How to Track Study Progress: Metrics That Actually Matter

Track study progress with five output metrics: revision coverage, accuracy on attempted questions, performance under exam timing, the number of overdue reviews, and counts by error category. Each of these changes what you do next week. Hours logged, pages read and videos watched do not, because they measure effort rather than what you can now produce reliably under pressure.

The distinction matters because input metrics almost always look healthier than output metrics. A week of ten solid hours can coexist with falling accuracy, and only one of those two numbers tells you something is wrong.

What should I measure to know if my study plan is working?

Measure what you can produce, not what you consumed. A plan is working when the share of syllabus you can retrieve is rising, your accuracy on fresh questions is rising, and your overdue backlog is not.

The practical test for any metric is whether a bad reading has an obvious response. "Twelve hours this week" has no response. "Accuracy fell from 74 to 61 percent on mixed questions" has several: check whether the mix changed, whether the topics were newer, or whether you were rushing. Metrics that generate a next action are worth collecting, and the rest are decoration.

Syllabus coverage versus revision coverage

Syllabus coverage is the proportion of topics you have studied at least once. Revision coverage is the proportion you have successfully recalled without notes since you studied them. They are usually reported as one number by people who are, in effect, reporting only the first.

The gap between them is the single most honest indicator in your whole system. Someone at 90 percent coverage and 35 percent revision coverage is much less prepared than someone at 60 and 55, even though the first number looks better and feels better. To collect revision coverage you need a per-topic record of the last recall attempt and its result, which is exactly what the spaced revision tracker column layout is for.

Count a topic as revised only if the most recent attempt was a genuine recall without notes and it succeeded. A topic you reread yesterday does not count.

What is attempted-versus-correct accuracy, and how do I calculate it?

Accuracy is the number of correct answers divided by the number you attempted, and it must always be recorded alongside the attempt count. Correct over attempted, with attempted over total stated next to it.

Reporting accuracy alone hides the most common failure in competitive papers. Consider two results on the same 100-question paper:

All three have identical scores by one measure and completely different next actions. A needs speed work and a skip rule. B needs to identify which categories the 19 losses fall into. C needs to stop attempting questions they cannot justify. Where negative marking applies, accuracy on attempted questions matters more than raw attempt count, so record both and never average them away.

How do I measure timed performance?

Timed performance is accuracy and attempt rate measured under the real paper's clock, and it is the only version of those numbers that predicts anything. Untimed practice measures whether you know the material; timed practice measures whether you can retrieve it fast enough.

Collect three figures per timed section: questions attempted, questions correct, and the number where you spent noticeably longer than your planned per-question budget. That third figure is where the time went, and it usually points at two or three topics rather than a general slowness. The speed and accuracy guide covers what to do once you have identified them.

Sit at least some practice at the same time of day as the real paper. Performance at 9am and 9pm is not interchangeable for most people.

Why does the overdue review count matter?

The overdue review count is the number of scheduled recall attempts whose date has passed without an attempt. It is a load metric: it tells you whether your plan fits your actual week, which no coverage number can.

Its value is in the direction, not the absolute figure. A steady ten is a working system with a small buffer. Six that becomes eighteen over a fortnight means you are scheduling more revision than your week absorbs, and the correct response is to schedule fewer reviews or merge related prompts, not to book a catch-up weekend that will not happen. Keeping that count visible is what a planner and calendar are for: in MyStudyPlanner the review dates sit on the calendar and the progress view keeps the outstanding count in front of you rather than in a file you have stopped opening.

If the backlog has already grown past what the remaining weeks can hold, the backlog recovery guide covers triage properly.

How do I use error-category counts?

Error-category counts are the number of lost marks grouped by cause rather than by topic. Four categories cover nearly everything, and each has a different repair.

The reason to group by cause is that topic-level counts mislead. Losing eight marks in algebra sounds like an algebra problem, but if six of them are timing errors, more algebra revision will not fix it. Record one row per question, not one row per test, using the columns in the mock-test error log. The mock-test analysis guide works through a full review.

Which metrics are vanity metrics?

Any number that goes up whether or not you learned anything. Hours logged, pages read, videos watched, chapters ticked, notes written and unbroken streaks all qualify.

They are not worthless, but their honest use is narrow. Hours are a capacity check: if you planned twelve and logged five, your schedule is built on time that does not exist, and that is useful for replanning. What hours cannot do is tell you whether the studying worked. Streaks are a motivation device, not evidence. Pages and videos measure exposure, and exposure is the input least correlated with what you can produce on a bad morning in an exam hall.

The failure mode is subtle: vanity metrics feel like progress precisely when real progress has stalled, because they are the easiest numbers to keep moving. If you want your goals stated in terms you can check, the measurable study goals guide covers how to phrase them.

How do I run a weekly review of my study metrics?

Book 20 to 30 minutes at the same time each week, ideally before you plan the following week rather than after.

1. Record the raw numbers first, without interpreting them: topics covered, topics successfully recalled, questions attempted and correct, overdue reviews, hours actually studied.
2. Compare revision coverage with syllabus coverage and note the gap. If the gap widened, your plan is front-loading new material over retention.
3. Check accuracy against attempt rate for any timed practice. Decide which of the three patterns in the table above you are in.
4. Total the error categories from the week's log and pick the single largest. That category, not the largest topic, sets next week's repair work.
5. Read the overdue count as a trend, comparing it with last week's. Rising means reduce the scheduled load now.
6. Compare planned hours with actual hours and rebuild next week on the actual figure, not the planned one.
7. Write down exactly one change for the coming week. One change is testable against next week's numbers; four are not.

Keep the record in the same place every week. A weekly sheet, a note or the progress view in MyStudyPlanner all work; what matters is that last week's figures are visible next to this week's so you are reading a trend rather than a snapshot.

What does a bad reading look like in practice?

Four patterns come up repeatedly, and each has a specific response.

None of these can be diagnosed from a single week, which is why the weekly record matters more than any individual number. Two consecutive weeks pointing the same way is a signal; one week is noise.

Frequently Asked Questions

What is the best way to measure study progress?

Measure output, not input. The five readings worth keeping are revision coverage, accuracy on questions you attempted, performance under exam timing, the number of overdue reviews, and counts by error category. Each one changes what you do next. Hours studied and pages read do not, because they describe effort rather than what you can now reliably produce.

Are study hours a useful metric?

Only as a capacity check. Hours tell you whether the time you planned actually existed, which is worth knowing when you rebuild a schedule. They tell you nothing about whether the studying worked. Two people can log identical hours with completely different retention, so never treat an hours total as evidence of progress on its own.

What is the difference between syllabus coverage and revision coverage?

Syllabus coverage is the share of topics you have studied at least once. Revision coverage is the share you have successfully recalled without notes since studying them. Coverage almost always outruns revision, and the gap between the two numbers is the most honest single indicator of how prepared you actually are.

How do I calculate accuracy on practice questions?

Divide the number correct by the number attempted, and record attempted separately from the total in the paper. An 80 percent accuracy on 40 of 100 questions is a very different result from 80 percent on 95, because the first hides an attempt-rate problem. Always log both numbers or the figure becomes uninterpretable.

How often should I review my study metrics?

Once a week, in a fixed 20 to 30 minute slot, plus a short check after every mock test. Daily review is noise, because single sessions vary too much to signal anything. Monthly is too slow to correct a drifting plan. A weekly cadence is frequent enough to act on and infrequent enough to show a trend.

What counts as a bad overdue review count?

Any count that is rising week on week, whatever its absolute value. A stable backlog of ten overdue reviews is manageable; six that became eighteen in a fortnight means your plan is scheduling more revision than your week can absorb. The fix is to reduce scheduled reviews or combine related prompts, not to promise a catch-up weekend.

Your next step

Pick two metrics only: revision coverage and error-category counts. Record them this week using the free CSV templates, then compare them with next week's. Two metrics tracked consistently beat eight tracked for a fortnight and abandoned.