United States
Hong KongChoosing an HKSI Paper 1 question bank: what to check before you pay
Check nine things: syllabus version, whether questions are original, whether explanations cite a source, whether every wrong option has a stated reason, whether mock items are kept separate from practice, language, dating, disclosed review status, and access period. Question count is the least useful number on the page.
Question banks for this paper are sold on one number, and it is the wrong number. Two thousand questions sounds like more preparation than five hundred, and it usually means the same material recycled with the wording changed, no citations, and wrong options nobody would pick.
Here is what to check instead. Nine things, all of them visible from a free sample or a product page if the provider is being straight with you.
The nine checks
| Check | What to ask | Why it matters |
|---|---|---|
| Syllabus version | Which version is this written against, and from what date? | The current Paper 1 syllabus is v3.5. Material written to an older version will teach superseded detail. |
| Original questions | Did you write these, or are they reconstructed past papers? | HKSI releases no past papers. Anything described as real questions is somebody's reconstruction, with somebody's errors. |
| Cited explanations | Does each explanation say where the answer comes from? | An uncited explanation cannot be checked. On a licensing exam that is the difference between studying and taking someone's word for it. |
| Reason for every wrong option | Does the explanation say why A, B and D are wrong, individually? | This is the single strongest quality signal. Most banks tell you the answer is C and stop. |
| Practice and mock separated | Are mock papers built from questions I will already have seen? | A mock made of reviewed items measures recognition. You need at least one clean form for a real reading. |
| Language | Is the material available in the language I will sit in? | Paper 1 is offered in English and Traditional Chinese, and the regulatory vocabulary differs enough that switching late costs time. |
| Dating | When was this last updated, and does it say so on the page? | Subsidiary legislation moves. Undated material is a warning about the whole operation, not just that page. |
| Disclosed review status | Has a qualified person reviewed this, and does the provider say either way? | Providers who say nothing are not thereby better reviewed. Ask, and treat a clear no as more trustworthy than silence. |
| Access period and resit terms | How long do I keep access, and what happens if I fail? | A resit costs HKD 1,800, so the terms attached to a failed attempt are worth reading before you buy. |
How to test a bank in ten minutes
Three probes, using whatever free sample is on offer.
- Read three wrong options. Would a candidate who had studied genuinely consider them? If you can cut two options on every sample question without reading the stem carefully, the bank is flattering you.
- Find a scenario question. If every sample is "which body regulates X", the bank is built on naming questions, which are the easiest to write and the least like the hard part of the paper.
- Check one explanation against the source. Pick a fact, open the Ordinance or the SFC codes, and verify it. If the explanation cites nothing, you have learned something about the whole product.
That third probe takes five minutes and almost nobody does it. It is the most informative five minutes you will spend before buying anything.
Why question count is a bad metric
The syllabus has 66 second-level headings. A bank that covers every heading with two or three well-built items, each with a stated reason for every wrong option, is around two hundred questions. Beyond that you are getting repetition, and repetition of items you have already reviewed does very little.
What actually varies between banks is not volume. It is whether the wrong options were designed or generated, and whether the explanation teaches you a habit or hands you a fact. Three good questions on Part X will do more for you than thirty weak ones.
Open any sample explanation. If it explains only the correct answer, the bank has no diagnosis layer, and you will learn facts one at a time with nothing transferring between them. This is the most common gap in the category and it is visible in ten seconds.
What we do, and where we fall short
- Questions
- 131–66 in practice modules, 65 held for mocks
- Mock forms
- 5, built only from the mock-only items
- Explanations
- Cited to the Part, rule or code paragraph
- Wrong options
- Every one carries a named misconception
- Syllabus
- Written against v3.5
- Section numbers
- Not used. We cite by Part and by name
- Independent legal review
- Not yet completed. We say so on every page
- Price and access
- HKD 680 for 6 months, or HKD 1180 bundled
The row about legal review is there because you should apply the ninth check to us as well. Our explanations are written against the Ordinance and the SFC codes and have been cross-checked internally, but That is a genuine limitation and we would rather you knew it than found out later.
The second limitation is calibration. We cannot show you that our questions match HKSI's difficulty, because HKSI publishes no items and no statistics. Nobody outside the Institute can. Any provider claiming a verified match to the real paper is claiming access that does not exist, and that claim should make you trust the rest of their material less rather than more.
Free material versus paid
A short diagnostic and a full free mock will tell you where you stand and whether the questions are any good. That is often enough to decide. What free material rarely includes is the explanation layer, because explanations are where almost all the writing cost sits: a good one takes longer to produce than the question it belongs to.
So the honest framing of what you are buying is not questions. It is the reasons. If a paid bank does not give you materially better explanations than its free sample, there is nothing to buy.
One opinion, stated plainly
The best predictor of whether a question bank will raise your score is whether its wrong options were written on purpose. Everything else, the count, the interface, the analytics dashboard, the adaptive algorithm, is downstream of that. A bank with designed distractors and a plain list view will beat a bank with generated distractors and a beautiful one, every time, because the wrong options are what teach you.
The concession: interfaces do matter at the margin, and a tool you dislike using is a tool you will not open. If two banks pass the nine checks, pick the one you will actually sit down with. Just do the checks first.
Common questions
How many practice questions do I need for HKSI Paper 1?
Enough to cover all 66 second-level syllabus headings two or three times, which is around two hundred well-built items. Beyond that, additional volume adds little compared with reviewing what you have already answered properly.
Are there official HKSI Paper 1 past papers to buy?
No. The HKSI Institute does not release past papers or retired items. Material sold as real past questions is a third-party reconstruction, and its wording, difficulty and legal accuracy all depend on whoever assembled it.
What should an explanation in a question bank contain?
Why the correct answer is correct, a source you can check, and a separate reason for each wrong option. The last part is what turns a fact into a transferable habit, and it is missing from most banks in this category.
Is a free HKSI Paper 1 mock good enough?
For measuring where you stand, often yes. Free mocks usually give a score and a topic breakdown without full explanations, and the explanations are the expensive part. Use free material to test question quality before paying for anything.
How much does HKSI Paper 1 preparation cost?
The examination fee itself is HKD 1,800, and a resit costs the same again. Study material varies widely, so weigh any course price against the cost of a second attempt rather than against zero.
Does a question bank need to match the current syllabus version?
Yes. Paper 1 follows syllabus v3.5. Material written against an earlier version can teach superseded detail, and any provider who cannot tell you which version they wrote against has answered the question anyway.