Here's the thing: you can write perfectly grammatical sentences, use sophisticated vocabulary, and organize your response like a pro, but one single number mistake can knock your band score down half a point or more. Not because the examiner is harsh, but because Task Response counts for 25% of your writing mark, and it explicitly rewards accurate data interpretation while penalizing misreported figures.
This is where most students mess up. They scan a graph quickly, jot down the first number they see, and never look back. By the time they realize they've written "sales increased by 40%" when the chart clearly shows 14%, the damage is done.
This guide walks you through exactly how to catch those mistakes before you submit, why they tank your score more than you think, and how to build a personal checking system that works during the exam.
The IELTS band descriptors for Writing Task 1 are explicit. Band 8 requires you to "select, present and appropriately highlight key features and exceptions." Band 6 is "generally accurate selection of key features." Band 5 says "selects some appropriate information but may miss some key features."
Notice something? There's no mention of "getting the numbers roughly right" or "being close enough." You either report the data accurately, or you don't.
When you misreport a number, even accidentally, you're making a false claim about the graph. An examiner reads "unemployment rose to 8.5%" when the graph shows 5.8%, and they immediately know you either didn't read carefully or you can't interpret visual data. Both damage your score.
Here's what makes it worse. One number error compounds. Say you write "sales increased by 40%" when it was actually 14%, then use that figure to support your next sentence about trends. Now you've built incorrect analysis on top of an incorrect fact. Your whole paragraph becomes unreliable.
Weak: "The chart shows that coffee consumption in the UK rose from 2010 to 2015, increasing by roughly 50 million cups annually."
Why this fails: You've invented a unit (cups annually) and a figure (50 million) without checking the axis labels or data points. The examiner knows you guessed.
Good: "Coffee consumption in the UK increased from approximately 4.2 million tonnes in 2010 to 6.8 million tonnes in 2015, a rise of 2.6 million tonnes over the five-year period."
Why this works: You've read the axis labels (tonnes), cited specific figures, and done basic math to show the increase. Every number is verifiable from the graph.
Not all misreporting is the same. Some errors are careless slips; others show you've genuinely misread the axis or confused categories. Knowing which mistakes tend to catch you helps you build a personal checking system.
A graph shows sales in "thousands of pounds" (the y-axis clearly says "Sales (£000s)"). You read 45 on the scale and write "sales reached £45." Wrong. It's £45,000.
This happens because you're reading quickly and your brain auto-corrects to what "sounds normal." Forty-five pounds for sales sounds too low, so you mentally adjust. But the examiner doesn't care what sounds normal. They care what the axis says.
Weak: "Smartphone sales reached 45 units in 2023."
Good: "Smartphone sales reached 45 million units in 2023." (after confirming the axis label says "millions")
A table has columns for three regions: North, South, and East. You scan quickly and pull figures from the South column but attribute them to North. The numbers are real, but they're assigned to the wrong place.
This is particularly common in bar charts with multiple colors or dense tables with lots of rows and columns. Your eye lands on a number, but your brain hasn't registered which row or category it belongs to.
Weak: "In the North region, healthcare spending increased to 32%." (You've actually read the South region's figure.)
Good: "In the North region, healthcare spending increased to 24%." (You've traced the row and column intersection carefully.)
The chart shows 67.4% and you write "approximately 67%." That's fine. But then you write "nearly 70%" in the next sentence, and suddenly you've shifted the figure again. Or you round 19.8% to 20% when it's clearly closer to the lower number.
Rounding itself isn't wrong. Inconsistent rounding is sloppy.
Weak: "Unemployment rose from 4.8% to approximately 7%, representing a rise of around 2.5 percentage points." (You've rounded the starting figure but not the ending figure, and the math doesn't add up.)
Good: "Unemployment rose from approximately 4.8% to 7%, representing a rise of roughly 2.2 percentage points." (All figures are at the same precision level, and the math checks out.)
You have 20 minutes for Task 1. You can't spend 10 of them fact-checking every figure. But you can spend 90 seconds doing a focused accuracy check before you submit.
Here's the process:
This takes about two minutes of your 20. That's 10% of your time, and it protects your Task Response score.
Tip: On exam day, use your pencil to physically point at the number on the graph while you read your sentence. This creates a physical connection between what you've written and what you're checking. It sounds simple, but it catches the mistakes your eyes would normally skip over.
Let's work through a realistic scenario. Say you have a bar chart showing average temperatures across four cities: London, Paris, Berlin, and Madrid, over three years (2019, 2020, 2021). The y-axis shows temperature in degrees Celsius, ranging from 0 to 20.
A weak response might say:
Weak: "All four cities experienced significant temperature increases over the three-year period, with Madrid reaching approximately 25 degrees and London remaining relatively stable at around 10 degrees. The data suggests warming trends across Europe."
The problem: Madrid never reached 25 degrees. The axis only goes to 20. You've invented a fact. An examiner reading this against the actual chart will immediately downgrade you for inaccuracy.
A strong response would say:
Good: "The chart illustrates average temperatures in four European cities from 2019 to 2021. Madrid was the warmest city throughout the period, ranging from approximately 17 degrees in 2019 to 18 degrees by 2021. London remained the coolest, at approximately 9 degrees in 2019 and rising slightly to 10 degrees in 2021. Paris and Berlin showed moderate temperatures between 12 and 15 degrees. Overall, all four cities showed either stable or modest temperature increases over the three-year period."
Why this works: Every number is plausible given a y-axis that ranges to 20. The figures are specific enough to show you've read the graph carefully, but not so specific that you're inventing decimals you can't see. The conclusion (modest increases) is directly supported by the figures you've cited.
Manual checking is essential, and you should practice it regularly. But when you're working at home, using an IELTS writing checker designed to flag numerical errors will speed up your learning.
A quality IELTS essay checker should do more than just catch grammar. It should prompt you to verify that numbers in your response make sense given the source data, or at least alert you to figures that stand out as potentially wrong.
Some tools can't do this automatically because they don't see your original image. But better writing correction tools will nudge you with questions like: "You've written that Figure X increased by 45%. Have you verified this calculation?" It's a prompt, not an answer, but prompts catch mistakes.
When you're writing Task 1 practice essays, get into the habit of pasting both your response and the key figures from the graph into a checker. You'll quickly develop intuition for which numbers are at risk and which are solid.
Both Band 7 and Band 8 responses for Task 1 are generally accurate with numbers. The difference is that Band 8 also "appropriately highlights key features," which means you're not just getting numbers right. You're choosing the most important ones to emphasize and connect them strategically.
A Band 7 might write: "Sales in Q1 were £2.3 million, Q2 were £2.8 million, and Q3 were £3.1 million."
A Band 8 might write: "Sales rose progressively from £2.3 million in Q1 to £3.1 million in Q3, representing a 35% increase. Q2 marked a turning point, when growth accelerated from 21% in Q1 to 35% by Q3."
Both are accurate. Band 8 just selects and connects the numbers more strategically. Start by getting accuracy right. Then worry about selection and emphasis.
Tip: Before you write a new Task 1 response, spend 60 seconds reading the graph and jotting down 5-7 key figures on scrap paper. Use these notes as your reference while you write. This reduces the temptation to glance at the graph and guess numbers.
Before your next practice test, try this exercise. Take any IELTS Task 1 graph. Write your response normally. Then, before you'd normally move on, spend exactly 90 seconds doing the four-step accuracy check above.
Do this five times. By the fifth time, you'll have built a habit. On exam day, that 90 seconds will feel automatic. You won't have to think about it. Your hand will automatically reach for the pencil and start checking.
This habit alone can protect 0.5 to 1 full band point, depending on how many number errors you were making before.
Task 1 is about accuracy. Task 2 is about argument strength. They're different skills, but the checking mindset carries over. If you're working on both tasks and want to improve overall, focus on strengthening your Task 1 overview paragraph first. Once that's locked in, the rest of your response will naturally be more accurate.
For additional support, try using an IELTS writing evaluator to practice both Task 1 and Task 2 under realistic conditions. Getting feedback on data accuracy in graphs now will help you when you move to IELTS essay topics that require numerical evidence.
Get instant feedback on accuracy, grammar, and band score. Our free IELTS writing checker highlights misreported figures and helps you build better checking habits.
Check My Essay Free