IELTS Task 2 Topics That Keep Coming Back

Most candidates prepare for Task 2 by collecting questions. They end up with a list of fifty prompts and no more confidence than when they started. The list is not the problem. The problem is that a list tells you what was asked and nothing about what to say.

Task 2 questions do repeat, but they repeat in themes rather than in wording. Nine themes cover almost every question in circulation. Prepare the themes and your vocabulary carries from one question to the next. Prepare the questions and nothing carries.

Task 2 is the second writing task in both Academic and General Training. You write at least 250 words in about 40 minutes, and it counts for twice as much as Task 1.

Why a topic list on its own is not preparation

A topic list gives you the noun. "Education." "Environment." "Technology." A Task 2 question never asks about a noun. It asks you to take a position on a claim about that noun, and the claim is where the marks live.

Look at three real question shapes on the same theme:

Some people think children should start school at the age of four. Others believe seven is better. Discuss both views and give your own opinion.

Many schools now teach coding instead of a second language. Is this a positive or a negative development?

University education should be free for all students. To what extent do you agree or disagree?

All three are education. None of them shares an argument with the others. The first is about child development, the second about curriculum priorities, the third about public spending. A candidate who memorised "education vocabulary" has words like curriculum and enrolment ready and still has nothing to say in the second body paragraph.

What actually transfers is one level up: the kind of reasoning each theme rewards, and the ten or twelve precise phrases that reasoning needs.

The nine themes and what each one asks you to argue

Read the middle column again. That is the part worth memorising. Almost every environment question in the last decade can be answered by arguing that individual choices matter but only regulation changes behaviour at scale, or the reverse. The examples change. The reasoning does not.

The vocabulary that transfers, and the vocabulary that does not

Lexical Resource, one of the four marking criteria, rewards the accurate and precise use of words. It does not reward rarity on its own. A long word used in the wrong collocation scores below a common word used correctly.

So the phrases worth learning per theme are the ones that are specific and hard to paraphrase:

1. Environment. "carbon emissions", "single use plastic", "renewable energy sources", "environmental degradation". These are the ordinary technical names for the things being discussed. They are not decoration.
2. Work. "job security", "work life balance", "the gig economy", "in the long term". Again, plain and exact.
3. Government spending. "allocate public funds to", "at the taxpayer's expense", "a better return on investment".

Compare that with the phrases candidates usually memorise: "it is a double edged sword", "every coin has two sides", "in a nutshell", "last but not least". These say nothing about the topic. They appear in essays on every theme, which is exactly why they earn nothing. A marker who reads the same opening sentence in a hundred essays learns nothing about this candidate's English from it.

One question, worked through the theme

Take the technology question above: schools teaching coding instead of a second language, positive or negative development.

The theme says: argue what is gained set against what is displaced. That gives you both body paragraphs before you have thought about coding at all.

Paragraph one, what is gained:

Coding is now a basic form of literacy in most technical professions, and students who learn it early enter the labour market with a skill employers pay for directly.

Paragraph two, what is displaced:

A second language, however, develops something coding does not. Learning one requires a student to understand how another culture organises meaning, and that capacity does not become obsolete when a programming language does.

Neither sentence uses a memorised phrase. Both are specific to the question. The structure came from the theme; the content came from thirty seconds of thinking inside that structure.

What this does not fix

Theme preparation gets you ideas. It does not get you the other three criteria.

Task Response still depends on answering the exact question asked. If the prompt says "to what extent do you agree", a balanced essay with no stated position loses marks no matter how good the ideas are. There is more on this in our guide to to what extent do you agree questions.

Coherence and Cohesion still depends on paragraphing and on linking that is not mechanical. Grammatical Range and Accuracy still depends on sentence variety and on error density. Ideas do not rescue any of those.

What to do this week

Pick three themes, not thirty questions. For each one, write a single sentence stating the argument the theme usually wants, in your own words. Then find two real questions on that theme and check whether your sentence works for both. If it does, you have something that transfers. If it only works for one, your sentence is still about the question rather than the theme.

Then write one full answer, timed, on the theme you find hardest. You can paste it into Bandly for an estimated band on each of the four criteria and the sentences behind each estimate, which is usually a faster way to find the gap than reading another topic list. The bands are machine estimates rather than examiner marks, so treat the reasoning as the useful part.