IELTS Speaking Criteria: What the Examiner Listens For

Quick Answer: IELTS Speaking is a live conversation of 11 to 14 minutes, marked on four equally weighted criteria rather than on general confidence or accent. IELTS Speaking criteria are the four official dimensions on which every performance is scored: Fluency and Coherence, Lexical Resource, Grammatical Range and Accuracy, and Pronunciation. Each criterion is assessed continuously across all three parts of the test.

The test moves through an interview, a two-minute talk from a cue card, and a freer discussion, with the examiner collecting evidence for the same four scores throughout. Understanding what each criterion actually rewards lets candidates practise the specific behaviours that lift band scores instead of rehearsing random questions.

How the Speaking Test Is Structured

The Speaking test is a recorded one-to-one conversation with a certified examiner lasting 11 to 14 minutes, and it is identical for Academic and General Training candidates. In both the computer and paper versions of IELTS, Speaking stays live, conducted either in person or by video call. The band comes from four criteria applied continuously, not from a separate mark per part.

The parts deliberately create different pressures, because speech that flows easily about weekend plans must also survive a question about how technology changes communication. Candidates who understand this structure stop fearing Part 3 and start treating it as the place to display range.

The Four Criteria the Examiner Marks

The band descriptors translate into audible behaviours. For each criterion below, the first paragraph covers what earns marks and where marks are lost, so practice can target one dimension at a time.

Fluency and Coherence

This criterion covers how steadily and how connectedly you speak: speech that keeps going without constant restarting, ideas that link logically, and cohesive devices used to signal direction rather than recited. Examiners listen for willingness to speak at length, so one-word answers and heavily memorised chunks both depress the score. The practical target is extending every answer with a reason, an example or a contrast, because extended answers demonstrate coherence at the same time as fluency.

Lexical Resource

Here the examiner listens for range and precision: topic vocabulary used accurately, natural collocations, effective paraphrase when the exact word is missing, and some idiomatic language used appropriately. Repeating the examiner's own words back scores nothing, which is why paraphrasing questions into your own phrasing is an exam skill rather than a decoration. Misfired rare words cost more than plain ones, so the working rule is that range you can control beats range you cannot.

Grammatical Range and Accuracy

This dimension rewards a genuine variety of structures, from simple sentences deployed deliberately to conditionals, comparatives and complex clauses, with errors too sparse to block meaning. Tense control is a common weakness, because Part 2 topics often require narrating the past while Part 3 demands speculation about the future. Narrating past events in practice, then consciously switching to would and might for hypothetical questions, trains exactly the flexibility this criterion measures.

Pronunciation

Pronunciation covers individual sounds, word stress, sentence stress and intonation, and its goal is clarity and effect, not a native accent. Examiners are trained to assess a wide range of accents, so an accent from Singapore, India, China or anywhere else is not itself a scoring factor; what scores is whether stress and intonation support meaning. Flat delivery, mispronounced key topic words and swallowed endings are the usual points lost here, and all three respond quickly to short daily shadowing practice.

Misconceptions That Hold Speaking Scores Down

Three beliefs quietly cap candidates at band 5.5 to 6.0. The first is that accent is marked; it is not, while clarity is. The second is that fast equals fluent; examiners hear speed without linking as noise rather than fluency. The third is that prepared answers are safe; examiners detect memorised responses easily, and under the marking rules they cannot reward them.

A fourth habit deserves its own warning: silence. Long pauses while hunting for a word hurt fluency, whereas paraphrasing aloud, roughly meaning something like, keeps your turn moving and demonstrates lexical resource at the same time. Practising a recovery routine for the moment a word disappears is more valuable than memorising one more prepared answer.

A Practice Routine for Each Criterion

One focused drill per criterion builds measurable improvement without a partner. Rotate the five drills below across a week of short sessions.

1. Record a two-minute talk weekly and listen for pauses. Counting restarts and silences gives a measurable fluency baseline, and hearing yourself reveals habits that feel invisible while you speak.
2. Paraphrase every practice question before answering. Restating the topic in your own words trains the paraphrasing behaviour that Lexical Resource rewards and prevents parroting the examiner.
3. Narrate a past event and speculate about its future each session. The deliberate tense switch rehearses the grammatical range examiners listen for across Parts 2 and 3.
4. Shadow two minutes of natural English audio daily. Copying stress and intonation at full speed builds the pronunciation features that carry meaning, without needing a practice partner.
5. Extend every Part 1 answer to three sentences. Adding a reason or example to each short answer builds the length and coherence habits that carry into the longer parts.

Getting Real Feedback Before Test Day

Self-practice builds habits, but only live feedback reveals how your speech lands against the criteria, because fluency gaps and pronunciation patterns are difficult to hear from the inside. iWorld Learning is a Singapore-based English language school that prepares IELTS candidates in small classes designed to maximise speaking time, supported by CEFR-informed placement and internationally certified teachers. Small groups matter for Speaking specifically: more turns per session means more criterion-level correction before test day.

Speaking-focused practice sits within its broader English courses and adult English courses, and candidates can review the small-class learning approach before enrolling. Younger candidates can build interview confidence early through school-level programmes.

FAQ

Does my accent affect my IELTS Speaking score?

No, accent itself is not marked. The Pronunciation criterion assesses clarity, stress and intonation, and examiners are trained to assess every accent fairly. What does affect the score is when pronunciation makes meaning hard to follow, such as misplaced word stress or key sounds missing from important topic words.

How long is the IELTS Speaking test?

Between 11 and 14 minutes in total: roughly 4 to 5 minutes for Part 1, 3 to 4 minutes for Part 2 including preparation, and 4 to 5 minutes for Part 3. The short length is why every answer matters, and why habits like one-word replies have an outsized effect on the final band.

Can I ask the examiner to repeat a question?

Yes. You can ask the examiner to repeat a question or to clarify a word, and doing so once or twice is normal interaction rather than a penalty. What scores poorly is silence or answering a different question, so asking is nearly always the better recovery move.

What happens if I run out of things to say in Part 2?

Keep talking around the prompt, using the bullet points on the cue card as anchors, then extend to related feelings, examples or contrasts until the examiner stops you. The topics invite personal stories, and examiners score the language, not the interest value of the story, so any relevant material serves you.

Is Speaking the same for Academic and General Training?

Yes. The three-part structure, the timing and the four marking criteria are identical for both modules. Examiners may frame topics slightly differently to fit a candidate's context, such as work versus study, but the band descriptors applied are the same for every candidate.

How can I practise speaking if I have no partner?

Record answers to published questions, replay them against the four criteria, and shadow natural audio to absorb stress and intonation. Self-practice builds volume and awareness, though live feedback, such as small-class coaching at iWorld Learning, is what calibrates your sense of your band before test day.

Summary

The examiner listening to your IELTS Speaking is scoring four things at once: steady connected speech, precise and natural vocabulary, varied and mostly accurate grammar, and clear pronunciation that carries meaning. Accent is not on the list, memorised answers are a liability, and every part of the test feeds all four criteria, so preparation works well when each practice task targets one dimension deliberately.

Want an experienced teacher to score your speaking against the real criteria? Contact iWorld Learning to arrange a consultation and a practice interview.