Listening is the module where preparation converts into marks fastest, because most lost marks are procedural rather than linguistic. The recording plays once, the answers arrive in order, and the questions on screen tell you what to listen for before a single word is spoken. The habits below are about using that preview time and never letting one missed answer cost you the next three.
Most Listening marks are lost after you have already heard the answer: to the detail the speaker withdrew a sentence later, to a plural typed as a singular, and to an article that pushed a correct answer over the word limit.
The format at a glance
Parts
4 parts, 10 questions each
Questions
40, one mark per question
Audio
About 30 minutes, played once
Checking time
2 minutes at the end on the computer-delivered test
Versions
Identical for Academic and General Training
Contexts
Parts 1-2 everyday, Parts 3-4 education and training
The traps, worked through
Four traps account for the bulk of the gap between what candidates hear and what they score. Each one is built deliberately, which means each one is predictable.
The correction you did not wait for
Part 1 speakers give a detail and then change it. The first version is said slowly and lands while you are still typing; the correction arrives inside a throwaway clause.
On screen
Form line — Collection day: __________
What you hear
“We could have it ready for you on the Tuesday — oh, hang on, the van is out that morning, so let us say Thursday instead.”
Most candidates write
Tuesday, typed the moment it was said.
Earns the mark
Thursday
Why: The mark goes to whatever survives to the end of the speaker's turn, and “let us say … instead” is the standard signal that the earlier detail has been withdrawn.
In Parts 2 and 3 every option gets mentioned. The one repeated in the recording's exact words is usually being conceded and then dropped; the correct option is the paraphrased one.
On screen
What does the speaker say has changed most at the centre? A the opening hours · B the membership fee · C the range of classes
What you hear
“Everyone asks about the opening hours, and yes, those did shift a little. But the real difference is how much more there is to actually do here — we run three times the timetable we used to.”
Most candidates write
A, because “opening hours” is the only phrase the candidate definitely heard.
Earns the mark
C
Why: “Three times the timetable we used to” is the paraphrase of range of classes, and the verbatim phrase sits next to a concession — “yes … but” is where distractors are planted.
Part 3 matching turns on who ends up with the job, not who first offers to do it, and the offer and the reassignment are usually two turns apart.
On screen
Match each task to A Priya, B Daniel or C both of them. Task: book the seminar room.
What you hear
Priya: “I could sort the room out, I suppose.” Daniel: “You have got all the survey data to write up though — leave the room to me, I am in the department office every morning anyway.”
Most candidates write
A, marked the instant Priya offered.
Earns the mark
B
Why: The first offer is routinely overturned, so the marked answer is the name attached to the settled arrangement rather than to the opening suggestion.
Two limits are counted mechanically: how many words the instruction allows, and which words are already printed around the gap. Break either and a correct word scores zero.
On screen
NO MORE THAN TWO WORDS AND/OR A NUMBER. Note — Members bring their own __________.
What you hear
“Everyone brings their own goggles — we only lend out the wetsuits.”
Most candidates write
their own goggles, a pair of goggles, or the singular goggle.
Earns the mark
goggles
Why: Repeating the printed words or adding an article breaks the two-word limit, and the singular contradicts the plural the speaker said — either one alone loses the mark.
Around thirty minutes of audio, forty questions, and two minutes at the end. Every one of those numbers is a decision you can make now instead of under pressure.
Each preview pause
Ten questions to prepare, so a few seconds each. Read the instruction line first, ring the word limit, then write the expected word class beside every gap: noun, plural, number, date. Predicting the shape of the answer beats reading the questions twice.
While a part is playing
Eyes on the question you are answering, finger on the next number. One question of thinking time is affordable; two is not, because the recording does not pause while you decide.
The second you know you missed one
Type something grammatical in the right word class and jump to the number the speakers have reached. A blank costs one mark; hunting for it costs every question that goes past while you look.
Between parts
Do not review the part you just finished — preview the next one. Review time exists at the end of the test; preview time never comes back.
The final two minutes
120 seconds across 40 boxes is three seconds each, so this is not the moment to re-think content. Three passes instead: fill every empty box, then check plurals and word counts, then check the spelling of days, months, names and street types.
Use every second of preview time
Before each part the recording gives you time to read the questions. That gap is the single highest-value moment in the module.
Read the instruction line first and note the word limit. NO MORE THAN TWO WORDS AND/OR A NUMBER means a three-word answer scores zero even if it is factually right.
Predict the word class for every gap: a noun, a number, a date, a plural, an adjective. Your ear then filters for the right kind of word instead of any word.
Underline the words either side of each gap. Those are the signposts the speaker will paraphrase just before the answer.
Read ahead into the next set of questions whenever a speaker is on a stretch you have already answered.
For map, plan and diagram labelling, find the starting point and the compass or entrance before the audio begins.
Track the recording, not your memory
Answers always come in the order of the questions. If you hear something matching question 14 while you are still on 12, you have missed two answers, so move on immediately.
Expect corrections. Speakers say a number or date, then change it: the second version is the answer. Listen for 'sorry', 'actually', 'no wait', 'in fact'.
Expect distractors. Two or three options get mentioned and only one survives, usually signalled by 'but', 'although', 'in the end', 'we decided instead'.
Never leave a blank. An unanswered question is guaranteed zero, and there is no negative marking for a wrong guess.
If you lose your place, jump forward to the next question you can still catch rather than trying to reconstruct the one you missed.
Spelling, numbers and word limits
This is where marks quietly disappear. The answer is graded on what you typed, not on what you heard.
Spelling must be correct. Either British or American spelling is accepted, but a misspelled word scores zero even when the meaning is obvious.
Singular and plural matter. If the sentence needs a plural, type the plural you heard.
Write numbers as digits: 15 is faster and safer than fifteen, and counts as one word.
A hyphenated word such as part-time counts as one word. A number plus a word, such as 20 minutes, counts as two.
Type only the words the gap is missing. If the sentence already reads 'in the ______' and the answer is 'morning', typing 'in the morning' is marked wrong.
Do not add articles, units or extra description that the gap does not require. Copy the recording's own words.
Standard number, date, currency and abbreviation formats are accepted, so 15 May, 15/5 and £20 are all safe. Spelling still has to be right.
Practise spelling the high-frequency Part 1 categories: days, months, common surnames, street types, accommodation words, payment words.
Part-by-part tactics
Part 1 is a transactional conversation, usually a form. Expect names spelled out letter by letter, phone numbers, prices, addresses and dates.
Part 2 is a monologue about an everyday topic and often carries a map or a matching task. Orientate yourself on the diagram before it starts.
Part 3 is a discussion between up to four people. Keep track of who holds which opinion, because matching and multiple-choice questions turn on exactly that.
Part 4 is an academic monologue with no break in the middle, so the concentration demand is highest. Follow the headings in the notes as a map of where the talk is going.
In the two minutes at the end, check spelling, plurals and word counts rather than trying to recall missed answers.
How to practise between now and test day
Do full 40-question sections rather than isolated questions, because the skill being tested is sustained attention.
After marking, sort every error into one of three buckets: did not hear it, heard it but mis-spelled it, heard it but broke the word limit. Each needs a different fix.
Re-listen with the transcript to find the exact moment the paraphrase happened. That trains recognition better than replaying the whole clip.
Use dictation on short clips to build typing accuracy under time pressure.
Listen to a range of accents. British, Australian, New Zealand, North American and Canadian speakers all appear.
Advice to ignore
Four pieces of standard Listening advice that cost more than they return.
“Type every answer in capitals so the marker cannot misread it.”
Case is not marked. On a keyboard, capitals buy you nothing and cost you shift keys and typing slips. Spend that attention on spelling, which is marked.
“Write the whole phrase, so the right words are definitely in there somewhere.”
An over-limit answer scores zero even when the correct word is sitting inside it. Type only what the gap is missing.
“Use the first preview to skim all forty questions.”
There is a preview before every part. Spending Part 1's pause on Part 4 means meeting question 1 cold — and Part 1 is the set most candidates can actually get right.
“Train on the fastest audio you can find so the real test feels slow.”
Speed is rarely the problem; paraphrase recognition is. Re-listening to a section you have already marked, transcript in hand, to find the exact second the wording changed, does more than any amount of double-speed practice.
Common mistakes and how to fix them
Reading question 1 while the introduction to Part 1 is still playing.
Fix: Read the whole set during the pause. The introduction carries no answers, so use it to preview instead.
Writing the first number heard when the speaker later corrects it.
Fix: Leave the answer until the sentence finishes. Corrections are deliberate and common.
Exceeding the word limit by including an article or a unit.
Fix: Re-read the instruction line and cut the answer to its core words before moving on.
Freezing after a missed answer and losing the next two as well.
Fix: Accept the loss, look at the next question number, and re-enter the recording there.
Typing a plural when the sentence needs a singular, or the reverse.
Fix: Check the grammar of the sentence around the gap during your final two minutes.
Choosing an option because you heard the exact word in the recording.
Fix: Correct answers are usually paraphrased. Word-for-word matches are often the distractor.
Question type by question type
The single mistake that costs most marks on each task. Each guide then covers what the task looks like on screen, how it is marked, four strategies and the two traps specific to it.
Is the Listening test different for Academic and General Training?
No. Both versions take the same Listening test, with the same four parts and the same 40 questions.
How much time do I get to transfer answers?
On the computer-delivered test you type answers directly and get two minutes at the end to check them. The ten-minute transfer time applies to the paper-based test only.
Does capitalisation matter?
Answers are accepted in either upper or lower case. Spelling is what has to be right.
Should I guess if I did not hear the answer?
Yes. There is no penalty for a wrong answer, so a plausible guess that fits the grammar and the word limit is always better than a blank.
Put these tips under exam conditions with a full timed section.
Drill Form Completion and Note Completion until you can type a corrected answer without pausing, then take a full section so all four parts run back to back.