There is a common way to use a mock test. You sit it, look at the score, feel pleased or worried, and sit another one. It feels like study. It is mostly measurement.
A thermometer does not lower a fever. The score tells you that something is wrong; only the review tells you what. Done properly, reviewing one test is worth more than sitting three.
Here is a method that works for both the JLPT and JFT-Basic.
First, sit it honestly
A review can only be as good as the data. That means:
- Timed. Pace is one of the things being tested.
- One sitting, no pauses.
- No dictionary, no notes, no looking back at a listening script.
- Answer everything. Neither exam has negative marking, so a blank is a wasted chance.
One extra habit makes the review far better: mark every question you guessed. A dot on scratch paper is enough. You will need it in a moment.
Read the sections before the total
The total is the least informative number on the results screen.
Look at the per-section breakdown first. The same total can mean very different things: three strong sections and a collapsed fourth, or four mediocre ones. Those need different plans.
On the JLPT this is more than advice. Every level has a sectional minimum — 19 out of 60 in each scoring section, or 38 out of 120 in the combined section at N4 and N5 — and a candidate who clears the overall pass mark still fails if one section falls under it. See JLPT scoring explained.
On JFT-Basic, the threshold of 200 out of 250 is high enough that a badly weak section leaves you very little room to make it up elsewhere.
Either way: find the weakest section, and start there.
Sort every mistake into one of four types
This is the core of the method. Go through each wrong answer — and each guess — and decide which of these it was.
Type 1: You did not know it
A word you had never seen, a kanji you could not read, a grammar pattern you have not studied.
The fix is study. Add the item to your deck or your notes. This is the simplest kind of mistake and the least worrying; it shrinks on its own as you cover more material.
Type 2: You half-knew it
You recognised everything and still chose wrong. You confused 持 with 待, used おつかれさまでした where お先に失礼します belonged, or picked 〜ために where the answer was 〜ように.
This is the most valuable type to find, because exams are built from exactly these confusions — the wrong options are chosen to attract people who half-know the material. The fix is not more exposure. It is contrast: write the two items side by side and state the difference in one sentence.
Type 3: You misread it
You knew the Japanese and missed a condition. You skipped the ただし clause, read the weekday table on a Sunday question, answered what happens before when the question asked about after.
The fix is a habit, not knowledge: read the question first, read to the end, check the condition explicitly. If most of your reading and listening errors are this type, more vocabulary will not help you at all.
Type 4: You ran out of time
You rushed the last questions, or never reached them.
The fix is pacing practice: timed drills, and a rule for letting go of a question that has taken too long. It is a separate skill from knowing Japanese, and it responds quickly to practice.
And the guesses
Now return to the questions you marked as guesses and got right. Treat each one as wrong and sort it the same way. A lucky guess is a mistake that has not happened yet.
Keep an error log
One line per mistake. Four columns are enough:
| Question | Type | What happened | Rule to remember |
|---|---|---|---|
| Reading, sale notice | 3 | Took the 20% headline, missed ただし | Read to the end before answering |
| Conversation, leaving work | 2 | Swapped the two phrases | お先に失礼します = the one leaving |
| Script, にもつを もつ | 2 | Chose 待 | 持 has the hand radical |
| Listening, meeting time | 3 | Heard 30分 but added it | 早く = earlier = subtract |
The last column is the point. Write the pattern, not the question. “Question 13 was B” will never help you again. “A ※ note overrides the numbered steps” will help on every paper you ever sit.
After two or three tests, read down the Type column. It usually tells a clear story:
- Mostly type 1 — you need more coverage. Keep studying; you may be sitting the test early.
- Mostly type 2 — you have covered the material but not separated it. Work in contrasts.
- Mostly type 3 — your Japanese is ahead of your exam technique. Slow down by ten per cent and read to the end.
- Mostly type 4 — practise at speed.
Read the explanations for the ones you got right
It takes a few extra minutes and catches something nothing else does: being right for the wrong reason.
If the explanation says the answer depends on a ※ note and you chose it because it “sounded right”, you did not know it — you were fortunate. Every explanation on our mock tests says why the correct option is correct and why the tempting wrong one is wrong, for that reason.
Decide what to do next
A review is finished when it has changed your plan for the next two weeks. Before you close the page, write down:
- The weakest section — it gets the largest share of your study time.
- The dominant mistake type — it tells you how to study, not just what.
- Three specific patterns from the error log to fix first.
Then study those. Not another mock test.
When to retake
Leave it at least two weeks. Sooner than that and you are measuring your memory of the answers.
When you do come back, change the task. Do not just pick the right option — say why each of the other three is wrong. If you can do that for every question, you have finished with that test. If you cannot, the question still has something to teach you.
Meanwhile, use a set you have not seen for your next honest measurement. There are four JFT-Basic sets for exactly that reason:
And one for each JLPT level on the mock tests page.
What a mock score can and cannot tell you
It can tell you which section is weakest, what kind of mistakes you make, and whether you are improving from one test to the next.
It cannot tell you your real score. The JLPT’s scaled scoring cannot be reproduced from raw marks, JFT-Basic publishes no conversion from correct answers to points, and a 20-question set is a sample, not a simulation. Treat the percentage as a direction.
What it can do is tell you when you are clearly ready: when you are comfortably above the bar, on sets you have never seen, in every section, more than once. A single narrow pass is not that.
Next steps
- Take a free mock test — and mark your guesses
- JFT-Basic study plan — eight weeks, with a mock test at each stage
- Improve your listening — the section review alone cannot fix