Listening is the section most candidates neglect and the one most likely to end an attempt, because it carries its own sectional minimum: sectional minimums.
It is also the slowest skill to move, which means it needs to start earliest.
Why it feels impossible
Three separate problems, usually diagnosed as one:
- You do not know the words. A vocabulary problem wearing a listening costume.
- You know the words but cannot segment the stream. You would recognise them written down but cannot hear where one ends and the next begins.
- You can segment but not fast enough to keep up while also thinking about the question.
The training is different for each, and applying the wrong one wastes months.
Diagnose with a transcript
Take a clip you failed to understand. Read the transcript.
- Understood it immediately on paper? Problem 2 or 3 - segmentation or speed.
- Still did not understand it on paper? Problem 1 - vocabulary and grammar. Listening practice will not fix this.
Do this once a week. It stops you doing listening drills when the real gap is words.
The four methods, in order
1. Intensive listening. Short clip, two plays, write down what you heard, compare with the transcript. Twenty minutes, three times a week. This is the one that builds segmentation, and it is uncomfortable, which is why most people skip it.
2. Shadowing. Speak along with the audio, a beat behind, matching rhythm rather than meaning. Fixes segmentation and pronunciation together: the shadowing method.
3. Extensive listening. Large volumes of comprehensible material - podcasts for learners, simple news, anything you mostly follow. Builds speed and endurance. This is the background work: an hour a day while cooking or commuting.
4. Exam-format practice. Past-paper listening sections under timing, with the note-taking system you will use on the day: taking notes in the listening section.
Add them in that order. Starting with method 4 alone is why some candidates plateau: exam practice measures the skill, it does not build it.
Never replay more than twice
The exam plays each clip once. A study routine that replays five times trains a skill you cannot use.
Two plays, then read the transcript, then move on. This is also exactly how our free exam demo works for JFT-Basic: the audio starts on its own, plays twice, and there is no replay button.
Numbers, times and negatives
Three specific things cause most listening errors at every level, and all three respond to targeted drilling:
- Numbers, especially the sound changes - さんぜん, はっぴゃく, ろっぴゃく. Numbers and time.
- Times, including ～じはん, ～ふん / ～ぷん, and までに.
- Negatives, which flip an answer with one syllable: とまります against とまりません.
Ten minutes a week on these three alone shows up in scores faster than any other listening work.