Is your IELTS Task 1 answer a data dump, or does it actually compare?

Your IELTS Task 1 answer is a data dump if most of its sentences read figures out one at a time, and it actually compares if most of its sentences relate those figures to each other. The dividing line is not how many numbers you use — it is what each sentence does with them.

That distinction matters because it is written into the marking. The public Task Achievement descriptors describe the band 5 answer as one in which "the recounting of detail is mainly mechanical", while band 7 requires that "the data are appropriately categorised, and main trends or differences are identified" (IELTS Writing band descriptors, ielts.org). Below are two answers to the same chart, a check you can run on your own answer in two minutes, and three ways to repair the sentences that fail it.

Same chart, same figures. One answer reads the bars aloud; the other says how they relate.

The line is what each sentence does, not how many numbers it has

Every Academic Task 1 prompt ends with the same instruction: summarise the information by selecting and reporting the main features, and make comparisons where relevant. Most candidates read that as two separate jobs — report the numbers, then add a comparison or two. Markers read it as one job. Selecting the main features is comparing, because a feature only counts as "main" relative to the others.

So the useful question about any sentence in your answer is not "is this accurate?" but "what is this sentence for?" There are two possibilities:

• It reads. The sentence takes one item on the chart and states its value, or its value at two points in time. "Rail was used by 14% of commuters in 2000." Nothing in the sentence connects that figure to any other part of the chart.
• It compares. The sentence puts two or more items, periods or groups in relation to each other — higher, faster, overtook, unchanged while others moved, together, by contrast. The figures in it are evidence for that relation.

A reading sentence is not wrong; it is accurate, and accuracy counts. But an answer built mostly from reading sentences leaves the comparison for the examiner to do. Band 6 already expects that "a relevant overview is attempted" and "information is appropriately selected and supported using figures/data". Band 7 adds categorisation and identified differences. A run of reading sentences supplies figures/data and nothing else.

This does not mean you need a minimum number of comparisons — no such count exists — or that you should cut figures. The band 6 line asks for figures; it just does not want them standing on their own.

Two answers to the same chart

The chart below is invented for this article. The numbers are not real statistics and should not be quoted as such. It is a bar chart showing the percentage of commuters in one city using four modes of transport, 2000 and 2020.

Mode	2000	2020	Change
Car	58%	41%	−17
Bus	22%	19%	−3
Rail	14%	22%	+8
Bicycle	6%	18%	+12

Both answers below share the same introduction ("The bar chart shows the percentage of commuters in one city who used four types of transport in 2000 and 2020") and both are trimmed to the overview and body. A complete answer would also need to clear the 150-word minimum.

Answer A

(1) Overall, the car was the most popular type of transport in both years. (2) In 2000, 58% of commuters travelled to work by car. (3) By 2020, this figure was 41%, a fall of 17 percentage points. (4) The bus accounted for 22% of commuters in 2000 and 19% in 2020. (5) Rail was used by 14% of commuters in 2000. (6) In 2020, it was used by 22%. (7) The bicycle was used by 6% of commuters in 2000. (8) This rose to 18% in 2020.

Answer B

(1) Overall, the car remained the most common way to commute, but its share shrank considerably as cycling and rail gained ground. (2) The largest change was the fall in car use, from 58% to 41%, a drop of 17 percentage points that no other mode came close to. (3) Cycling grew fastest, tripling from 6% to 18%. (4) Rail use also rose, from 14% to 22%. (5) Together, these two modes gained 20 percentage points, matching the combined losses of the car and the bus. (6) The bus, by contrast, was almost unchanged, slipping from 22% to 19%, so by 2020 rail had overtaken it as the second most common choice.

The two answers are close in length, and Answer A quotes nine figures to Answer B's ten. Every figure in both is correct, which is why the candidate who wrote Answer A would see no mistake in it.

Here is how the difference breaks down across the four criteria:

• Task Achievement. This is where the gap is widest. Answer A's overview is a true statement, but it describes one item rather than the chart's movement, and the body that follows recounts each mode in turn. Answer B's overview names the main trend and the main difference; its body ranks the changes (sentence 2), groups the gainers (sentence 5) and identifies the one mode that behaved differently (sentence 6). That is the band 7 wording — categorised data, identified differences — happening sentence by sentence.
• Coherence and Cohesion. Answer A is sequenced by the chart's legend: car, bus, rail, bicycle, two sentences each. Answer B is sequenced by importance, and its referencing ("these two modes", "by contrast", "so") carries the logic between sentences.
• Lexical Resource. Answer A rotates "was used by", "accounted for" and "travelled by". Answer B's verbs carry meaning: shrank, gained ground, tripled, slipping, overtaken.
• Grammatical Range and Accuracy. Both are accurate. Answer B uses more varied structures — a participle clause, a result clause with "so" — but this is the smallest of the four differences. Answer A is not being held back by its grammar.

Is my Task 1 answer a data dump? The sentence-by-sentence check

You can run this on your own answer without a teacher. Take your overview and body paragraphs (leave out the introduction, which paraphrases the question and is not meant to compare anything) and put a letter next to every sentence:

• C — the sentence relates two or more data points to each other: a ranking, a grouping, a contrast, a change in order, a shared pattern.
• R — the sentence restates values for one item, even if it gives two dates or calculates the change.

Then count. If R is more than half, the answer reads as a data dump. This is a working rule for self-checking, not an official threshold — the descriptors do not count sentences. But it tracks the descriptor wording closely: an answer that is mostly R is an answer in which the recounting of detail is mainly mechanical.

Applied to the two answers above:

Sentence	Answer A	Answer B
1	C — compares car with the others	C — trend plus main difference
2	R	C — ranks the changes
3	R — calculates, but compares nothing	C — "fastest" ranks it
4	R	R
5	R	C — groups two modes
6	R	C — contrast and change of rank
7	R	—
8	R	—
Total	1 C, 7 R	5 C, 1 R

Three things are worth noticing. Answer A sentence 3 calculates the 17-point drop and still counts as R, because the calculation stays inside one series — arithmetic is not comparison. Answer B keeps one plain R sentence, and sentence 5 immediately puts it to use. And Answer A's overview is a C, but one comparison at the top does not change what the other seven sentences are doing.

Be strict with sentences that merely sound comparative. "This rose to 18%" points back to the previous sentence but compares nothing. Ask what the sentence would lose if every other bar on the chart were deleted. If the answer is "nothing", it is an R.

Three ways to turn an R sentence into a C sentence

Most dumped answers do not need new information. The comparisons are already in the figures the candidate has written down; they just have not been stated. Three repairs cover almost every case, and each one keeps the figures.

Ranking, grouping and contrasting. The figures stay; the sentence gains a job.

1. Rank. Two reading sentences about separate items become one sentence that orders them.

Before: "Rail rose from 14% to 22%. The bicycle rose from 6% to 18%."

After: "Cycling grew fastest, tripling from 6% to 18%, while rail's rise from 14% to 22% was smaller but enough to change the rankings."

The figures are identical. What changed is the word "fastest" and the second clause, which together tell the examiner you have looked at both bars and decided which one matters more.

2. Group. Several items that behave the same way become one category.

Before: "Car use fell from 58% to 41%. Bus use fell from 22% to 19%. Rail use rose from 14% to 22%. Bicycle use rose from 6% to 18%."

After: "The four modes split into two groups: the car and the bus lost share, while rail and cycling gained it."

This is categorisation in the descriptor's sense. It also frees your body paragraphs: once the groups are named, each paragraph can take one group and add the figures, instead of walking through the legend.

3. Contrast. The item that did not follow the pattern is singled out against the ones that did.

Before: "The bus accounted for 22% of commuters in 2000 and 19% in 2020."

After: "While every other mode changed substantially, the bus barely moved, slipping only from 22% to 19%."

The before sentence is accurate and unremarkable. The after sentence identifies a difference — precisely what band 7 asks for — and makes the small change interesting rather than easy to skip.

None of these repairs needs advanced vocabulary. They need a decision about the data before you write, so the most reliable fix is upstream: spend a minute marking on the question paper what ranks, what groups and what stands out. If you cannot name those three things, you do not yet have an overview either — the overview sentence examples show the same decision from the overview's side.

Check which side your answer falls on

The C and R check is quick, but it has a weak point: it relies on you labelling your own sentences honestly, and most candidates find their own reading sentences sound more comparative than they are. "Also", "this" and "meanwhile" make an R sentence feel connected even when it compares nothing.

If you are not sure which way your answer leans, paste it below. Bandly marks it against all four criteria, and the Task Achievement feedback points to the specific sentences that recount detail without comparing it.

If Task Achievement comes back as your lowest criterion, the repairs above are where your next point is. If it is level with the other three, comparison is not your bottleneck.

FAQ

How many comparisons does a Task 1 answer need?

There is no official number. The descriptors ask for a clear overview, sensibly categorised data and identified trends or differences. What matters is whether most of your sentences relate figures to each other.

Is it wrong to include a lot of numbers?

No. The band 6 wording asks for information supported using figures and data. The problem is a sentence whose only job is to read figures aloud, not the figures themselves.

Is a data dump the same as having no overview?

They usually travel together, but you can write a passable overview and still dump data in every body sentence — as Answer A does above.

Does every sentence have to compare something?

No. The introduction compares nothing, and an occasional reporting sentence is normal. The check is about the balance of the overview and body.

Can comparing too much make an answer worse?

Only if the comparisons stop being accurate or go beyond the data. Comparing what is on the chart is what the task instruction asks for.

Related reading

• The IELTS Task 1 overview sentence: 10 examples scored — the same comparing decision, compressed into one sentence
• IELTS Task 1 Academic band 7 sample: bar chart — a full answer that groups and ranks its data
• IELTS Task 1 table band 7 sample — the chart type that most tempts candidates to list
• Task Achievement vs Task Response — why Task 1 is marked on a different criterion

The full public band descriptors are published on ielts.org.