Where a Writing band actually comes from
Writing is two tasks, and the band is decided on four criteria across both: task achievement, coherence and cohesion, lexical resource, and grammatical range and accuracy. Each is a quarter of the score.
A candidate who improves at grammar can move one criterion by maybe half a band. A candidate who fixes how the answer is structured moves task achievement, coherence, and lexical resource at once, because those three all depend on organisation.
So the improvements below are roughly in order of how much band they are worth.
1. Learn what the descriptors say
Read the four public band descriptors and count how many of their phrases you could not have written yourself.
This is first because it is free and it changes everything after it. Most candidates are aiming at a description of IELTS writing they formed in their teens, not the one the examiner is applying. Band 7 in the descriptor is not "good English". It is a specific set of things, including "uses a sufficient range of vocabulary without obvious attempts to use uncommon items" and "uses a variety of complex structures". Those are checkable, and a candidate who has read them knows what to write towards.
2. Time your two tasks properly
Twenty minutes for Task 1, forty for Task 2, of the sixty you have. This is not a suggestion, and most candidates under-time Task 2 badly.
The usual failure: Task 1 overruns because the chart is confusing, Task 2 gets rushed, and a rushed Task 2 is the one that costs the band, because it is where coherence suffers. Decide the split before you start, and stick to it even if Task 1 is awkward. An awkward chart is still describable in twenty minutes.
3. Write the overview paragraph for Task 1
It is required, it is usually missing, and it carries marks on its own.
One or two sentences on the most striking feature of the whole chart, with no numbers in it. Find the grouping the chart is inviting before you write anything else, because the overview sentence and the paragraph order both come from that.
4. Spend the first five minutes planning Task 2
Five minutes of planning, four points, is worth more than two extra sentences written without a plan.
Write your four ideas down as keywords before you write a sentence. If you cannot find four distinct points in five minutes, you do not have a Task 2, you have a two-point essay, and stretching it to fit a word count produces the repetition that coherence is marked down for.
5. Keep an error log
This is the habit that separates candidates who improve from candidates who keep writing answers.
One file. Every time you write something you are not sure of, note it. Once a week, look up the rules for this week's entries and write three correct sentences for each.
Over a month you get a list of around fifteen items that are genuinely yours. It is a much shorter list than a grammar book, and it is built from the errors you actually make rather than the ones a textbook assumes.
6. Get your work marked by a person
Writing is judged against descriptors by a person. Self-assessment does not work, because the errors are exactly the ones you cannot see while you are making them.
One properly marked answer is worth five written and not marked. This is the item on the list with the highest cost and the highest return, and it is the one most candidates try to avoid.
If a tutor is not affordable, a study partner who scored well is a genuine option. What is not useful is a language tool, which marks grammar and ignores the two criteria worth half the score.
7. Read the question, every time
The most basic discipline and the most commonly skipped. The question states exactly what is required, and candidates frequently answer a nearby question instead.
The phrase "discuss both" means both. "To what extent" means state your position, then concede the opposite position seriously rather than mentioning it. "Do the advantages outweigh the disadvantages" needs a verdict, not a list.
Underlining the instruction and the two or three required parts of the answer, in the question itself, takes thirty seconds and is the cheapest way to protect the task-achievement criterion.
A two-week plan
| Day | Work |
|---|---|
| 1-2 | Read the descriptors, write one Task 1 and one Task 2, mark them against the criteria |
| 3-6 | One timed Task 1 and one timed Task 2, with a full plan written first |
| 7 | Mark everything. Find your two weakest criteria. |
| 8-12 | Work only on those two criteria |
| 13-14 | Two full timed tests, reviewed for errors and for criteria |
The point of the last fortnight is that you are not practising more. You are practising one specific thing, which is the only way a band moves on a criterion you are weak on.
The thing that is not on this list
Vocabulary size. It matters, and it is worth building, but it is the slowest of these seven to move a band, because the descriptors reward a *range* that is used accurately, not a large list that is remembered. Adding three words a week for a year changes your writing far less than fixing how your answers are organised, and it changes it much less than getting your work marked.
π Visit Step Ahead Institute
Ready to start your IELTS journey? Visit us today:
- Address: 519B, Darshan Lane, Model Town, opp. Guru Nanak Girls College, Yamunanagar, Haryana 135001
- Call: 78766-66588 | 74171-13474 | 94677-77196
- WhatsApp: Chat with us
- Timings: Mon-Thu: 10 AM - 4 PM | Fri-Sat: 10 AM - 2 PM
- Get Directions: Google Maps
Learn directly from our 8.0 Band certified trainer - teaching since 2009. Small batches, personal attention, and student practice portals.