A remark only moves a script that sat on a band boundary. How to count your eight criterion marks and work out whether yours did, before you apply.
A remark is worth applying for when your essay was sitting on a band boundary, and close to worthless when it was not. A remark is not an appeal and not a second opinion on your ideas: a certified examiner marks the same script again against the same public band descriptors, so the only scripts it can move are the ones where two trained readers could reasonably have placed a criterion differently.
That turns the decision into something you can work out rather than something you gamble on. Below: what a remark actually does, how to tell from your own writing whether your 6.5 was a boundary 6.5 or a comfortable one, when the same money does more as a retake, and how to check the shape of your marks before you decide.
One thing you will not find here is fees, deadlines, processing times or success rates. Those vary by centre and by year, and the percentages that circulate in forums are not published figures. Take the procedural details from the centre where you booked your test, and from nowhere else.
One reported band, eight separate judgements behind it.
The single number on your report form is the end of a chain of eight separate judgements. Each of your two tasks is marked on four criteria — Task Achievement or Task Response, Coherence and Cohesion, Lexical Resource, and Grammatical Range and Accuracy — and each of those eight marks is a whole band awarded by one examiner reading against one set of descriptors. The four are averaged into a task band, and the two task bands are combined into the number you were sent.
Two candidates can both be sent 6.5 with completely different scripts behind it. One was marked at 7 on six of the eight criteria and at 6 on two. The other was marked at 6 on four and at 7 on four. The report form cannot tell them apart — it does not show criterion scores at all — but a remark treats them very differently.
For the first candidate there are two judgements in play, and if either was genuinely arguable, a second examiner reading the same evidence could land a band higher on it. For the second, four judgements would all have to move the same way. That is not bad luck being corrected; that is a different script.
So the question worth answering before you apply is not was I unlucky. It is how many of my eight marks were close enough to argue about. That number is the only thing about your result you can actually estimate, and this article is about estimating it.
A remark sends your script to a second certified examiner, who marks it again against the same public band descriptors using the same method. Nothing else about it changes. It is not a re-read for effort, not a chance to explain what you meant, and not a review of whether the question was fair.
The method matters, because it is where the room for a different result comes from. Examiners place a script at the band it fits best, not at a band whose every line it matches. A real essay almost always shows evidence from two adjacent bands at once — a paragraph that develops its idea properly next to one that only states it, complex sentences that are accurate in three places and wrong in the fourth. The examiner weighs which band the balance of that evidence sits closest to.
On a script where the evidence for one criterion is genuinely split down the middle, two trained readers applying that same rule can reasonably land on adjacent bands. That is the entire space a remark operates in. On a script where the evidence is not split — where three criteria are plainly at band 6 and read as band 6 from the first paragraph to the last — a second examiner reading the same wording will see the same thing, because they are looking for the same things.
This is also why "the examiner's score" is not a single fixed target that your result either hit or missed. Where marking can and cannot reasonably differ is worked through in more detail in how accurate AI writing scores are against a real examiner, and the descriptor wording every one of these judgements is made against is set out line by line in IELTS writing band descriptors explained.
You cannot get your marked script back, so this is a reconstruction rather than a lookup. Take the most recent essay you wrote under real conditions — an unseen question, 40 minutes, no editing afterwards — and mark it yourself on each of the four criteria separately, writing down a whole band for each one with the descriptor line you are relying on next to it. Do it for a Task 1 answer too. You now have eight marks in the same shape as the ones behind your result.
Then count the sixes.
Count how many of the eight sit a band below the rest.
Consider two scripts that both come out at 6.5.
Script A. Task 1 marked 7, 7, 7, 7 — a task band of 7.0. Task 2 marked 6, 6, 7, 7 — a task band of 6.5. Task 2 counts double, so the arithmetic is (7 + 6.5 × 2) ÷ 3 = 6.67, reported as 6.5.
Script B. Task 1 marked 6, 6, 7, 7 and Task 2 marked 6, 6, 7, 7 — 6.5 and 6.5, giving (6.5 + 6.5 × 2) ÷ 3 = 6.5.
Same reported band, and the candidates are not in the same position. Script A has two arguable marks, both of them in the task that counts double. Script B has four, spread across both tasks. If you are Script A and one of those two Task 2 sixes was a close call, a remark has something to find. If you are Script B, it does not, and the four sixes are telling you something the remark cannot fix.
Two cautions on the arithmetic. First, where your four criteria average to a quarter — 6, 7, 7, 7 gives 6.75 — stop there. How a quarter resolves into a reported band is not published, so running your own quarters through the weighting produces a number that looks precise and is not. Count the judgements instead of predicting the new band. Second, where a six sits changes what it is worth. Task 1 at 7.0 with Task 2 at 6.5 gives 6.67, reported as 6.5. Swap them — Task 1 at 6.5 with Task 2 at 7.0 — and the same two task bands give 6.83, reported as 7.0. A borderline mark in Task 2 carries twice the weight of the same mark in Task 1. The full weighting and what it does to the boundary is worked through in how the IELTS writing band is calculated.
The count you have just done answers this as well.
Four or more of your eight marks at 6. This is a band 6 script with strong patches, not a band 7 script that was marked harshly. A second examiner has no route to a different result, because the evidence is not split anywhere.
The same criterion lowest in both tasks. A criterion that comes out lowest in a Task 1 report and again in a Task 2 essay is a habit rather than an accident of one question — the same missing overview, the same ideas stated and not developed, the same three complex sentences that break. Habits are visible to every examiner who reads the script, which is precisely why a remark will not move them. Where that gap between how an essay looks and how it is marked comes from is set out in my essay looks band 7 but scored 6.
Nothing on the page to find. This is the hard one. A remark re-reads what you wrote; it cannot credit an overview you did not write, a position you never stated outright, or a second part of the prompt you only mentioned. If your honest reconstruction says a required element is absent, no reading of that script produces the mark.
The opposite case is worth naming too. If your count comes out at one or two sixes, both in Task 2, and you can point to the specific paragraph where the evidence for each was genuinely mixed, then the script is doing what a boundary script does, and you are deciding between a remark and sitting the test again from scratch. Note that a retake is not a guaranteed step up either — scores move in both directions, for reasons set out in why an IELTS writing score drops on a retake.
This article will not tell you which to pick. It gives you the count; the cost of each option, which this article deliberately does not quote, is the other half of the decision and only you have it.
The weak point in everything above is that you are marking your own writing, and self-marking runs generous — you know what you meant, so you read the intention rather than the sentence, and a paragraph that only gestures at an idea reads to you as if it develops one.
That is the part worth outsourcing before you spend anything. Paste an essay below and you get a band for each of the four criteria separately, with the sentences each judgement is tied to. Use something written under test conditions, not a draft you polished — a polished essay gives you a reading of your editing.
Then read it the way this decision needs it read. Do not start with the overall estimate; start by counting how many criteria came out a band below the rest, and check the sentences behind each of those. Run a second essay through and see whether the same criterion is lowest again. One low criterion on one essay is a boundary signal. The same low criterion twice is a habit, and a habit is a writing problem rather than a marking one.
Is an IELTS writing remark worth it if I got 6.5?
It depends on whether your script was on a band boundary. A remark is a second certified examiner marking the same script against the same public descriptors, so it can only change marks that were genuinely arguable. Mark your own writing on the four criteria for both tasks and count how many came out a band below the rest — one or two, in Task 2, is a boundary script; four or more is not.
Can a remark give me a lower score?
The rules on what happens to your original result, who is eligible and by when are set by the centre where you booked the test, and they change. This article does not quote them for that reason — ask your centre directly before you apply, rather than relying on forum answers.
What are the chances a remark changes my band?
No official success rate is published for IELTS writing remarks, and the figures that circulate online are not sourced. That is exactly why the useful thing to estimate is your own script: how many of your eight criterion marks were close calls, and whether they sat in Task 1 or in Task 2.
Why would two examiners give the same essay different bands?
Examiners place a script at the band it fits best, and a real essay usually shows evidence from two adjacent bands at once. Where that evidence is split down the middle for a criterion, two trained readers can reasonably land on adjacent bands. Where it is not split, they agree.
Does one criterion at band 6 keep my writing at 6.5?
It can, because the four criteria are weighted equally inside each task and one mark a band below the others pulls the task average down by a quarter. It matters more in Task 2, which counts double in the reported band.
Is the band estimate on this page official?
No. It is an estimate made against the public band descriptors. Only a certified examiner produces an official band, and only your test centre can act on your result.
Descriptor wording referred to here is from the public Writing band descriptors published by IELTS and the British Council.
Bandly grades your essay or letter against all four criteria and shows the gap to your target band, sentence by sentence.
The IELTS Test Report Form gives one Writing band, not four. What the report leaves out, why a remark will not recover it, and how to get the breakdown.
ChatGPT often scores IELTS essays higher than an examiner. Why it happens, which criterion the gap opens on, and three prompts that work better.