I get 7 in Task 1 but 6.5 in Task 2 — here's exactly why
Scoring 7 in Task 1 but 6.5 in Task 2 is the usual direction, not a fluke. The two tasks differ on one criterion — here is where the half band goes.
An error-free essay can still score 6. Accuracy is half of one criterion out of four — here is what your correct sentences earn under each of them.
An essay with no grammar mistakes can score band 6. Accuracy is half of one criterion out of four, and the other three — Task Response, Coherence and Cohesion, Lexical Resource — never ask whether your sentences are correct.
That is the gap behind a low IELTS writing score with no grammar mistakes anywhere in the script. Error-checking is the one form of self-marking every candidate knows how to do, so it is the one that gets done, and it reports back that the essay is fine. What follows is what error-free writing is worth under each of the four criteria, an error-free paragraph marked at band 6, and four passes to run over your own essay instead.

Nothing in the script is wrong. Three of the four criteria are still at 6.
Examiners do not subtract marks for errors. Each criterion is awarded the band whose descriptor is the best fit for the script, so a clean essay does not accumulate credit — it avoids the wording at band 5 and below that describes frequent errors. Avoiding that wording is not the same as meeting the wording at band 7.
The arithmetic makes the point faster than the theory. Within a task the four criteria are weighted equally. Suppose your grammar is genuinely strong and scores 8, while Task Response, Coherence and Cohesion and Lexical Resource each sit at 6. The average is (8 + 6 + 6 + 6) / 4 = 6.5. You can have the cleanest script in the room and still report a 6.5, because three quarters of the mark was never about cleanliness.
It compounds across the two tasks, too. IELTS publishes that "The assessment of Task 2 carries more weight in marking than Task 1", so an immaculate Task 1 cannot rescue a Task 2 that is correct but undeveloped.
Which leaves the real problem: grammar is the only one of the four you can check without knowing the descriptors. The rest of this article makes the other three visible.
Even inside grammar, "no mistakes" only answers half of the name.
Band 6: "A mix of simple and complex sentence forms is used but flexibility is limited. Examples of more complex structures are not marked by the same level of accuracy as in simple structures."
Band 7: "A variety of complex structures is used with some flexibility and accuracy. Grammar and punctuation are generally well controlled, and error-free sentences are frequent."
The safest way to write with no errors is to write only what you are certain of: short sentences, one clause each, the same three or four patterns repeated down the page. It also produces a script that the band 5 line describes precisely — "The range of structures is limited and rather repetitive" — while containing nothing anyone could mark as wrong.
Band 7 asks for both halves at once: a variety of complex structures, used with accuracy. Range with errors is the band 6 line, where "more complex structures are not marked by the same level of accuracy as in simple structures". Accuracy without range sits below it. Only both together reach 7, and a candidate trained out of every risky sentence has optimised one half at the cost of the other.
Punctuation belongs to this criterion too, not a separate one, so comma splices and missing commas around subordinate clauses are counted here.
How to check. Count the sentences in your last essay that contain a subordinate clause — one opening with although, which, because, while, if, or a participle. If well under a third of your sentences do, your range is the problem and no amount of further proofreading will find it. The four structures that reliably count are set out in complex sentences for band 7.
Task Response. Band 6: "The main parts of the prompt are addressed (though some may be more fully covered than others). A position is presented that is directly relevant to the prompt, although the conclusions drawn may be unclear, unjustified or repetitive." Band 7: "The main parts of the prompt are appropriately addressed. A clear and developed position is presented. Main ideas are extended and supported."
Nothing in either line concerns correctness. The words separating them are developed, extended and supported, and a flawlessly written paragraph that states a claim, gives an example and restates the claim satisfies none of the three. Task 2 conclusions that reach band 7 scores several endings against the "repetitive" half of that band 6 line.
Coherence and Cohesion. Band 6: "Information and ideas are generally arranged coherently and there is a clear overall progression. Cohesive devices are used to some good effect but cohesion within and/or between sentences may be faulty or mechanical due to misuse, overuse or omission." Band 7: "Information and ideas are logically organised, and there is a clear progression throughout the response. A range of cohesive devices including reference and substitution is used flexibly but with some inaccuracies or some over/under use."
Mechanical is the operative word, and an error check cannot see it. There is nothing incorrect about Firstly, Moreover and In conclusion at the head of consecutive paragraphs; they are correctly spelled, correctly punctuated and correctly used. They are also labels announcing that a paragraph has started rather than links carrying a relationship, which is what the band 6 line calls mechanical. Note too what band 7 names and band 6 does not: reference and substitution — this, such, the latter, doing so. Neither is a connector, so adding connectors cannot supply them. Linking words are keeping you at band 6 takes that descriptor apart in full.
Lexical Resource. Band 6: "The resource is generally adequate and appropriate for the task. The meaning is generally clear in spite of a rather restricted range or a lack of precision in word choice." Band 7: "The resource is sufficient to allow some flexibility and precision. There is some ability to use less common and/or idiomatic items. An awareness of style and collocation is evident, though inappropriacies occur."
"A rather restricted range" and "a lack of precision" describe absences, not mistakes. An essay assembled from important, problem, good, many people and affect passes every check you can run on it and stops at 6, because band 7 asks for the right word rather than a serviceable one.
Three of the four criteria, then, are invisible to the only check most candidates know how to run.

The same paragraph, equally accurate in both versions, developed in only one of them.
Take a prompt of the familiar shape: Some people think governments should spend money on public transport rather than on building new roads. To what extent do you agree or disagree? Here is a body paragraph with no grammar mistakes in it.
Public transport is very important for modern cities. Many people use buses and trains every day to go to work. If the government spends money on public transport, more people will use it. For example, in my country the government built a new metro line and many people started to use it. Therefore, I think spending money on public transport is a good idea.
Every sentence is correct. Marked criterion by criterion:
Four sixes, no mistakes, band 6. Now the same idea written for band 7:
Cities that invest in buses and metro lines are buying capacity rather than space. A new road absorbs the traffic it attracts and returns to congestion within a few years, whereas the same money spent on a metro corridor moves several times as many people along the same route. That is the case for prioritising public transport, and it survives the obvious objection: even where car ownership is rising, the binding constraint in a dense city is not how many vehicles people own but how much room there is to move them.
The accuracy is identical — neither version contains an error. What changed is everything the other three criteria measure. Task Response: the claim is carried two steps past itself, first to a mechanism and then through an objection. Coherence and Cohesion: whereas, That and it carry the joins, which is reference and substitution rather than labelling. Lexical Resource: absorbs, capacity, corridor, constraint are precise rather than impressive, which is what "flexibility and precision" asks for. Grammatical Range and Accuracy: the complex structures are longer and still controlled, so the criterion moves on range without losing accuracy.
Band 6.5 vs band 7 essay runs the same comparison across a full script, and band 6 to band 7: the same essay rewritten tags every edit to the criterion it moved.
Take one essay you wrote under time, not a corrected version, and read it four times. One criterion per pass, one question each.
Do not run a fifth pass looking for mistakes. You have already run that one, and it is what brought you here.
Fix in order of speed. Task Response and Coherence and Cohesion respond to planning decisions taken before you write, so they can move within a week; Lexical Resource and Grammatical Range and Accuracy track your English level and move over months. If a pass in each pair turns up a problem, start with the planning one.
If more than one pass comes back clean and the score still has not moved, the issue may not be spread across the criteria at all: why am I stuck at band 6.5 in IELTS writing sorts the four patterns by how often they turn out to be the single cap. The full band 5 to 8 wording is in IELTS Writing band descriptors explained, and how to get band 7 in IELTS Writing states the four as conditions rather than advice.
My essay has no grammar mistakes. Why is my IELTS writing score still low?
Because grammar is one criterion of four, and accuracy is half of that one. Task Response, Coherence and Cohesion and Lexical Resource do not measure whether your sentences are correct — they measure whether ideas are developed, whether the joining carries information, and whether the word choice is precise. An error-free essay can score 6 on all three.
Do examiners deduct marks for each grammar error?
No. Each criterion is awarded the band whose descriptor best fits the script, so marking is a matching process, not a subtraction. Having no errors means you avoid the descriptor wording about frequent errors; it does not earn credit towards band 7 on its own.
Can I reach band 7 with only simple sentences?
Not in Grammatical Range and Accuracy. Band 7 asks for "a variety of complex structures used with some flexibility and accuracy", so an essay built entirely from short, correct, repeated patterns meets the accuracy half and fails the range half.
Is it better to write safely or to attempt complex sentences?
Attempt them, but only the structures you can control. Range with errors and accuracy without range both sit at band 6; band 7 requires both halves together. Adding a structure you cannot punctuate reliably moves the criterion down, not up.
Which criterion should I check first if my grammar is already clean?
Task Response. It is the criterion most often at 6 in a clean script, because a correct paragraph that states a claim, illustrates it and restates it satisfies nothing in the band 7 line, and it is also the fastest to change, since it depends on planning decisions taken before you write.
All descriptor wording quoted here is from the public Writing band descriptors updated in May 2023, 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.
Scoring 7 in Task 1 but 6.5 in Task 2 is the usual direction, not a fluke. The two tasks differ on one criterion — here is where the half band goes.
A 6.5 is rarely general weakness. It is usually one or two criteria scoring 6 while the rest sit at 7 — here are the four patterns, and how to tell which is yours.