United States
Hong KongSpaced repetition for HKSI Paper 1: about 100 cards, not a thousand
Build about 100 cards, not a thousand. Cards belong to facts with no structure to reason from: periods, deadlines, thresholds, licence types, code names. Review at 1, 3, 7 and 14 days, 15 minutes daily. Scenario reasoning does not fit on a card and should stay in practice questions.
What spaced repetition is actually for
It solves one problem: things you learn in week two are largely gone by week eight unless you meet them again on a schedule. That is not a discipline failure. It is how memory works, and it is the single biggest structural risk in any Paper 1 plan longer than about a month.
What it does not solve is reasoning. Paper 1 mostly asks you to read a scenario and pick which of four consequences follows, and no flashcard trains that. Cards handle the raw material the reasoning runs on. Practice questions handle the reasoning. Confusing the two is how people end up with a beautiful deck and a mock score in the thirties.
What deserves a card
The test is simple. If you could work the answer out from first principles given thirty seconds, it does not need a card. If it is arbitrary, it does.
| Make a card | Do not make a card |
|---|---|
| Record-keeping periods, notification deadlines, contract note timing | "What is the purpose of the Code of Conduct?" - reason it out |
| The regulated activity types and what each one covers | Whole paragraphs from the study manual |
| Which Part of the Ordinance governs licensing, conduct, misconduct, compensation | Section numbers. Paper 1 does not test citation |
| The names of the SFC codes and who each one binds | "Explain the SFC's supervisory powers" - too big for a card |
| Capital and threshold figures that appear in the syllabus | Anything you have already answered correctly three times |
| The pairs: client money against client securities, corporation against institution | Scenarios. Those belong in a question bank |
Run that filter honestly across all nine topics and you land somewhere around 100 cards. Perhaps a few more if Topic 4 is new to you. If your deck passes three hundred, you have started carding things you understand, and the daily review will become long enough that you stop doing it.
Abandoned decks are almost always abandoned because they got too big, not because the method failed. Fifteen minutes a day is sustainable for eight weeks. Forty is not, and the forty-minute deck usually gets dropped in week three, taking the review layer with it.
Card design that survives contact with the exam
- One fact per card. A card asking for three things is three cards pretending to be economical.
- Ask in the direction the exam asks. The paper gives you a scenario and wants the rule, so the front should be the situation and the back the rule - not the other way round.
- Write the back in your own words. A card copied verbatim from the manual tests whether you can recognise the manual's phrasing, which is not what happens on the day.
- Put the confusable pair on one card. Client money on one side of the answer, client securities on the other. Separate cards teach both and distinguish neither.
- Delete aggressively. Three correct recalls in a row and the card leaves the deck. Keeping it is sentimentality, not study.
The intervals
You do not need software to do this properly, though software makes it easier. The schedule below is simple enough to run on paper with four envelopes or four sections of a notebook.
| Review | When | What happens |
|---|---|---|
| First | Same day you make the card | Confirms you understood it, rather than copied it |
| Second | 1 day later | Catches the ones that never really landed |
| Third | 3 days later | Most failures happen here. That is the card doing its job |
| Fourth | 7 days later | Anything surviving this is reasonably secure |
| Fifth | 14 days later | Passed three times in a row? Retire the card |
| Failed at any stage | Back to 1 day | No shame in it. A card you keep failing is a heading you should reread |
Fifteen minutes a day, ideally in a slot you would otherwise waste. The morning commute is close to perfect for this - it is short, it repeats, and recall is the one study activity that works on a crowded train.
Fitting it to your plan
Start building cards in your first week, not your last. A deck assembled in the final fortnight has no time to space anything and is really just a cramming list with better branding.
On an eight-week plan the deck grows for six weeks and then stops. On twelve weeks it matters more, because the gap between meeting Topic 3 and sitting the paper is long enough for real decay. On a two-week crash plan, skip cards almost entirely - there is no interval to space across, and the hours are better spent on questions.
The opinion, and where it stops
Spaced repetition is the second most valuable thing you can do for this paper. Practice questions are first, by a distance, and it is not close. If you have four hours and must choose, spend three on questions and one on cards.
The reason is what Paper 1 rewards. Sixty questions, 90 minutes, four plausible options each, and the wrong options are written to look like the right one. That skill - separating the nearly-true from the true under time pressure - comes from doing it repeatedly and reviewing your errors. Cards give you the vocabulary to reason with. They do not give you the reasoning.
The concession: for the numbers, cards win outright. Periods, deadlines, thresholds. There is nothing to reason about, no structure to fall back on, and drilling them is the only method that works. Put every one of them on a single sheet as well as in the deck, and review that sheet three times a week from the halfway point of your plan.
Common questions
How many flashcards do you need for HKSI Paper 1?
Around 100. Cards belong to arbitrary facts - periods, deadlines, thresholds, licence types, code names - not to anything you could reason out. A deck past three hundred usually contains material you already know, and it grows long enough that people stop reviewing it.
What review intervals work for this exam?
Same day, then 1 day, 3 days, 7 days and 14 days, with a failed card going back to a 1-day interval. Retire anything you recall correctly three times in a row. Fifteen minutes daily is enough to run about 100 cards across an eight-week plan.
Should you use an app or paper cards?
Either. Software handles the scheduling for you, which matters once a deck passes about a hundred cards, but four envelopes and a notebook work perfectly well. The method is what does the work, not the tool, and the tool people actually open is the right one.
Do flashcards help with scenario questions?
Only indirectly. Paper 1 mostly asks you to read a situation and choose the consequence, and that skill comes from timed practice questions and reviewing your errors. Cards supply the facts the reasoning uses. They cannot substitute for the reasoning itself.
Is spaced repetition worth it on a short plan?
Not on two weeks - there is no interval to space across, and the hours pay better on questions. On four weeks, keep a small numbers-only deck. On eight or twelve weeks it becomes genuinely important, because that is long enough for early topics to decay before the exam.