---
title: "8 Common Mistakes in IELTS Speaking (And Fixes)"
description: "The 8 IELTS Speaking mistakes examiners actually penalize, mapped to the 4 band criteria, with do/don't examples for each."
lang: "en"
slug: "common-mistakes-ielts-speaking"
pubDate: 2026-07-17
---
Common IELTS Speaking mistakes include memorized answers, one-word Part 1 responses, ignoring cue card bullets, filler and connector overuse, excessive self-correction, asking to change the topic, under-speaking in Part 2, and treating Part 3 like Part 1. Each one maps to a specific band criterion examiners are trained to penalize.

Most candidates don't lose bands because their English is weak—they lose bands because of test-taking habits that examiners are specifically trained to notice. Understanding which mistake damages which of the four criteria (see the [test structure and criteria breakdown](/blog/en/ielts-speaking-test-structure-criteria)) lets you fix the right thing instead of over-studying vocabulary when your real problem is pacing.

## Mistakes That Damage Fluency and Coherence

### Giving memorized-sounding answers

Many candidates prepare "model answers" for common topics and recite them word-for-word, complete with unnatural pauses at memorized phrase boundaries. Examiners are trained specifically to detect recitation—the rhythm, intonation, and vocabulary don't match the rest of the candidate's spontaneous speech, and the answer often doesn't quite fit the question asked.

This is penalized under **Fluency and Coherence** because natural fluency requires real-time language processing, not playback. A memorized answer that's technically fluent still scores poorly because it isn't responsive to the actual question.

- **Don't:** Recite a pre-written 150-word answer about "a person you admire" regardless of what the cue card actually asks.
- **Do:** Prepare ideas and vocabulary in advance, but construct the sentences live, adapting to the specific bullet points on the card.

### Overusing fillers and connector spam

Some candidates pad every answer with "um," "you know," "like," and "basically" to buy thinking time. Others overcorrect by stacking formal written connectors—"moreover," "furthermore," "in addition to this"—into spoken answers, which sounds like a scripted essay rather than a conversation.

Both extremes are penalized under **Fluency and Coherence**: excessive fillers signal hesitation and disrupt flow, while written-register connectors like "moreover" in speech sound unnatural and can even suggest memorization.

- **Don't:** "Moreover, I would like to say that, furthermore, my hometown is, um, quite, you know, big."
- **Do:** "My hometown's actually pretty big—it's one of the largest cities in the region."

### Self-correcting every sentence

Stopping to fix minor grammar mistakes mid-sentence, over and over, breaks the natural flow of speech and signals anxiety rather than control. A small amount of self-correction is normal and even positive, but constant restarting turns a 2-minute answer into a stop-start ordeal.

This damages **Fluency and Coherence** directly, since the criterion explicitly rewards smooth delivery with only occasional hesitation.

- **Don't:** "I go—I went, I mean I have went, no—I went to the beach last summer."
- **Do:** "I went to the beach last summer." (Let small errors go if the meaning is clear.)

## Mistakes That Damage Task Response and Coherence

### Giving one-word answers in Part 1

Part 1 rewards natural, complete answers of roughly 30-60 seconds, but nervous candidates often answer with a single word or short phrase ("Yes." "I like reading.") and wait for the next question. This starves the examiner of enough language to assess.

Short answers limit every criterion at once, but the primary damage is to **Fluency and Coherence**, since there isn't enough connected speech to demonstrate smooth, logical delivery.

- **Don't:** Examiner: "Do you enjoy cooking?" Candidate: "Not really."
- **Do:** "Not really, actually—I never learned properly growing up, so I tend to rely on simple meals or eating out with friends."

### Ignoring the cue card bullet points

The Part 2 cue card always includes 3-4 specific prompts ("what it is, where you got it, who gave it to you, and how you felt about it"). Candidates who talk generally about the topic without addressing the bullets leave the examiner unable to credit organized, developed speech.

This is penalized under **Fluency and Coherence**, which rewards a clear structure and logical progression through ideas—something the bullets are designed to scaffold.

- **Don't:** Talk for two minutes about "a gift" in general terms without ever saying who gave it or how you felt.
- **Do:** Use the bullets as a loose outline, moving through each one with a specific example, as modeled in our [Part 2 cue card model answers](/blog/en/ielts-speaking-part-2-cue-card-model-answers).

### Speaking too little in Part 2

The Long Turn gives you up to two minutes, but many candidates stop after 30-45 seconds, unsure what else to say. Since the examiner cannot ask follow-up questions to extend a weak answer, running short leaves large gaps in the evidence available for scoring.

Under-speaking mainly damages **Fluency and Coherence** (insufficient evidence of sustained, developed speech) but also caps **Lexical Resource**, since there's less language sampled overall.

- **Don't:** Speak for 40 seconds, then say "That's all I can think of" and stop.
- **Do:** Add a specific anecdote, a sensory detail, or a reflection on why the topic matters to reach closer to the full two minutes.

### Treating Part 3 like Part 1

Part 3 questions are abstract and discussion-based ("How has technology changed the way people learn?"), expecting extended, reasoned answers with examples and justification. Candidates who default to short, factual Part 1-style responses ("I think it's good.") underperform badly here.

This limits both **Fluency and Coherence** (no development) and **Lexical Resource** (no opportunity to show range), since Part 3 is designed to elicit more sophisticated language than Part 1. See our [Part 3 questions and answers guide](/blog/en/ielts-speaking-part-3-questions-and-answers) for the expected answer shape.

- **Don't:** "How has technology changed education?" → "It's changed a lot, yes."
- **Do:** "It's changed enormously—online resources mean students aren't limited to what's in a textbook anymore, though I think it's also made it harder to focus, since there are so many distractions available at the same time."

## The Mistake That Damages Task Handling Directly

### Asking to change the topic

Occasionally a candidate gets a cue card on an unfamiliar topic and asks the examiner for a different one, or announces they "don't know anything about this." Examiners cannot swap cards, and doing this signals an inability to handle the task under pressure—a skill the test is explicitly assessing.

While there's no single "task response" criterion in Speaking the way there is in Writing, this behavior drags down **Fluency and Coherence** (the derailed, hesitant restart that usually follows) and indirectly **Lexical Resource**, since the candidate typically falls back on very basic vocabulary once flustered.

- **Don't:** "I don't have a public place to talk about—can I do a different topic?"
- **Do:** Adapt: invent a plausible example or draw on a loosely related experience. Examiners reward resourcefulness, not literal truth.

## Mistake-to-Criterion Mapping

| Mistake | Criterion Damaged | Why |
|---|---|---|
| Memorized-sounding answers | Fluency & Coherence | Recitation rhythm doesn't match spontaneous speech; often off-question |
| Filler/connector spam ("moreover") | Fluency & Coherence | Excessive hesitation markers or unnatural written-register connectors |
| Self-correcting every sentence | Fluency & Coherence | Breaks smooth delivery; signals low confidence |
| One-word Part 1 answers | Fluency & Coherence | Insufficient connected speech to assess |
| Ignoring cue card bullets | Fluency & Coherence | No clear structure or development of ideas |
| Speaking too little in Part 2 | Fluency & Coherence, Lexical Resource | Insufficient evidence of sustained, varied speech |
| Treating Part 3 like Part 1 | Fluency & Coherence, Lexical Resource | Underdeveloped answers miss the task's higher demands |
| Asking to change the topic | Fluency & Coherence, Lexical Resource | Derails flow; candidate falls back on basic language |

Most of these mistakes cluster around Fluency and Coherence because it's the criterion most sensitive to test-taking behavior rather than raw English ability. Fixing them is often about practice habits, not vocabulary lists—see our guide on [how to reach Band 7 in IELTS Speaking](/blog/en/how-to-get-band-7-ielts-speaking) for a criterion-by-criterion improvement plan, and [practicing without a teacher](/blog/en/practice-ielts-speaking-without-teacher) for structured self-study routines that build these habits before test day.

## Practice Under Real Exam Conditions

The mistakes above are hard to catch on your own—memorized phrasing and pacing problems are much easier for a trained ear (or a calibrated AI examiner) to spot than for the candidate speaking in the moment. OralPrep is an AI oral exam simulator that runs the full IELTS Speaking test with an AI examiner across all three parts, including the timed one-minute Part 2 preparation, then scores your performance against the official band descriptors criterion by criterion. [Try OralPrep now](https://oralprep.app)