Most study apps and school platforms show some version of a progress bar — a percentage of topics “covered,” a completion ring, a checklist filling in green as you go. It feels good to watch it climb. The problem is that it's measuring the wrong thing. Completion percentage tells you what you've looked at. It doesn't tell you what you'd actually get right on an exam paper tomorrow, and those two numbers can be a long way apart.
Why “Completion” Isn't Readiness
A 100% completion mark measures exposure, not readiness. It usually just means you opened the material, watched the video, or clicked through the notes once — it says nothing about whether you can still produce that content a week later without looking at it, apply it to a question phrased differently from how you first learned it, or do it under the time pressure of a real exam. Exposure and retention are genuinely different things, and a tracker built purely around “have you seen this” is measuring exposure while quietly implying it's measuring readiness.
This isn't a minor technicality — it actively misleads students at the point where it matters most. A student sitting at 95% “complete” across their subjects two weeks out from exams can reasonably feel like the hard part is done, when in reality they might be able to recall almost none of it without their notes open. An inflated sense of readiness right before an exam is arguably worse than no tracker at all, because it actively discourages the kind of last-stretch review that would have actually helped.
What Genuine Readiness Actually Depends On
A meaningful readiness measure has to be built from evidence of actual performance, not just exposure. A few things it should reasonably account for:
- Retention over time — not just whether you got a concept right the day you learned it, but whether you can still produce it after the kind of gap an exam schedule actually involves.
- Application, not just recognition — being able to answer a genuinely new question that uses a concept, not just recognise the concept when it's described to you directly.
- Performance under time pressure — how you actually do on timed practice, not just untimed review, since exam conditions are a real factor in performance, not a minor detail.
- Consistency, not a single good result — one strong quiz score can be luck or a familiar question; a genuine readiness signal should reflect a pattern across several attempts.
None of that can be inferred from whether a topic page was opened. It has to come from actual performance data — homework answers, quiz results, flashcard recall, and how much focused time has genuinely gone into a topic — which is a fundamentally different (and harder to build) kind of signal than a checklist.
Why a Conservative Score Is More Useful Than an Inflated One
A conservative, honest readiness score is more useful than an inflated one, even though inflated scores feel more motivating in the moment. There's a real temptation, from a product perspective, to make readiness scores look better than they are — students like seeing high numbers — but an inflated score is actively harmful precisely because of when it gets used: in the final weeks before an exam, when the entire point of the number is to tell a student where to actually spend their limited remaining time. A score that says “you're fine” when you're not doesn't just fail to help — it actively directs effort away from the topics that need it.
A conservative, honest score is more useful even when it's less flattering, because it does the one thing a readiness score actually exists to do: tell you the truth about where you stand while there's still time to act on it. A score that starts low on a topic with little evidence behind it, and only climbs as real performance data accumulates, is doing its job correctly — even if it's less satisfying to look at early on.
How to Use a Readiness Score Well
A readiness score is most useful as a prioritisation tool, not a verdict. What matters is what the number is telling you to do next — a low score on a topic is a signal to spend more time there, not a judgment on your ability. Treating a low score as discouraging rather than informative defeats the purpose of tracking it in the first place.
It's also worth checking a score against your own sense of how a topic actually went, rather than trusting it blindly. If a score seems out of step with how a recent test or practice session felt, that's worth paying attention to — it might mean there isn't enough performance data yet for the score to be confident, or it might mean your own sense of how well you understood something doesn't match how you actually performed under test conditions, which is itself useful to know.
Finally, a score is only as current as the activity behind it. A readiness number that hasn't moved in three weeks because you haven't done any homework, flashcards, or quizzes on that topic isn't wrong — it's just increasingly out of date. Checking a score without feeding it any recent evidence and assuming it still reflects where you stand is its own way of fooling yourself, just with real data instead of a completion percentage.
How Clarby's Prediction Works
Clarby's exam readiness prediction is deliberately built from actual performance signals rather than exposure. It combines real homework performance, flashcard confidence ratings (whether you genuinely knew an answer or were guessing), quiz scores, and time actually spent studying a topic — not just whether a topic was opened or a lesson was marked as watched. Because it's built this way, it starts conservatively on topics with little evidence behind them, and only becomes confident as real performance data builds up, which is the opposite of how a simple completion tracker behaves.
That also means it isn't static — it moves as your actual performance does, which is the point. A score is only useful if it reflects where you genuinely stand right now, not where you stood when you first opened the topic. You can see it applied to your own subjects by setting up Clarby with your current topics, and it starts working from your first homework session, flashcard set, or quiz rather than requiring weeks of data before it says anything useful.
Whatever tool you use to track your progress, the underlying question worth asking of any readiness number is simple: is this built from what I've actually demonstrated I can do, or just from what I've looked at? Those are very different numbers, and only one of them is worth trusting the week before an exam.