IELTS Speaking Band Descriptors: Band 6 vs Band 7

Two candidates answer the same cue card with the same story, the same examples and the same amount of detail. One is marked Band 6 and the other Band 7. The difference is not what they said; it is how the examiner can hear them say it.

This guide puts one answer next to itself. Transcript A is a Band 6 version of an answer and Transcript B is a Band 7 version of the same answer to the same cue card. Every line of both is mapped to the descriptor wording examiners actually work from, so you can see which criterion is doing the work - and which four edits separate the two versions.

Why the same answer can be heard two ways

IELTS Speaking is marked on four criteria - fluency and coherence, lexical resource, grammatical range and accuracy, and pronunciation - and the examiner awards one band for each of them. The published descriptors add the rule that matters here: a candidate must fully fit the positive features of the descriptor at a particular level, and the rating is based on their average performance across all parts of the test.

Two consequences follow. First, an answer is not scored line by line; the examiner asks whether the performance as a whole fits one band. Second, a band works as a ceiling as much as a mark: if hesitation keeps breaking the flow, the fluency criterion can stay at 6 even when the vocabulary in the same answer looks like 7.

That is why the comparison below runs criterion by criterion, not sentence by sentence. No amount of good content lifts a criterion that the delivery does not satisfy.

Every wording quoted in this guide comes from the official IELTS Speaking Band Descriptors for Academic and General Training, read on 29 October 2026: Speaking Band Descriptors, ielts.org (PDF). If you want the whole nine-band scale explained first, read the full guide to the IELTS Speaking band descriptors before you compare transcripts.

The cue card and the two transcripts

Describe a time when you had to explain something complicated to someone. You should say what you explained, who you explained it to, how you explained it, and how you felt about it.

Both transcripts tell one story: a first job, a new colleague, a reporting system, three steps, and a second attempt that finally works. The information is identical. Only the wording and the delivery change, and that is the whole point of the exercise.

Transcript A and Transcript B are constructed teaching examples, written to isolate one descriptor at a time. They are not recordings of a named candidate, and no official examiner has marked either version.

Transcript A: the Band 6 version

• I want to talk about, um, a time when I explained something... it was about my first job.
• So my manager, he asked me to teach a new colleague about the reporting system, and I say yes.
• But the problem is, the colleague, she don't know anything about it, so I need to start from the beginning and, and explain very slowly.
• I show her the steps one, two, three, and she write down everything, but I think she is still confusing.
• So I decide to sit with her and do the same thing again on the computer, and then it is ok.
• I think I learn that it is important to be patient and to check if the person understand.

Read this version for what the criteria describe rather than for its mistakes. The speaker keeps going and does produce long turns, which is exactly what the Band 6 fluency wording asks for. What it also does is lose coherence at the moments where repetition and self-correction pile up, keep its connecting words at the level of so, and and but, and place its errors where the structures become complex - she don't know, the person understand, she is still confusing.

Transcript B: the Band 7 version

• I'd like to talk about the first time I had to train someone properly, which was in my first job.
• My manager asked me to bring a new colleague up to speed on our reporting system, and I agreed straight away.
• The thing is, she had never seen the tool before, so I had to start from scratch and take it slowly.
• I walked her through the three steps, and although she wrote everything down, she still looked a bit unsure.
• So at that point I sat next to her and we repeated the whole process on the computer together, which is what finally made it click.
• Looking back, I'd say that taught me to check understanding rather than assume it.

The story has not changed. What changed is that the speaker now marks the turns of the argument instead of chaining clauses with and: The thing is, So at that point, Looking back. The register shifts into phrases a fluent speaker reaches for - bring someone up to speed, from scratch, walked her through, made it click - and the complex sentences arrive without the errors that interrupted version A.

Transcript A against the published Band 6 wording

Criterion	Line in Transcript A	Published Band 6 wording
Fluency and coherence	I want to talk about, um, a time when I explained something...	Able to keep going and demonstrates a willingness to produce long turns.
Fluency and coherence	so I need to start from the beginning and, and explain very slowly	Coherence may be lost at times as a result of hesitation, repetition and/or self-correction.
Lexical resource	I show her the steps one, two, three	Resource sufficient to discuss topics at length. and Vocabulary use may be inappropriate but meaning is clear.
Grammatical range and accuracy	she don't know anything about it	Produces a mix of short and complex sentence forms and a variety of structures with limited flexibility.
Grammatical range and accuracy	I think she is still confusing	Though errors frequently occur in complex structures, these rarely impede communication.

Every line here sits comfortably inside Band 6, and that is the diagnosis a Band 6 answer needs: the criterion is not failing because the answer is short or off topic, it is failing because the positive features of the next band are not fully present.

Transcript B against the published Band 7 wording

Criterion	Line in Transcript B	Published Band 7 wording
Fluency and coherence	I'd like to talk about the first time I had to train someone properly, which was in my first job.	Able to keep going and readily produce long turns without noticeable effort.
Fluency and coherence	The thing is, she had never seen the tool before, so I had to start from scratch	Flexible use of spoken discourse markers, connectives and cohesive features.
Lexical resource	My manager asked me to bring a new colleague up to speed on our reporting system	Resource flexibly used to discuss a variety of topics. and Some ability to use less common and idiomatic items and an awareness of style and collocation is evident though inappropriacies occur.
Grammatical range and accuracy	although she wrote everything down, she still looked a bit unsure	A range of structures flexibly used. Error-free sentences are frequent.
Grammatical range and accuracy	which is what finally made it click	Both simple and complex sentences are used effectively despite some errors. A few basic errors persist.

The same story now fits the Band 7 wording in three criteria at once, and it does so without extra information. Note what the table does not show: nothing here depends on the cue card being harder, the vocabulary being rarer or the answer being longer. Only the delivery changed.

The four edits that move A to B

1. Replace the filler with a discourse marker. Version A opens with um and keeps restarting; version B opens the same idea with I'd like to talk about the first time..., which fits the Band 7 requirement for Flexible use of spoken discourse markers, connectives and cohesive features.
2. Move the self-correction out of the middle of the sentence. In version A the repetition sits inside the clause - and, and explain very slowly. Band 7 allows some hesitation and repetition, but only where it does not affect coherence.
3. Trade a general verb for the phrase a fluent speaker uses. show her the steps becomes walked her through the three steps, which is the less common and idiomatic side of the lexical resource wording.
4. Deliver the complex sentence without an error. Version A attempts complexity and slips - if the person understand; version B produces which is what finally made it click cleanly, which is what turns errors frequently occur into Error-free sentences are frequent.

None of the four edits adds a fact. If you try them in this order, you are changing one criterion at a time, which is the only way to hear which change actually moved the band.

Why pronunciation breaks the pattern

Three criteria can be checked on paper. The fourth cannot. Pronunciation is judged from the recording, not from the transcript, and its wording is written differently from the others: the Band 7 descriptor reads Displays all the positive features of band 6, and some, but not all, of the positive features of band 8.

That single sentence tells you two things. The Band 6 pronunciation features stay in place as a baseline - Uses a range of phonological features, but control is variable. - and progress above 6 is measured by picking up some of the Band 8 features, such as Can sustain appropriate rhythm. Flexible use of stress and intonation across long utterances, despite occasional lapses.

Practical consequence: when you review your own recording, do not ask whether the pronunciation was good. Ask which of the Band 6 features is unstable - chunking, rhythm, stress-timing - and treat that one as the target for the next attempt. The full wording for every band sits in the official descriptor document.

How to record the comparison yourself

1. Record the same cue card twice, a day apart, without reading your first transcript in between.
2. Transcribe both attempts as text and compare only the facts. If the second version drops or invents content, re-record it before comparing wording.
3. Work through the criteria in order, in the same order the tables above use: fluency and coherence first, then lexical resource, then grammatical range and accuracy.
4. Change one thing and record again. Pick the first criterion that does not yet fit the Band 7 wording, change one feature of it, and keep everything else identical so the difference is audible.
5. Keep a one-line log: date, criterion, the change, and whether you can hear the difference in the new recording.

Two attempts compared in this way are more informative than ten unreviewed recordings, because each pair answers one question: which criterion is holding the answer down.

Frequently Asked Questions

How to improve IELTS Speaking from 6 to 7?

Find the first criterion whose wording you do not yet fit, and work on that one alone. In the transcripts above the entry point was fluency and coherence: the story was there, but repetition and self-correction kept interrupting it. Practise the same answer until the discourse markers do the work that and was doing, then move to the vocabulary criterion and replace two general phrases with idiomatic ones. Grammar usually improves as a side effect of speaking in longer, planned turns rather than short chains of clauses.

Can the same IELTS Speaking answer be Band 6 and Band 7?

Yes, and the published descriptors explain why. Each of the four criteria is judged on whether the performance fully fits the wording of one band, and the delivery decides that, not the information in the answer. The same story told with hesitation that breaks coherence, general vocabulary and unstable complex structures fits Band 6; told with flexible discourse markers, idiomatic phrases and frequent error-free sentences, it fits Band 7. What it cannot do is fit both at once, because the descriptors describe two different performances of the same content.

Where the descriptors leave you

A band score is not a verdict on your ideas. It is a description of how you delivered them, criterion by criterion, and that is why the same answer can be rewritten upward without learning anything new about the topic.

Keep the four criteria in view when you practise. Record the same cue card a day apart, compare it against the published wording rather than against a model answer, and change one criterion at a time. Your speaking score is the average of the four criteria rounded to the nearest half band, so lifting one criterion from 6 to 7 moves the section score by 0.25 - and a second criterion moved in the same direction is usually what carries an overall score over the line. The rounding rules are set out in IELTS scoring in detail.