IELTS Reading band score: correct answers table (2026)
IELTS Reading raw score to band for Academic and General Training: how many correct answers you need for band 6, 7, 8, and why one mark can shift your band by 0.5.
Two people sit the same Academic Reading paper. One gets 29 answers right, the other 30. A single mark separates them, and on the band scale they land on opposite sides of a line: 6.5 and 7.0. That one question was worth half a band.
Reading is scored the way Listening is. No examiner, no opinion, no Band Descriptors. Forty questions, one mark each, and a table that turns your raw score into a band. But Reading adds a twist Listening doesn't have: the score you need shifts depending on whether you sit Academic or General Training. Thirty correct answers is a 7.0 on Academic and a 6.0 on General Training. Read the wrong row and you'll aim at the wrong number.
The short answer
Reading has 40 questions, one mark each, and no penalty for a wrong answer, so a blank never beats a guess. Your raw score out of 40 converts to a band on the 0–9 scale. The catch is the conversion itself: Academic and General Training use different cutoffs, because the two papers aren't equally hard. General Training texts are everyday material like notices, adverts, and workplace documents, while Academic passages run denser and more abstract. To balance that, General Training asks for more correct answers to reach the same band. The official IELTS scoring page confirms the format: Reading is marked out of 40, one mark per question.
Both tables are below. As with Listening, the bands aren't evenly spaced, and that uneven spacing is where your strategy lives.
Quick lookup: what band is my raw score?
Sitting Academic and want a fast read on a number? Out of 40, 23 correct is a band 6.0, the floor for a 6. From there, 27 is a 6.5, 30 a 7.0, 33 a 7.5, 35 an 8.0, 37 an 8.5, and 39 a 9.0. So 32/40 is a 7.0, and 38/40 is an 8.5. General Training runs stricter: the same 30/40 is only a 6.0 there, and a 7.0 needs 34. The full tables, and the reason the two papers differ, sit just below. Listening works the same way, 40 questions to a band, though its cutoffs differ slightly (Listening band score).
None of this changed for 2026: Reading is still 40 questions, one mark each, converted on the tables that follow.
The Academic Reading raw-to-band table
This is the most widely cited Academic conversion, reproduced consistently across major preparation sites:
| Band score | Correct answers (out of 40) |
|---|---|
| 9.0 | 39–40 |
| 8.5 | 37–38 |
| 8.0 | 35–36 |
| 7.5 | 33–34 |
| 7.0 | 30–32 |
| 6.5 | 27–29 |
| 6.0 | 23–26 |
| 5.5 | 19–22 |
| 5.0 | 15–18 |
| 4.5 | 13–14 |
| 4.0 | 10–12 |
Treat these as approximate. IELTS doesn't publish a fixed public table and notes that the exact marks needed shift slightly between test versions, so read each cutoff as roughly ±1. What holds steady is the shape: the bands near the top are narrow.
The General Training table is stricter
Same test length, higher bar. General Training Reading also has 40 questions, but because the texts are easier, you need several more correct answers for the same band:
| Band score | Correct answers (out of 40) |
|---|---|
| 9.0 | 40 |
| 8.5 | 39 |
| 8.0 | 38 |
| 7.5 | 36–37 |
| 7.0 | 34–35 |
| 6.5 | 32–33 |
| 6.0 | 30–31 |
| 5.5 | 27–29 |
| 5.0 | 23–26 |
| 4.5 | 19–22 |
| 4.0 | 15–18 |
Line the two tables up and the gap is stark. Thirty correct answers is a 7.0 on Academic but only a 6.0 on General Training, a full band apart for the identical raw score. A 6.0 costs 23 marks on Academic and 30 on General Training. So if you're sitting General Training for a visa or migration application, never borrow an Academic target. Around the 6.0 to 7.0 range, that's aiming four to seven marks too low.
Read the gaps, not just the numbers
Look at how wide each Academic band is. A 6.0 runs from 23 to 26, four marks of room. A 6.5 covers 27 to 29, three marks. Then 7.0 is 30 to 32, and every band above it narrows. The closer you get to the top, the less slack each band gives you.
That's why one answer decides more up there. At 29 correct on Academic you're a 6.5. Find one more, hit 30, and you're a 7.0, the same half-band jump that one question bought in the opening example. Most competing score guides stop at printing the table and never explain this. The cutoffs aren't a smooth slope; they're steps, and standing one mark below a step is the most expensive place to be. It's the Reading version of the rounding that decides your overall result, which I break down in how IELTS band scores are calculated.
How many you actually need for your target
Turn the tables into a single number you can aim at. On Academic:
- A 6.0 needs 23 correct.
- A 6.5 needs 27.
- A 7.0 needs 30.
- A 7.5 needs 33.
On General Training, the same targets cost more: a 6.0 needs 30, and a 7.0 needs 34.
Now your mock scores mean something. If you keep landing on 27 on Academic practice tests, you're a 6.5 sitting three marks under a 7.0. That isn't "read faster." That's "find three answers," and three answers have a location. They're usually cheaper to recover in Passage 1, which carries the most accessible text, than in Passage 3, where the argument is denser and the question types harder. If 6.5 is the wall you keep hitting, I wrote a whole piece on breaking the 6.5 ceiling.
Reading is a race against the clock
Here's the trap Listening doesn't set. In Listening, the audio paces you. In Reading, you pace yourself, and most lost marks here aren't comprehension failures at all. They're time failures. Sixty minutes, 40 questions, three passages, and no extra transfer time on either format. On the computer test you answer on screen; on paper you write straight onto the answer sheet as you read. There's no separate 10 minutes to copy answers across, the way Listening hands you on the paper version.
That works out to about 90 seconds a question, but spending it evenly is a mistake. A sensible, widely used split is 15 minutes on Passage 1, 20 on Passage 2, 25 on Passage 3, because the passages get harder as you go. Bank time early on the readable text so you still have some when the questions start fighting back.
One habit saves more marks than any other: skip and come back. When a question won't give up its answer in a reasonable stretch, mark it, move on, and return once the easy ones are in. Sitting on a single stubborn matching-headings question for four minutes is how people leave five gettable answers blank at the end. And never hand one in empty. There's no negative marking, so a guess is free.
The marks you lose with the right answer in your head
Some of the most frustrating drops in Reading come from answers you found and then wrote wrong. The conversion table can't see these, but your score feels them.
Spelling is marked. Copy a word out of the passage as your answer and misspell it, and it's wrong, even though the correct word was sitting right there on the page. Reading is as strict as Listening on this.
Word form trips people who understood the text perfectly. If the passage supports "tickets" and you write "ticket," that's a lost mark. Same with the word-count limits: when the instruction says "NO MORE THAN TWO WORDS," three words scores zero even if the right two are buried in there. None of this needs extra reading skill. It needs a checklist you run on every answer before you move on.
Capitalization, on the other hand, isn't scored. The Reading answer sheet isn't case sensitive, so don't waste a second worrying about it. Put that attention into copying the word exactly and counting how many words the instruction allows.
Why there's no Band Descriptor for Reading
Writing and Speaking get judged on the four Band Descriptors: Task Response, Coherence and Cohesion, Lexical Resource, and Grammatical Range and Accuracy. An examiner reads the criteria and assigns a band, which leaves room for interpretation, the kind I get into with the Speaking Band Descriptors. Reading has none of that. You either picked the right answer or you didn't.
That bluntness is good news, because it makes Reading one of the most diagnosable sections you have. A stuck 6.5 is rarely a flat, even 6.5. It's a pattern: you lose four marks on matching headings, two more on True/False/Not Given, and you're clean everywhere else. The fix isn't reading three more articles a day. It's the one or two question types that keep leaking marks.
That's the loop IELTS Drills is built around. Instead of running another full timed paper that re-confirms the same 27, you do short, targeted reps on the exact question type costing you marks, five minutes a day, until that specific leak closes. Reading rewards the approach because, like Listening, it has no Band Descriptor to hide behind.
Frequently asked questions
Is 30 out of 40 a band 7 in Reading? On Academic, yes, 30 correct is a 7.0. On General Training the same 30 is a 6.0, and you'd need 34 for a 7.0. Treat both as approximate, since cutoffs shift slightly between versions.
Why do Academic and General Training have different score tables? Because the General Training texts are easier, so the same band demands more correct answers. The band itself means the same thing on your certificate; only the raw-score route to it differs.
Is there negative marking in IELTS Reading? No. A wrong answer costs nothing, so fill in every question and never leave a blank.
Does spelling count in Reading answers? Yes. A misspelled answer is marked wrong even when you clearly found the right word in the text. Capital letters, though, aren't scored.
Is this conversion table official? Not to the exact mark. IELTS doesn't release a fixed public table and says the marks needed vary slightly by version. Both tables here are the widely reproduced approximations, accurate enough to set targets with.
How does my Reading band affect my overall score? It's averaged with the other three sections and rounded, so a strong Reading band can lift a weaker one across a rounding line. That mechanic is the whole point of how IELTS band scores are calculated.
The bottom line
Reading doesn't reward general effort. It rewards correct answers, in a quantity that depends on which paper you're sitting. Find your real practice score, read it against the right table (Academic or General Training, not a blur of the two), and measure the exact gap to your target. Then go spend your minutes on the question types where those few marks keep slipping away, and check your spelling before you call an answer done.


