Direct current-format answer

The current IELTS Speaking test lasts 11–14 minutes and has three parts. The Speaking test is the same for IELTS Academic and IELTS General Training. In standard test-centre delivery, it is a recorded, one-to-one, face-to-face interview with a certified IELTS examiner, who assesses your performance throughout the test.

There is one narrow delivery exception worth knowing. Where IELTS Online is currently available, the Academic Speaking test is still conducted live with a certified IELTS Examiner, but the conversation happens by video call. It is not an automated speaking task. Availability varies, so check the official IELTS booking information for your location rather than assuming the online option exists everywhere.

IELTS Speaking format at a glance
Format point Current official position What it means for preparation
Total time 11–14 minutes Train concise answers, a sustained long turn, and deeper discussion—not a 30-minute speech marathon.
Number of parts Three Practise changing gear from familiar questions to a long turn and then to abstract discussion.
Academic versus General Training The Speaking test is the same You do not need a separate Academic or General Training speaking strategy.
Interaction Recorded interview with an examiner Practise responding to a person in real time, including follow-up questions and natural changes of direction.
Assessment Four equally weighted criteria Do not spend all your preparation time collecting vocabulary while ignoring coherence, grammar, or intelligibility.

For the source-of-truth version, check the official IELTS Academic: Speaking test format, IELTS test types, and IELTS scoring information. Those pages should win any argument with a prep blog, a five-year-old video, or your cousin’s extremely confident group-chat advice.

What happens in Parts 1, 2, and 3

The three parts are not three random piles of questions. They create a deliberate progression: speak about familiar life, organise a longer individual turn, and then discuss a related issue in a more general and abstract way. Your preparation should follow that progression.

The task, timing, and useful preparation target for each Speaking part
Part What happens Official timing Best preparation target
Part 1: Introduction and interview The examiner checks your identity and asks general questions about familiar topics such as home, work, studies, family, or interests. 4–5 minutes Answer immediately, develop the answer enough to show real language, and stop before it turns into a miniature Part 2 speech.
Part 2: Long turn You receive a task card, have one minute to prepare with pencil and paper, then speak for up to two minutes. The examiner may ask one or two questions on the same topic. 3–4 minutes, including preparation time Turn brief notes into a clear route through the topic and keep speaking until the examiner stops you.
Part 3: Discussion You and the examiner discuss issues connected to the Part 2 topic in a more general, abstract, and sometimes deeper way. 4–5 minutes State a position, explain it, test it with an example or contrast, and respond flexibly when the examiner changes the angle.

Part 1: familiar does not mean one-word answers

Part 1 questions are about everyday topics, but the examiner still needs enough language to assess. A useful response shape is direct answer → reason or detail → optional example. It is a shape, not a script.

Original Part 1 practice question: What is one part of your daily routine that you genuinely enjoy?

A dedicated IELTS Speaking Part 1 guide is planned at /learn/guides/ielts-speaking-part-1/. Until it is live, focus on answering the exact question first and adding one relevant layer rather than stuffing every answer with a memorised story.

Part 2: use the minute to build a route

The task card gives you a topic and points to cover. Your one-minute preparation is for notes, not for writing a perfect paragraph in microscopic handwriting. Choose a beginning, two or three content stops, and an ending. Then speak from the route rather than reading your notes in your head.

Original Part 2 practice cue card—not an official IELTS test item

Describe a small decision that improved your daily life.

You should say:

  • what the decision was
  • when and why you made it
  • what changed afterwards
  • and explain why the decision was worthwhile.

For a focused method, use IELTS Speaking Part 2: what to do during the 60-second preparation time. The goal is not to predict every cue card. It is to make your planning method reusable when the topic is unfamiliar.

Part 3: move from your experience to the wider issue

Part 3 is linked to the Part 2 theme, but it asks you to analyse, discuss, explain, and speculate rather than simply tell another personal story. Listen for the level of the question: is it asking about individuals, society, causes, consequences, comparisons, or the future?

Original Part 3 abstract follow-up: Why do some people find it difficult to change everyday habits even when they understand the benefits?

A practical answer could move through position → reason → example → qualification. For help developing that kind of answer without wandering into the wilderness, use IELTS Speaking Part 3: how to extend an answer without rambling.

The four equally weighted criteria

IELTS Speaking is assessed through Fluency and Coherence, Lexical Resource, Grammatical Range and Accuracy, and Pronunciation. The four criteria carry equal weight. Examiners assess your average performance across the parts, so there is no sensible strategy in “sacrificing” one criterion and hoping a mountain of fancy vocabulary will rescue everything else.

What each official assessment criterion asks you to demonstrate
Criterion What the examiner is listening for A useful review question
Fluency and Coherence Continuity, an appropriate rate and level of effort, logical sequencing, relevant development, and effective linking within connected speech. Could a listener follow the route of my answer without doing detective work?
Lexical Resource Range, precision, appropriacy, collocation, expression of attitude, and the ability to paraphrase around a vocabulary gap. Did I choose words that fit the meaning, or did I reach for impressive language that did not quite fit?
Grammatical Range and Accuracy A useful range of sentence structures and accurate, appropriate grammar, including how frequent errors are and whether they affect meaning. Did my grammar help me express the relationship between ideas, and which repeated error made communication less precise?
Pronunciation Meaningful chunking, rhythm, stress, intonation, connected speech, word and sound production, and overall intelligibility. Could someone understand me without a transcript, and did my voice make the important meaning easy to hear?

One answer, four different lenses

Return to the Part 1 question: What is one part of your daily routine that you genuinely enjoy?

“I actually enjoy the walk to the station in the morning. It only takes about fifteen minutes, but it gives me a clean break between home and work. I usually listen to a podcast, and if the weather is good, I take a slightly longer route through a park. It sounds ordinary, but I arrive feeling more awake and less rushed.”

This is not a band-score verdict. It is a demonstration of what changes when you review the same language through four criteria.

How to review the same answer through all four criteria
Lens What is visible or audible in this answer What still needs a recording to judge
Fluency and Coherence The answer moves from a direct choice to duration, purpose, routine detail, and result. The ideas stay relevant and form a clear route. Whether pauses interrupt the flow, whether the rate is natural, and whether self-correction damages coherence.
Lexical Resource “Clean break,” “slightly longer route,” “more awake,” and “less rushed” express precise everyday meanings without circus-trick vocabulary. Whether the speaker can paraphrase if a word disappears and whether the word choices sound natural in delivery.
Grammatical Range and Accuracy The answer mixes simple sentences with coordination and an if-clause, using grammar to show contrast and condition. Whether errors appear in the spoken version, how dense they are, and whether they interfere with meaning.
Pronunciation The text offers natural meaning groups such as “the walk to the station / in the morning” and contrastive words that could carry stress. Almost everything important: chunking, rhythm, stress, intonation, sound clarity, and listener effort.

The full official descriptions are in the Speaking Band Descriptors (PDF) and IELTS Speaking Key Assessment Criteria (PDF). A separate learner-facing guide, IELTS Speaking Band Descriptors Explained, is planned at /learn/guides/ielts-speaking-band-descriptors-explained/. This hub deliberately does not duplicate that future page band by band.

For skill work beneath the exam labels, use how to organise ideas while speaking English for coherence and the pronunciation hub for intelligibility, stress, rhythm, and connected speech.

A diagnostic by failure symptom

“My speaking is bad” is not a diagnosis. It is fog. Record one answer from each part, then look for an observable symptom: where the answer stops, loses its route, becomes vague, collapses grammatically, or becomes difficult to understand. The table below gives you a useful starting hypothesis—not a medical scan of your English soul.

Map a repeated speaking symptom to the next useful practice focus
Observable failure symptom Likely bottleneck Start here next
Your Part 1 answer ends after one short sentence even when you know the topic. Fluency and Coherence: limited development, or no reliable way to add a reason or detail. Use four elaboration moves for longer answers without rambling, then practise the planned Part 1 route at /learn/guides/ielts-speaking-part-1/ when it goes live.
You spend the Part 2 preparation minute writing full sentences and still begin without a route. Task management and coherence: notes are consuming the planning time instead of organising it. Use the 60-second Part 2 preparation method; practise with keywords and a beginning-middle-end route.
You reach 40–60 seconds in Part 2 and run out of content. Topic development: too few content layers, weak sequencing, or no concrete detail. Add place, people, action, change, and result. For personal-event topics, use the setup, turning point, and result storytelling structure.
Your Part 3 answer has several ideas, but they arrive in a confusing order. Coherence: the listener cannot tell which idea is the position, reason, example, or qualification. Use a simple speaking structure for organising ideas, then apply it to the Part 3 extension guide.
Your opinion sounds absolute, vague, or strangely aggressive even when your idea is reasonable. Lexical Resource and discourse control: the stance is not expressed with enough precision. Practise how to state an opinion with appropriate stance markers.
You repeat broad words such as “good,” “bad,” “interesting,” or “important” across unrelated topics. Lexical Resource: limited precision or slow access to alternatives. After recording, replace only two repeated words with more precise, natural alternatives. Also practise paraphrasing one missing word instead of stopping to hunt for it.
Every attempt at a complex sentence creates a long pause, restart, or loss of meaning. Grammatical Range and Accuracy is overloading Fluency and Coherence. Return to one accurate main clause, then add one controlled relationship—because, although, if, when, or which. Build range gradually instead of launching a grammar chandelier and dropping it on the sentence.
Your answer looks clear in a transcript, but listeners need the transcript to understand it. Pronunciation: likely problems with chunking, stress, rhythm, individual sounds, or overall intelligibility. Start with the pronunciation hub and work on one repeated intelligibility problem at a time.
A memorised answer sounds polished, but an unexpected follow-up causes a complete breakdown. Flexible language access: the prepared script is stronger than spontaneous response ability. Keep the topic but change the question angle. Answer again without reusing the opening sentence, then compare whether the same ideas survive in fresh language.

Diagnostic rule: do not redesign your entire preparation plan because of one awkward answer. Choose the first symptom that appears in at least two recordings, make it your main target for several sessions, and then record again under similar conditions. Repeated evidence beats post-practice panic.

A four-week preparation plan

This is a practical study model, not an official IELTS prescription and not a promise of a particular band. It assumes roughly 30–45 minutes on five days each week. With only three days, keep the same sequence—record, review, repair—but combine the shorter sessions.

A realistic four-week IELTS Speaking preparation roadmap
Week Primary job Five-session plan Evidence that the week worked
Week 1: Learn the test and find the leak Understand the current format and capture a cold baseline.
  1. Read the official format and criteria.
  2. Record several Part 1 answers without scripts.
  3. Record one Part 2 long turn with one minute of notes.
  4. Record several Part 3 answers on one connected theme.
  5. Use the failure-symptom diagnostic and choose one primary issue plus one secondary issue.
You can describe the three parts accurately and point to a repeated, observable problem in your own recordings.
Week 2: Build answer shape Improve Fluency and Coherence before trying to decorate every sentence.
  1. Practise Part 1 with direct answer → reason/detail.
  2. Practise Part 2 notes using a beginning, two or three stops, and an ending.
  3. Practise one Part 2 topic twice: first for route, then for richer detail.
  4. Practise Part 3 with position → reason → example or contrast.
  5. Record a mixed session and check whether the listener can follow each answer’s route.
Your answers have a recognisable structure without sounding like the same template wearing different hats.
Week 3: Strengthen the four criteria Repair the weakest criterion while keeping the others active.
  1. Lexical Resource: replace vague repetition and practise paraphrasing gaps.
  2. Grammatical Range and Accuracy: add one controlled complex structure to accurate speech.
  3. Pronunciation: mark chunks and meaning stress in a short extract, then re-record it.
  4. Fluency and Coherence: answer a changed follow-up without restarting from a memorised opening.
  5. Compare the first and second recording of the same task; keep only repairs that improve real delivery.
At least one repeated problem is less frequent or less disruptive in comparable recordings.
Week 4: Stabilise under timing Make your method reliable when topics and question angles change.
  1. Run one timed mixed sequence covering all three parts.
  2. Review it and choose only the two highest-impact repairs.
  3. Repeat with fresh topics rather than polishing the same answers.
  4. Do one final mixed recording with no pauses or restarts.
  5. Finish with light review: format, personal error log, flexible structures, and sleep—not a midnight idiom-shopping spree.
You can keep going, organise ideas, and remain understandable across unfamiliar prompts without depending on a fixed script.

The session pattern that makes the plan work

  1. Warm up for five minutes: answer easy familiar questions aloud.
  2. Record one target task: keep test-like timing for that part and do not stop halfway to repair.
  3. Review one short section deeply: identify the most disruptive repeated issue.
  4. Repair deliberately: change the route, wording, grammar, or delivery—not all four at once.
  5. Re-record once: compare clarity and control, then move on.

This keeps preparation active. Reading model answers may give you ideas, but only speaking exposes whether you can retrieve and organise language at conversational speed.

How to record and review answers

A recording is useful only when it produces a repair. Collecting 200 audio files without reviewing them is not a study system; it is a podcast archive nobody subscribed to.

The record–review–repair loop

  1. Record a clean first take. Use one prompt, respect the relevant timing, and do not pause the recorder when you hesitate.
  2. Write the message in one sentence. Before analysing language, note what you were trying to communicate. This protects you from “improving” an answer until its meaning disappears.
  3. Listen once without reading anything. Mark where the route becomes hard to follow or where understanding requires extra effort.
  4. Listen again through the four criteria. Note one piece of evidence for Fluency and Coherence, Lexical Resource, Grammatical Range and Accuracy, and Pronunciation.
  5. Use a transcript microscope, not a transcript swamp. Transcribe only a troublesome 30–60 second extract. Circle repeated words, unfinished clauses, avoidable self-corrections, and places where the written version hides unclear delivery.
  6. Choose one main repair. For example: make the opening direct, replace two vague words, simplify a broken clause, or re-chunk a sentence for intelligibility.
  7. Re-record with the same communicative goal. Do not read a rewritten paragraph. Say the answer again from brief notes and compare the two recordings.
A compact self-review sheet for one recording
Check Evidence to write down Possible one-step repair
Relevance and route The exact point where the answer drifts, jumps, or stops developing. State the answer in the first sentence and keep one reason plus one example.
Pausing and continuity Whether pauses are for thinking about content or searching for basic words and grammar. Practise the missing phrase family, or simplify the sentence you repeatedly abandon.
Vocabulary Two repeated or inaccurate words and one moment where you successfully paraphrased. Prepare natural alternatives in complete phrases, not isolated “advanced” words.
Grammar One repeated pattern of error and whether it changes or obscures meaning. Rebuild that idea with one accurate structure, then vary it in two fresh examples.
Pronunciation A phrase that required extra listener effort and the likely cause: chunking, stress, rhythm, or a sound. Mark the meaning groups, stress the key words, shadow a clear model if available, and re-record.
  • I answered the question that was asked, not the question I wished had been asked.
  • I can point to the route of the answer in a few words.
  • I identified evidence rather than guessing a band from how confident I felt.
  • I repaired one high-impact issue and made a second recording.
  • I kept the original and repaired versions so progress can be compared.

A dedicated IELTS speaking practice route is planned at /practice/ielts-speaking/. The live fallback is FunFluen general speaking practice. It is general speaking practice, not an official IELTS simulation or scoring service; an IELTS-labelled prompt used for practice remains practice.

Common myths

IELTS Speaking myths that waste preparation time
Myth Reality Better preparation choice
“Academic Speaking is harder than General Training Speaking.” The Speaking section is the same in both test types. Prepare for the same three-part format and four criteria. Choose Academic or General Training for the requirements of the full test, not for a different speaking interview.
“If I take IELTS on a computer, I will speak to a machine.” The standard Speaking test is an interview with an examiner. Where IELTS Online is available, the Speaking test is a live video call with a certified examiner. Practise live, responsive conversation rather than recording isolated sentences for a robot.
“I need a British or American accent.” The official pronunciation descriptors focus on intelligibility and the effective use of phonological features. Accent matters when it affects how easily you are understood, not because one national accent is the secret password. Work on chunking, rhythm, stress, intonation, and recurring sound problems that increase listener effort.
“More idioms and rare words automatically mean a higher score.” Lexical Resource includes range, but also precision, appropriacy, collocation, and paraphrase. A forced idiom used inaccurately is not a vocabulary victory. Prefer exact, natural language and build flexible phrase families you can retrieve under pressure.
“Speaking faster proves fluency.” Fluency includes continuity, rate, and effort, while coherence and pronunciation still need to survive. Very rapid speech can damage rhythm, chunking, clarity, and listener comprehension. Aim for connected speech at a manageable pace. Remove unnecessary restarts before trying to increase speed.
“One grammar mistake ruins the whole Speaking score.” Examiners consider performance across the test, including error density and the communicative effect of errors. One slip is not the same as a repeated pattern that obstructs meaning. Track repeated errors, especially those that change meaning, while continuing to speak and develop the answer.
“Part 1 is just a warm-up, so it does not really count.” Examiners assess speaking performance throughout the test, and candidates are rated on average performance across all parts. Use Part 1 to show clear, natural control—without turning every familiar question into a memorised monologue.
“Memorising enough model answers will cover every question.” The wording, topic angle, and follow-ups change. A polished script does not prove that you can respond flexibly when the conversation moves. Reuse structures and language resources, not whole answers. Change the question and practise rebuilding the message.

Choose the next part or criterion

Do not open nine tabs and “work on everything.” Pick the route that matches the failure you can actually hear.

The best next step is usually boringly specific: one part, one criterion, one repeated symptom, one new recording. That is how a vague goal such as “improve IELTS Speaking” becomes something you can actually train this week.