IELTS Speaking Part 1 and Part 3: Weekly Question Practice
Most IELTS speaking prep treats Part 3 like a debate club warm-up: teach students to argue, hedge, concede a point, build a case. It's a reasonable guess — Part 3 is officially labelled a "two-way discussion," and it comes right after the solo long turn of Part 2, so it feels like the moment the candidate finally gets to really talk. But a Conversation Analysis study of 60 transcribed IELTS Speaking tests, published by IELTS's own research programme, found something that should change how that prep time gets spent: Part 3 is interactionally almost identical to Part 1. Both are built from the same underlying shape — an examiner asks a question, the candidate answers, the examiner asks another. What changes between the two parts isn't the interactional structure at all. It's the topic. Getting IELTS Speaking Part 1 and Part 3 practice right means training candidates to answer confidently under examiner-led questioning on subjects that get progressively less concrete, not training them to hold the floor in a debate that the test never actually asks them to have.
What the research actually found
Seedhouse and Harris (2011), in a report titled "Topic development in the IELTS Speaking test" hosted on IELTS's own official research-reports page, ran a Conversation Analysis institutional-discourse methodology over a corpus of 60 transcribed test audio-recordings (what candidates and teachers usually just call IELTS speaking task 3, since "task" and "part" get used interchangeably in how people search for this material). Their finding, in their own words: "although Part 3 is termed 'two-way discussion,' it is almost identical to Part 1 interactionally." Both parts rely on the same structural unit — what they call a topic-based question-answer adjacency pair — where the examiner introduces every topic through a question and the candidate responds. The study also found that candidates have very little control over where the conversation goes: "there are hardly any opportunities for candidates to introduce or shift topic and they are generally closed down when they try to do so." Topic management, across both parts, sits almost entirely with the examiner's script.
That's a genuinely useful correction to the debate-prep instinct. A candidate who has practiced free-flowing argumentation — building a case, anticipating counterpoints, steering the conversation — has trained a skill Part 3 doesn't actually test. What Part 3 tests is the same skill Part 1 tests: answering a question fluently, on the spot, without a script. The only thing that shifts is the depth of reasoning the question demands.
Part 1: familiar, personal, fast
IELTS's own Speaking test format page (checked live 2026-08-05) describes Part 1 as 4 to 5 minutes of introduction and interview, covering familiar topics — home, family, work, studies, interests. The examiner asks, the candidate answers, and the exchange moves quickly from one question to the next. There's no preparation time and no expectation of extended reasoning; a good Part 1 answer is a direct response with a little natural elaboration, delivered without hesitation.
Part 3: same shape, harder content
Part 3 runs another 4 to 5 minutes, and the official format page describes it as exploring the Part 2 long-turn topic "in a more general and abstract way." Per Seedhouse and Harris's findings, the interactional demand doesn't actually change — it's still examiner-question, candidate-answer. What changes is what the question is about: instead of "Do you work or are you a student?" a Part 3 question looks more like "Why do you think some people find it harder to change careers later in life?" The candidate still isn't required to build an argument and defend it against pushback the way a debate would ask. They're required to answer an abstract question as fluently as they answered a concrete one in Part 1. That's a different, narrower skill than debate training, and one that's much easier to actually rehearse in a normal lesson.
Ask, Tell, Reveal: rehearsing the Part 1 shape
Ask, Tell, Reveal generates a speaking action — Ask, Tell, or Reveal — with a CEFR-levelled prompt on a rotating theme (family, work, hobbies, travel, and six more). It's built on exactly the question-answer rhythm Seedhouse and Harris describe: one student is handed a prompt, they respond, the app can chain a natural follow-up onto their answer. That chain-reaction mechanic is closer to a real Part 1 exchange than most speaking activities get, because a genuine examiner follow-up ("And do you enjoy that?") is exactly what the app's chain reaction generates from what the student just said.
The app's own theory grounding leans on Austin (1962) and Searle (1969)'s speech-act theory — the idea that Ask, Tell, and Reveal are three distinct illocutionary types, each demanding different discourse resources. For IELTS Part 1 specifically, the Ask and Tell prompts are the closer match: they're the two speech acts that mirror an examiner's actual question types ("Tell me about your hometown," "What do you do in your free time?"). The app's "Opinion Round" prompt tier at B1 is worth running deliberately for IELTS prep, since it pairs a Tell prompt with a scaffold that forces a stated opinion plus a reason — "I think... because..." — which is a near-literal rehearsal of the kind of real IELTS Part 1 interview questions that ask not just what a candidate does, but why they like it.
Say Mean Quotes: training the move from concrete to abstract
Say Mean Quotes generates a real, attributed quote and walks students through three analytical stages in strict sequence: Say (what does it literally mean), Mean (what is the author really saying), and Matter (why does this still matter today). That three-stage escalation — literal, then interpretive, then abstract and evaluative — is a close structural match for what happens across an IELTS test as a whole, and especially for the jump from Part 1 to Part 3. Candidates move from stating facts to interpreting implications to explaining why something matters more broadly, all while still just answering the question in front of them rather than mounting an argument.
The app's own citation, Fisher and Frey (2012)'s Say-Mean-Matter framework, sequences comprehension before analysis before evaluation — the same order Anderson and Krathwohl's (2001) revision of Bloom's Taxonomy places those cognitive demands in. Run the Matter stage specifically as abstract discussion practice for Part 3: once students have worked through what a quote says and means, ask them to answer "why does this matter?" cold, under a short time limit, exactly the way an examiner's abstract follow-up question would land with no warning. The Quote Disagreement activity — where students must take a side and justify it using two Matter keyphrases — is the closest the app gets to genuine debate language, and it's worth running occasionally, but it isn't the main event; most of a Part 3 answer is closer to the Matter stage alone.
Sentence Lab: justifying an answer without a fight
Sentence Lab gives students three or four CEFR-levelled, part-of-speech-tagged words and challenges them to build one grammatically correct sentence — then explain it to the class. It's a smaller, faster activity than the other two, and its value for Part 3 prep is specific. The "justify your sentence" step is a low-stakes rehearsal of the exact move an abstract IELTS answer needs — stating something, then being ready to explain the reasoning behind it in a sentence or two, without turning the explanation into an argument. Swain's (1985) Output Hypothesis, already cited in the app's own theory array, is the grounding: sustained, syntactically demanding output forces exactly the kind of on-the-spot reasoning a Part 3 answer requires, just compressed into a single sentence instead of a paragraph.
| App | IELTS part rehearsed | CEFR range | Formation | What it trains |
|---|---|---|---|---|
| Ask, Tell, Reveal | Part 1, Introduction & Interview | A1–C1 | Free Circle | Fast, examiner-style Q&A on familiar topics |
| Say Mean Quotes | Part 3, Two-Way Discussion | B2–C2 | Groups of Three | Escalating from literal to abstract, evaluative answers |
| Sentence Lab | Part 3, Two-Way Discussion | A1–C2 | Whole Class | Stating and briefly justifying a position |
A weekly rotation built on the real shape of the test
Building a rotation this way also means picking real IELTS speaking part 3 topics — career changes, technology, education, the environment — and treating each one as a fast Q&A round rather than a debate prompt. A rotation that mirrors the research finding, rather than the debate-club instinct, looks like this: open with a short Ask, Tell, Reveal round on a rotating personal theme — five or six exchanges, examiner-style, no notice given — to rehearse Part 1's pace and register. Follow with a Say Mean Quotes round, but don't stop at the Say and Mean stages; push through to Matter and hold students there, cold-calling for an answer the way an examiner would ask a genuine abstract follow-up. Close with a quick Sentence Lab round on a related, more concrete topic, using the "justify it" step as a fast, low-pressure abstract-reasoning rep. That sequence takes about twelve minutes and rehearses the question-answer shape both parts actually share, with the topic difficulty climbing from familiar to abstract exactly the way the real test does.
I ran a version of this with a B2 exam-prep group after reading the Seedhouse and Harris finding, and it reframed a problem I'd been misdiagnosing. The group I'd been running through mock Part 3 "discussions" kept freezing — not because they lacked opinions, but because they'd been coached to build an argument, and building an argument from a cold start under time pressure is genuinely hard. Once I switched to running Say Mean Quotes' Matter stage as straight rapid-fire Q&A instead — no debate framing, just "why does this matter, answer now" — the same students who'd frozen on "discussion" answered the abstract questions almost as quickly as they'd answered the personal ones in the Ask, Tell, Reveal round beforehand. The interactional shape hadn't changed for them; only my instructions had.
What this doesn't replace, and how it's different from the Part 2 post
None of these three apps produce an IELTS-format Part 3 question bank, and none puts a student in front of an examiner reading from the real test's asymmetrical script — the version where the candidate genuinely cannot redirect the topic. This rotation builds the underlying skill (fluent, examiner-led question-answer at rising levels of abstraction) without claiming to substitute for timed practice with real past Part 1 and Part 3 questions close to the test date.
It's also worth being explicit about how this post relates to IELTS Speaking Part 2 Practice. Part 2 is the test's one genuinely different task — a solo, notes-then-monologue long turn with no examiner interruption during the two minutes of speech. Parts 1 and 3 are a different animal entirely: continuous question-answer with the examiner, never more than a few sentences at a time, and — per the research above — structurally the same task wearing two different topic difficulties. Don't run Part 2's timed-monologue rotation for Part 1 or Part 3 prep, and don't run this post's rapid-fire Q&A rotation as long-turn practice; they train genuinely different things.
Every app here lives inside the standard ENK account, gated by the same monthly AI credits as the rest of the library rather than a separate exam-prep tier — 100 credits free every month, or from €5/month for 300 credits on the pricing page. Browse the full app library to see how these three sit alongside the rest of the speaking-focused apps covered in ESL Speaking Activities.
FAQ
Is IELTS Speaking Part 3 really a debate? No. A Conversation Analysis study of 60 transcribed IELTS tests (Seedhouse & Harris, 2011) found Part 3 is interactionally almost identical to Part 1 — both are examiner-led question-answer exchanges, not open discussion or debate. What changes between the parts is the abstractness of the topic, not the interactional format.
How much control does a candidate have over the topic in Parts 1 and 3? Very little, according to the same research. Candidates have "hardly any opportunities" to introduce or shift topic in either part, and attempts to do so are generally closed down by the examiner, who controls topic development through a scripted sequence of questions.
What's the difference between preparing for Part 1 and preparing for Part 3? The interactional skill is the same — answering an examiner's question fluently and quickly, without a script. What needs to change is the topic's abstractness: Part 1 asks about familiar personal facts, while Part 3 asks the candidate to reason about more general, evaluative questions connected to the Part 2 topic.
Do these three apps replace practicing with real past IELTS questions? No. They rehearse the underlying question-answer skill at rising levels of abstraction, but they don't generate authentic IELTS-format question sets or reproduce the examiner's real scripted sequence. Closer to the test date, students still need timed practice with real past Part 1 and Part 3 questions.
Written by Prof. Vito Schiuma, designer of English, no kidding.