Band 5.5 IELTS essay: the 4 mistakes that capped the score
A band 5.5 essay scored 5/6/5/6, plus the exact sentence holding each criterion down and the minimum fix that clears the band 6 line.
A band 7.0 essay and a band 7.5 essay on the same prompt. The half band is two criterion boxes, Coherence and Grammar — not a better essay overall.
The difference between band 7 vs 7.5 in IELTS Writing is not that the 7.5 essay is generally better written. It is that exactly two of the four criteria went up by one whole point — because an examiner awards a whole number on each criterion, and 7.5 is what four whole numbers average to when two of them read 7 and two read 8.
That arithmetic is the subject of this article. Below are two Task 2 essays on one prompt, both written from scratch for this article along with the prompt; nothing here comes from a past paper or another site, and no official IELTS answer exists at any band. The first is a 7.0 — 7 for Task Response, 7 for Coherence and Cohesion, 7 for Lexical Resource, 7 for Grammatical Range and Accuracy. The second is a 7.5 — 7, 8, 7, 8. Position, ideas and vocabulary are held constant. Only two boxes move.
Four criteria, each a whole number, averaged to give the band for that task.
Nothing was ever scored 7.5; there is no such thing as 7.5 for Coherence and Cohesion. It exists only as the average of four integers, so the useful question is not "how do I write half a band better?" but "which two boxes am I closest to moving?"
Notice what a single box is worth. Had Coherence and Cohesion gone up while Grammatical Range and Accuracy stayed at 7, the numbers would be 7 + 8 + 7 + 7 = 29 ÷ 4 = 7.25 — not a figure that appears on a report form, and the rule for resolving it is not published, so I will not guess. The consequence is enough: one improved criterion is no reliable route to a half band; two are.
One constraint before the essays. Task 2 requires at least 250 words — a floor, not a target. Writing 251 earns nothing; writing 230 gives the examiner a reason to hold Task Response down. The essays below run to 269 and 284 words. Throughout this article the criteria are described in my own words rather than quoted from the published descriptors.
Prompt, written for this article: In many older cities, local councils own land that can be used either to restore historic buildings or to build new housing. Some people believe historic buildings should always be protected. Others believe housing needs must come first. Discuss both views and give your own opinion.
Local councils in older cities often control land that could be used either to restore historic buildings or to build new homes. Some people think that heritage must be protected in every case, and other people think that housing is more urgent. In my opinion, housing should normally come first, although a small number of buildings deserve full protection.
Firstly, there are good reasons to preserve old buildings. Old buildings give a city its character, and tourists come to see them. Moreover, if a historic street is demolished, it cannot be rebuilt in the same form, so the loss is permanent. A city that keeps its old quarter also keeps a record of how ordinary people lived there, and schools and museums can use this record in their teaching. Many residents therefore oppose to demolition, even when they accept that new homes are needed.
However, the housing shortage is a more immediate problem. Rents in large cities have risen faster than wages in recent years, and young workers are forced to live far from their jobs. Restoration is also expensive. The money that is spent on repairing one stone facade could pay for several flats. Furthermore, many heritage buildings stand empty during years, while families are waiting for accommodation. This is why I believe councils should give the priority to housing.
In conclusion, both views have merit, because historic buildings are valuable and people also need somewhere to live. My own view is that councils should protect only the buildings that are genuinely irreplaceable, and that the remaining land should be released for new homes as quickly as possible.
Why each box reads 7:
Same prompt, same argument in the same order, same four ideas, almost the same content words.
Local councils in older cities often control land that could be used either to restore historic buildings or to build new homes. One group argues that heritage must be protected in every case; the other argues that housing is more urgent. The second position is closer to my own, although I would protect a small number of buildings completely.
The case for preservation rests on the fact that loss is permanent. If a historic street is demolished, it cannot be rebuilt in the same form, so a decision taken in one year closes that option for every year afterwards. Old buildings also give a city its character, which is why tourists come to see them, and a city that keeps its old quarter keeps a record of how ordinary people lived there, a record that schools and museums can use in their teaching. Many residents oppose demolition for this reason, even when they accept that new homes are needed.
The housing shortage, however, is the more immediate problem. Because rents in large cities have risen faster than wages in recent years, young workers are forced to live far from their jobs, and every year of delay adds to that distance. Restoration competes directly with this need: the money spent on repairing one stone facade could pay for several flats. What makes the comparison sharper is that many heritage buildings stand empty for years while families wait for accommodation. Priority, in my view, should go to housing.
Both views have merit, and the choice is not between valuing history and valuing people. It is a question of proportion. Councils should protect the buildings that are genuinely irreplaceable and release the remaining land for new homes.
Why two boxes moved and two did not:
| Band 7.0 version | Band 7.5 version | What moved |
|---|---|---|
| Firstly, there are good reasons to preserve old buildings. | The case for preservation rests on the fact that loss is permanent. | CC — the topic sentence carries the claim, not a label |
| Many residents therefore oppose to demolition | Many residents oppose demolition for this reason | GRA — error removed; CC — points back to something specific |
| Rents have risen faster than wages, and young workers are forced to live far from their jobs. | Because rents have risen faster than wages, young workers are forced to live far from their jobs, and every year of delay adds to that distance. | GRA — fronting replaces coordination; CC — that distance starts a chain |
| Restoration is also expensive. The money that is spent on… | Restoration competes directly with this need: the money spent on… | CC — colon subordinates example to claim; GRA — reduced relative |
| Furthermore, many heritage buildings stand empty during years | What makes the comparison sharper is that many heritage buildings stand empty for years | GRA — cleft, preposition fixed; CC — weight signalled, not linked |
| This is why I believe councils should give the priority to housing. | Priority, in my view, should go to housing. | GRA — fronting; article slip gone |
| Position, four supporting ideas, paragraph count, content vocabulary | Identical | TR and LR unchanged, by design |
Your Writing band is not your Task 2 score. Task 2 counts double: (Task 1 + Task 2 × 2) ÷ 3, rounded. Two candidates, both lifting Task 2 from 7.0 to 7.5:
Candidate A — Task 1 already at 7.0.
Candidate B — Task 1 sitting at 6.5.
Candidate B is common, and the Task 2 work there was real and invisible. If your Task 1 is half a band behind, the two boxes worth hunting are probably in Task 1 — the shorter, more mechanical task, where the same effort buys more. Both examples land exactly on a half band by design; I have avoided any averaging .25 or .75, since how those resolve is not public.
Stop asking whether your writing is good enough for 7.5. Ask which two of your four boxes are closest to tipping, and work on those.
Bandly marks your essay on all four criteria and returns four whole numbers, each tied to the sentences that produced it — so you can see which two boxes are one rewrite away.
What is the real difference between 7 and 7.5?
Two criterion boxes going up by one point each. 7 + 7 + 7 + 7 = 28 ÷ 4 = 7.0; 7 + 8 + 7 + 8 = 30 ÷ 4 = 7.5. Nothing was ever scored 7.5.
Can Coherence and Cohesion be a 7.5?
No. Every criterion is a whole number. Half bands come from the average only.
What if just one criterion improves?
You get 29 ÷ 4 = 7.25, which is not a figure on the report form, and the rule for resolving it is not published. The practical point stands: two boxes, not one.
Which boxes are the cheapest to move?
Coherence and Cohesion, and Grammatical Range and Accuracy — both respond to rewriting. Lexical Resource is slow, and Task Response moves only when your thinking does.
Does a better Task 2 always raise my Writing band?
No. With Task 1 at 6.5, lifting Task 2 from 7.0 to 7.5 still reports 7.0. The rounding eats it, and the repair work belongs in Task 1.
Official test format and assessment information is at ielts.org.
Bandly grades your essay or letter against all four criteria and shows the gap to your target band, sentence by sentence.
A band 5.5 essay scored 5/6/5/6, plus the exact sentence holding each criterion down and the minimum fix that clears the band 6 line.
The same IELTS Task 2 essay at band 6 and band 7, with every edit tagged to the criterion it moved — and the averaging trap that keeps you at 6.5.