WiderAI › Blog › Task Response: The Criterion Most Band 6 Essays Fail
Task Response: The Criterion Most Band 6 Essays Fail
Task Response is the criterion that measures how well your Task 2 essay actually answers the question: whether you address every part of it, take a clear position, and develop your ideas with enough depth to convince a reader. It carries a quarter of your Task 2 score, and it is the criterion most likely to hold a fluent writer at Band 6 — because it is the one that has nothing to do with how good your English is.
Here is the uncomfortable truth teachers see constantly: a candidate with Band 7 grammar and Band 7 vocabulary writes a general essay about the topic instead of a specific answer to the question, and Task Response drags the whole score down. The examiner is not asking "can this person write English?" under this criterion. They are asking "did this person answer what I asked, fully, with a position I can find and ideas that go somewhere?"
The good news is that Task Response is the fastest criterion to improve, because it is fixed by thinking and planning, not by months of language study. Three habits move it: answer every part of the question, develop fewer ideas further, and keep your position visible from the first paragraph to the last.
What does Task Response actually measure?
Strip the official language back and Task Response comes down to three questions the examiner asks while reading. First, coverage: does the essay address all parts of the task? Many prompts have two or more parts — "discuss both views and give your own opinion" contains three demands, and skipping any one of them limits the score no matter how well the rest is written. Second, position: can I tell what this writer thinks, and does that view stay consistent? Third, development: are the ideas extended and supported, or just announced and abandoned?
Notice what is missing from that list: sophisticated vocabulary, complex grammar, impressive length. Those belong to other criteria. An essay in plain, accurate English that fully answers the question outscores an ornate essay that half-answers it — on this criterion, every time.
Why do so many Band 6 essays fail here?
Band 6 on Task Response typically means the examiner could follow the essay but found it only partly did the job: some parts of the question got more attention than others, the position wandered or stayed vague, or ideas were relevant but thin. Three failure patterns account for most of it.
Answering the topic instead of the question
The prompt "Some people believe children should begin learning a foreign language at primary school. To what extent do you agree?" is not an invitation to write everything you know about language learning. It asks for a judgement about primary-school children specifically. An essay full of true, relevant-sounding sentences about the benefits of English still fails to respond to the task if it never engages with the age question. Examiners call this tangential response, and it is far more common than an off-topic essay.
The invisible position
"To what extent do you agree?" requires an extent. "This essay will discuss both sides" is not a position; it is a refusal to take one. Nor does a position have to be extreme — "largely agree, with one reservation" is a perfectly strong stance, often stronger than total agreement because it is easier to develop honestly. What kills the score is an essay where the examiner reaches the conclusion still unsure what the writer thinks, or where the conclusion contradicts the introduction.
Ideas announced, never developed
The classic underdeveloped paragraph lists three reasons in three sentences and moves on. Each idea is relevant, none is developed, and the paragraph reads like an outline. Development means taking one idea and pushing it: why is it true, what follows from it, what would it look like in practice? One idea explained across five sentences beats three ideas mentioned in one sentence each.
How do you develop an idea properly?
Use a simple chain for each body paragraph: claim, reason, consequence, example. State the idea, explain the mechanism behind it, show what results from it, then ground it in a concrete case. Take the claim "remote work benefits employers". The reason: it widens the pool of people a company can hire beyond one city. The consequence: positions are filled faster and with better-matched candidates. The example: a software company in Tashkent can hire a specialist in Samarkand without either side relocating. Four sentences, one idea, real depth.
Examples do not need statistics or famous cases — inventing figures actually reads badly. A plausible, concrete scenario from ordinary life carries the same weight and is much easier to write accurately. What matters is that the example genuinely illustrates the claim rather than sitting next to it decoratively.
How do you keep your position consistent?
State your position in the introduction in one clear sentence, in your own words. Then make each body paragraph visibly serve that position: a topic sentence that connects to your stance, not just to the topic. If the task asks you to discuss both views, you can present the view you reject fairly — but signal that you reject it ("Those who favour X argue that… However, this overlooks…") so the reader never mistakes explanation for endorsement.
The conclusion should restate the position and briefly say why, without introducing new ideas. A new argument in the conclusion is undeveloped by definition, and it lands in the worst place: the examiner's final impression. If a good idea occurs to you late, either work it into a body paragraph or leave it out.
One more trap: memorised introductions. Examiners recognise rehearsed openings that could sit on top of any essay ("In the contemporary epoch, this issue has sparked heated debate…"), and generic material earns nothing under Task Response because it responds to nothing. Every sentence should be one that only makes sense for this question.
Task Response has a direct cousin in the speaking test: Part 3 answers are judged on whether you address the question asked and extend your ideas, not just on fluency. Practising extended answers aloud — for instance in an IELTS speaking mock test — trains exactly the claim-reason-example habit that essays need. The other three writing criteria are covered in their own guides on the WiderAI blog.
WiderAI gives you AI-scored IELTS mock tests with instant band scores and examiner-style feedback.
Start practising freeFrequently asked questions
What is the difference between Task Response and Task Achievement?
Task Response is the first criterion for Writing Task 2 (essays); Task Achievement is its equivalent for Task 1 (letters in General Training, report writing in Academic). Both measure how completely and appropriately you did the task set, but essays are judged on position and argument development, while Task 1 is judged on covering the required content in the right format and tone.
Can I score well if I only discuss one side of an agree/disagree question?
Yes — a strong one-sided essay is entirely legitimate for "to what extent do you agree" questions, provided your position is clear and well supported. What you cannot skip are explicitly required parts: if the task says "discuss both views", both views must appear. Read the instruction, not just the topic statement.
Does writing more than 250 words improve Task Response?
Not by itself. Length is only penalised below the 250-word minimum; beyond it, extra words help only if they add development. A 270-word essay with two deep, well-supported ideas routinely outscores a 350-word essay of listed, unexplained points — and the shorter essay leaves more time for checking grammar.
How can I check my own Task Response before the exam?
After writing a practice essay, cover it and answer three questions from memory: what did the question ask (all parts), what is my position, and what are my two main supports? Then check the essay against your answers. Any part of the question your essay cannot account for is exactly what an examiner would find, and it tells you what to fix in the next attempt.