Guides / Scoring and band descriptors

Band 8 in IELTS Writing Task 1: The Eleven Mark Combinations, and the Word That Separates 7 From 8

Most pages about band 8 in IELTS Writing Task 1 tell you to write a clear overview and to learn a list of trend words. The published band descriptors say something different. The overview is named at band 7 and in the bands below it. At band 8 the word is illustrates, and that is a different instruction.

A band 8 in Task 1 is also not a single standard of writing. It is the average of four separate marks, and eleven different combinations of those four marks report 8.0. Only one of the eleven has no mark below 8.

This page gives those eleven combinations, quotes the band 7 and band 8 wording side by side, then shows a full answer where every key feature carries the figure that proves it.

The eleven combinations that report 8.0

Your Task 1 score comes from four criteria, each marked as a whole band: Task Achievement, Coherence and Cohesion, Lexical Resource, and Grammatical Range and Accuracy. The four are averaged, and the average is rounded to the nearest half band, with a .25 average rounding up.

I counted the combinations myself. Every one of the 6,561 ways to give four whole band marks from 1 to 9 was passed through that rounding rule, and the ones reporting exactly 8.0 were kept. Ninety one of them report 8.0. Sorting each set from high to low and removing the duplicates leaves eleven distinct mark profiles.

Profile, highest mark firstAverageReported
8 / 8 / 8 / 88.008.0
8 / 8 / 8 / 77.758.0
9 / 8 / 8 / 78.008.0
9 / 8 / 7 / 77.758.0
9 / 8 / 8 / 67.758.0
9 / 9 / 7 / 78.008.0
9 / 9 / 7 / 67.758.0
9 / 9 / 8 / 68.008.0
9 / 9 / 8 / 57.758.0
9 / 9 / 9 / 58.008.0
9 / 9 / 9 / 47.758.0

Two facts come out of that table.

First, ten of the eleven profiles contain a mark below 8, and six of them contain a 6 or lower. A reported 8.0 does not mean four band 8 marks. It usually means a mixture.

Second, the arithmetic is lopsided. The same count gives twenty profiles that report 7.5, eleven that report 8.0, and only five that report 8.5. The band gets narrower as you climb, which is why the last half band costs more work than the one before it. The same count for the top of the scale is in the post on what band 9 in Task 1 actually requires.

What band 8 asks for, in the published wording

Here are the two Task Achievement entries for Academic Task 1, quoted from the public version of the band descriptors.

Band 7:

covers the requirements of the task presents a clear overview of main trends, differences or stages clearly presents and highlights key features/bullet points but could be more fully extended

Band 8:

covers all requirements of the task sufficiently presents, highlights and illustrates key features/bullet points clearly and appropriately

Read those twice. The overview appears at band 7 and disappears at band 8. That is not because band 8 answers skip the overview. It is because by band 8 the overview is assumed, the way a driving test does not award points for starting the engine. Writing a clean overview is how you stop losing marks at 6 and 7; it is not how you gain the eighth band. The mechanics of that sentence are covered separately in the Task 1 overview post.

What is new at band 8 is one verb: illustrates. Band 7 says you present and highlight key features but "could be more fully extended". Band 8 says you illustrate them. In a chart report, illustration means the number. You name the feature, then you attach the figure that proves it, and you do it for every feature you claim.

That is the single most common gap in answers that otherwise read well. A sentence like "the Science Centre grew dramatically" highlights a feature. A sentence like "the Science Centre more than doubled, from 260,000 to 540,000" illustrates it. Both are accurate. Only one of them is doing the band 8 job.

The question

A practice Academic Task 1 question. The figures are invented for this exercise, so please do not quote them as real data.

The bar chart below shows the number of visitors, in thousands, to four museums in one city in 2015 and in 2022. Summarise the information by selecting and reporting the main features, and make comparisons where relevant. Write at least 150 words.
Museum20152022
City History Museum420310
Science Centre260540
National Art Gallery380395
Maritime Museum15090

An answer written to the band 8 wording, 204 words

The bar chart compares visitor numbers at four museums in a single city in 2015 and in 2022, with all figures given in thousands. Overall, the Science Centre was the only venue to grow substantially over the period, and it replaced the City History Museum as the busiest of the four. Two museums lost visitors, while the National Art Gallery was almost unchanged and had become the second most visited site by 2022. The sharpest movement was at the Science Centre, which more than doubled from 260,000 to 540,000. Its gain of 280,000 visitors was more than three times the Maritime Museum's entire audience in 2022. The two falls were steeper in proportional terms than the gallery's rise. The City History Museum, the most popular site in 2015 at 420,000, fell by 110,000 to 310,000, and the Maritime Museum lost 40 per cent of its audience, dropping from 150,000 to just 90,000. The National Art Gallery was the stable case, edging up from 380,000 to 395,000, an increase of under 4 per cent. By 2022 the distance between the busiest and the quietest venue had therefore widened considerably, from 270,000 to 450,000.

Count the figures in that answer. Every claim carries one. "More than doubled" is followed by 260,000 and 540,000. "Lost 40 per cent" is followed by 150,000 and 90,000. "Almost unchanged" is followed by both values and the percentage. The last sentence turns two numbers the chart does not state, the opening gap and the closing gap, into a comparison the reader could not make without doing the subtraction themselves. That is illustration, and it is also what the task means by "make comparisons where relevant".

The four marks, and where this answer could still lose one

Task Achievement. All requirements are covered: a summary, selected main features, comparisons. Every feature named is illustrated with a figure. The overview comes second, not last, and makes three separate claims rather than one vague one.

Coherence and Cohesion. The band 8 bullets ask that information be sequenced logically, that all aspects of cohesion be managed well, and that paragraphing be used sufficiently and appropriately. The answer groups by behaviour, not by museum: one paragraph for the rise, one for the falls and the flat case. Grouping by behaviour is what lets a reader see the pattern. Four paragraphs, one per museum, is the most common way a well written Task 1 answer stalls at 6 or 7, because the reader has to build the comparison themselves.

Lexical Resource. Band 8 asks for a wide range used "fluently and flexibly to convey precise meanings" and for skilful use of uncommon items, accepting occasional inaccuracy in word choice and collocation. The answer varies the verbs for movement: grew, doubled, fell, dropping, edging up, widened. Note that edging up is doing precise work, not decorative work. A rise of under 4 per cent is not a surge, and calling it one would be a word choice error, not a vocabulary flourish.

Grammatical Range and Accuracy. Band 8 asks for a wide range of structures with the majority of sentences error free. The answer uses a relative clause, two non finite clauses, a comparative structure, and a passive. A memorised template cannot do that, which is why the two language criteria are usually what caps a strong answer. Across the eight marked Task 1 answers published in Bandly's own library, Lexical Resource was the lowest or joint lowest of the four estimates on every one of them, which is the pattern you can look through on the band 8 samples hub and across the Task 1 Academic samples.

The weakest point in the answer above is its length. At 204 words it is comfortably over the minimum, but the Maritime Museum's decline gets one clause while the Science Centre gets a whole paragraph. If an examiner judged that the smallest venue was a key feature in its own right, "covers all requirements of the task sufficiently" would be the bullet under pressure.

How this compares with 7 and 6

The useful exercise is not reading one band 8 answer. It is reading the same question answered at three levels and seeing which sentences changed. A band 7 Task 1 answer marked in full usually has the right structure and vaguer language. A band 6 answer usually has the right content and visible grammar errors. Neither of them is short of ideas. They are short of precision.

What to do today

Take any Task 1 answer you have already written. Go through it sentence by sentence and underline every claim about the data. Then check whether each underlined claim is followed by a figure. Rewrite the ones that are not, and say the number out loud as you write it.

That one pass is the cheapest route from the band 7 wording to the band 8 wording, because it attacks the exact difference between them. You can then paste the rewritten answer into Bandly for a band estimate on each of the four criteria, which will tell you which of the eleven profiles above your answer currently looks like.

Find out what this essay would score

Paste your own answer and Bandly returns an estimated band for each of the four criteria, with the sentences that pulled the score down. Under a cent an essay, about six seconds.

Check my essay

Keep reading