled

Our own question bank suggests one of these is systematically over-reported, and it is worth knowing before you tag.

The largest single group of wrong answers in our SAT Math bank are choices that come from clean, correct work aimed at the wrong quantity. The arithmetic is right. The method is right. The number is a real result of the working, just not the one the question asked for.

Those feel careless when you look back at them, because the working is sound and the error is invisible in the steps. They are not careless. They are a target error, and the check that catches them sits before the first line of working rather than after the last.

So if your review comes back mostly careless slip on Math, look again at whether you solved for the wrong thing.

What one tap cannot do

The honest limits, because a feature like this invites more confidence than it earns.

It is self-reported. You are the only one who knows why you missed a question and you are also motivated to be generous with yourself. "Careless slip" is a more comfortable answer than "didn't know it", and we have no way to check.

It is optional and most people skip it, which means the totals in your review are a sample of the misses you felt like tagging rather than all of them.

And it measures your reasons, not your results. Nothing about tagging a miss makes you less likely to repeat it. The tag is only worth the two seconds if you then do something different, and the tag cannot make you.

We built it anyway, because a review session that ends with a shape ("mostly trap, some timing") points somewhere, and one that ends with a list of topics you got wrong usually points at redoing the same practice.

Where it fits

The pattern label on each wrong answer tells you what the test did. The button tells you what you did. Those are different questions and both are more useful than the score.

If you want the manual version, the four buckets work equally well on a paper test with a pen: sort your misses before you read any explanation, then spend your time on the largest pile. Read the explanations afterwards. College Board's own advice after a practice test is to review each question and its explanation and track the patterns in what you missed, which we agree with. Sorting first is our addition, and the order is the only thing we are arguing about.