United States
Hong KongHow to tell whether your HKSI Paper 1 preparation is working
Track four things: how many of the 66 syllabus headings you have covered with questions, your mock score trend, your accuracy per topic, and how often you repeat an error you have already reviewed. Hours studied tells you nothing about whether any of it is landing.
Hours studied is the metric everyone uses and the worst one available
It is easy to record, it goes up reliably, and it feels like evidence. It is also compatible with having learned almost nothing, because an hour spent highlighting the manual logs identically to an hour spent working through wrong answers on client money.
Track output instead. Not what you put in - what you can now do that you could not do a fortnight ago. Four numbers cover it, and none of them take more than a minute a week to record.
The four numbers
| Metric | How to measure it | What good looks like | When to worry |
|---|---|---|---|
| Headings covered | Tick off each of the 66 second-level syllabus headings once you have answered questions on it, not once you have read it | All 66 ticked by three weeks out | Fewer than half ticked at the halfway point of your plan |
| Mock score trend | Full 60-question timed papers only. Record the total and the date | Rising across three mocks, with the last one comfortably above 42 | Flat across two mocks despite study between them |
| Accuracy per topic | Percentage correct in each topic, from mocks and question sets | Above 70% on Topics 3, 4 and 5. Nothing below 50% | Any topic under 40% within a fortnight of the sitting |
| Repeat errors | Of the questions you get wrong, how many are on a point you have already reviewed once | Under a fifth of your errors | Half or more. Your review method is not working |
The fourth one is the one nobody tracks and it is the most diagnostic of the set. Getting a question wrong on unfamiliar material is progress. Getting it wrong on something you reviewed three weeks ago means the review did not stick, and that is a method problem no amount of extra hours will solve.
Headings, not chapters
Count against the syllabus rather than against whatever book you are using. HKSI publishes nine topics and 66 second-level headings in v3.5, and that list is the actual scope of the paper. A chapter you have read is not a heading you have covered.
The rule for ticking one off: you have attempted questions on it and you can state, from memory, what it requires and what happens when someone gets it wrong. If you cannot do that second part, it is not ticked. This is stricter than most people's instinct and it is why the count is useful.
Print the syllabus contents, put a box next to each second-level heading, and tick them as you go. It takes ten minutes to set up. Seeing 66 boxes at the start is also a far more honest picture of the workload than a page count is.
Reading a mock score properly
The total is the least informative part of a mock result and the part everyone fixates on. What matters is the topic breakdown and the direction of travel.
| Where you are | A score around this is normal | What it means |
|---|---|---|
| Halfway through your plan | 30 to 40 of 60 | On track. You have met the material but not consolidated it |
| Three weeks out | 40 to 45 | Fine. The remaining work is rotation and repair, not new content |
| One week out | Above 45 | Comfortable margin over the 42 you need |
| One week out | 42 to 45 | You would probably pass, with no room for a bad morning |
| One week out | Below 40 | Consider moving the sitting. HKSI runs the paper monthly and a resit costs HKD 1,800 |
Do not compare a score against another candidate's. Different question sets, different difficulty, no shared scale. Comparing which topics you each struggled with is useful; comparing totals is a way of feeling either falsely safe or needlessly awful.
And treat a single mock as noisy. A five-mark drop between two papers is usually the papers, not you. A consistent weakness in the same topic across two mocks is a real signal and should redirect your next fortnight.
The repeat-error log
Keep one line per error: the topic, the point you missed, and the date. When you get something wrong, check whether it is already on the list. If it is, mark it.
Two things come out of this. You get a ranked list of your actual weak points, which is worth more than any amount of intuition about which topics feel shaky. And you get an honest measure of whether your review method works - if a fifth or fewer of your errors are repeats, it is working; if half are, you are reviewing by re-reading rather than by retrieval, and that needs fixing before anything else.
What not to track
- Pages read. Measures the wrong thing entirely, and rewards the front-to-back manual read that is the most common way candidates waste six weeks.
- Cards in the deck. A growing deck is not progress. Cards retired after three correct recalls is a better number, and it goes up more slowly.
- Number of practice questions attempted. Volume without review is close to worthless. Two hundred questions with every error understood beats a thousand skimmed.
- Days of unbroken study. Streaks make people study when they are too tired to learn, and one missed day then feels like failure rather than a rest.
Am I ready?
The honest test is three-part and it takes no judgement to apply. All 66 headings ticked. Two consecutive timed mocks above 42, ideally above 45. No topic sitting below half marks.
Meet all three and you are ready, whatever your nerves say. Miss one and you have a specific thing to fix rather than a general anxiety, which is much easier to work with.
The concession is that these thresholds are ours, drawn from how candidates who pass tend to look rather than from anything HKSI publishes. What HKSI does publish is the pass mark, 70%, and the pass rate, which averaged 51.9% over the twelve months to June 2026. Everything else in this post is inference - useful inference, but you should know which is which.
Common questions
How do you know if you are ready for HKSI Paper 1?
Three tests: all 66 syllabus headings covered with questions, two consecutive timed mocks above 42 of 60, and no topic scoring below half marks. Meet all three and you are ready. These thresholds are ours, based on how passing candidates tend to look rather than on HKSI data.
What mock score should you be aiming for before the exam?
Above 45 of 60 in the final week gives real margin over the 42 needed to pass. Between 42 and 45 would probably pass but leaves nothing for a bad morning. Below 40 a week out is a reason to consider moving the sitting.
Should you track how many hours you have studied?
It is the least useful metric available. An hour highlighting the manual records the same as an hour working through wrong answers, and the two produce very different results. Track headings covered, mock trend, per-topic accuracy and repeat errors instead.
Is it normal for mock scores to go down?
A drop of a few marks between two papers is usually the question set rather than you. Look at the topic breakdown. A weakness appearing in the same topic across two consecutive mocks is a real signal; a lower total on its own generally is not.
What does a high repeat-error rate mean?
That your review method is not working. Getting a question wrong on new material is progress; getting it wrong on a point you already reviewed means the review was re-reading rather than retrieval. Fix the method before adding hours, or the hours will produce the same result.