How long it takes to improve your IELTS score: data, not stories

There is no fixed answer, and the stories you read online make it look faster than it usually is. IELTS itself says that in two classroom studies, learners took "on average" three months "to improve by up to half a band". Practice data points the same way: a higher mock score often shows up within a week, but a higher average that holds on new tests is much rarer, and rarer still the higher you start.

Below: what IELTS and Cambridge publish, practice data from over 9,000 learners, checkpoints for week 4, 8 and 12, and how to tell a real rise from a lucky day.

So where does "200 hours per band" come from? A 2021 study of UK university language policy by William Pearson traces it. The 1989 IELTS Handbook said one band took "between 100 and over 200 hours". The 2002 handbook said "up to 200 hours", and Pearson reports that this was the last time the figure appeared in official IELTS material. Research by Anthony Green in 2004 and 2005 concluded that the formula could no longer be supported. If a course promises one band in 200 hours, ask where that number comes from, along with the other questions to ask before you pay for an IELTS course.

Among histories that reached it, the typical wait was about a week, and in at least half of them it was the very next test. A higher score within a week can reflect learning, getting used to the format, or ordinary variation between tests. One result can't tell you which.

Retests show why it matters. Among 345 learners who took two full mocks between 28 July and 26 September 2026, 43% scored higher on the second, 27% the same and 31% lower. The median gap between attempts was four days. On their own, 30% of next Writing attempts (354 of 1,178) and 29% of next Speaking attempts (1,108 of 3,842) within two weeks were lower, in the same period. So one higher mock is a good sign, not proof.

Some histories never showed that higher score. Among those we could follow for at least 30 days, 12% starting at 4.0 to 5.0 did not record a score half a band higher during the observation window, compared with 25% at 5.5 to 6.0, 37% at 6.5 and 47% at 7.0 to 7.5.

What the table shows:

1. Longer histories showed more gains. At 5.5 to 6.0, the share whose average rose half a band was 32% among histories shorter than 3 weeks and 45% among those of 12 weeks or more. These are different people, so it doesn't prove that the extra weeks caused the rise.
2. Higher starters rose less often. From 6.5, 3 in 10 long histories rose half a band. From 7.0, the average barely moved.
3. Part of the pattern is statistics. Learners whose first two tests were unusually low tend to score higher later anyway, and the reverse happens at the top. That makes low starters look faster and high starters slower.

People who keep practising for three months may also be more motivated. These are patterns, not a forecast. Keeping tests below 4.0 in changes these shares by up to 4 points, not the pattern.

If your gap is in Writing or Speaking, give it the most time, and start there. For Writing, the guide to moving from 6.5 to 7 shows which criteria usually hold essays back.

More tests went with bigger rises. In the same group, the average rose half a band in 29% of skill histories with 4 to 5 different tests (130 of 441), 38% with 6 to 9 (196 of 512) and 48% with 10 or more (467 of 976). The pattern held in each skill separately. It doesn't prove that more tests cause the rise: people who test more may also study more in other ways. It does fit what IELTS reports, that the biggest gains came with more study hours.

Counting "best vs first", as below, is the kindest way to measure: with several mocks, one of them is likely to be a good day. That's why the checkpoints below use two tests in a row.

The same 6.0 to 6.5 row, skill by skill:

Use these checkpoints to review your plan, not as a forecast. Even among people still practising at week 12, about 1 in 3 starting at 6.0 had reached 6.5 on two tests in a row, and in Writing far fewer. If you're behind at week 4 or 8, you're in the majority. Change what you practise before you simply add more weeks:

• Listening and Reading: full timed tests, then go through every wrong answer and find the question type that costs you most.
• Writing: fix the criterion that holds you back; see what changes from 6.5 to 7.
• Speaking: record full tests and compare them with the band descriptors, not just more practice questions.
• Overall: your weakest skill pulls the overall band down, so it gets the most time.

This is an observation about practice tests. We haven't checked it against real IELTS results. As a practical guide for booking:

• Wait until your last two mocks, on tests you haven't seen, both reach your target. Check the overall band and every skill minimum you need.
• If your date is fixed, keep a fallback. A One Skill Retake must be taken within 60 days, only after IELTS on computer. Check with the centre of your original test that it offers one, and with your organisation that it accepts it. Results take 3 to 5 days, so make sure they would arrive before your deadline.

Dana's pattern is the one to look for: a higher two-test average on new tests, weeks after the start. Whether it lasts, her next tests will show. Aziz and Timur may still be learning, but their scores don't show it yet, and one good score can feel like a new level.