Search "IELTS study plan" and you will find the same skeleton everywhere: week 1, grammar; week 2, vocabulary; week 3, practice tests. Notice what is missing — nobody has measured anything yet. A plan that schedules the cure before running the diagnosis is a calendar, not a plan. IELTS reports four separate skill scores and averages them, so the only question that matters on day one is: which of the four skills — and which criterion inside it — is actually costing you marks?
The failure: allocating weeks by habit, not by gap
Most people study what they like studying, or what feels like "English study" — usually vocabulary lists and grammar drills. But IELTS does not award an overall grammar mark. Listening and Reading are 40 questions marked right or wrong — IDP: "each correct question will be awarded 1 mark". Writing and Speaking are examiner-marked against four named criteria each, weighted equally. General grammar work is one quarter of one criterion set — it can be exactly the wrong place to spend a month.
Why a targeted half-band beats a month of general study
The overall band is pure arithmetic. From the official scoring page:
"The overall band score is the average of the four section band scores rounded to the nearest half band."
"If the average of the four sections ends in .25, the overall band score is rounded up to the next half band, and if it ends in .75, the overall band score is rounded up to the next whole band."
Take test taker C from ielts.org's own worked examples: Listening 6.5, Reading 6.5, Writing 5.5, Speaking 6.0. Average 6.125 — overall 6.0. Now raise the weakest skill, Writing, by half a band to 6.0: the average becomes 6.25, and the .25 rule rounds it up to 6.5. One half-band, in one skill, moved the overall a full half-band.
Arithmetically, any half-band in any skill adds 0.125 to the average — but the weakest skill is usually where the next half-band is cheapest, because that is where the largest, most fixable errors live. Meanwhile a month of general study spread across all four skills can leave every individual skill short of its next half-band boundary — four small improvements, none of them enough to change any reported number. That framing is coaching, not an official rule — but the rounding arithmetic it rests on is quoted above, verbatim.
Week zero: measure
Before you schedule anything, sit one full timed mock — all four skills, real timings, 2 hours and 45 minutes of total test time, no pausing, no dictionary. Score Listening and Reading out of 40. Get your Writing and Speaking attempts assessed against the official criteria, or at minimum record and review them against the public descriptors.
The point is not the overall number. The point is the per-skill split — and, inside Writing and Speaking, the per-criterion split.
Read your result against the official anchors
Listening and Reading have official "average marks" anchors. From ielts.org, marks out of 40:
| Band | Listening | Academic Reading | GT Reading |
|---|---|---|---|
| 8 | 35 | 35 | — |
| 7 | 30 | 30 | 35 |
| 6 | 23 | 23 | 30 |
| 5 | 16 | 15 | 23 |
Carry the official caveat with you: "The precise number of marks needed to achieve these band scores will vary slightly from test version to test version." These anchors tell you what a half-band costs in raw marks. Sitting at 26 in Listening? Band 7 is roughly four more correct answers — a findable number of specific, fixable mistakes, not a vague resolution to "get better at English".
Writing and Speaking have no official raw-to-band conversion at all — nothing on ielts.org, British Council, IDP or Cambridge publishes one. What is official is the criteria. Writing: "Task achievement (for task 1) and task response (for task 2), Coherence and cohesion, Lexical resource, Grammatical range and accuracy" — and "The criteria are weighted equally and the score on the task is the average." Speaking: "Fluency and coherence, Lexical resource, Grammatical range and accuracy, Pronunciation" — again equally weighted. So your Writing diagnosis is not a number out of 40; it is which of the four criteria is dragging the average of the four criterion scores.
Then — and only then — allocate the weeks
Give weeks to skills in proportion to the gap, not in proportion to comfort. A plan built from a measurement might give three weeks to Writing coherence and one maintenance session a week to a Reading score already at target — which no habit-built plan would ever produce.
For the skill-specific failure modes, each skill has its own diagnostic page here: Listening's four failures, Reading is a clock problem, Writing's four criteria and Speaking's four criteria. Re-measure with another full mock partway through and reallocate — a plan that never updates against evidence has stopped being a plan again.
One more reason the per-skill view matters: some receiving organisations set per-skill minimums, not just an overall — for example, canada.ca lists Listening and Speaking thresholds for Canadian citizenship. Check the exact requirements on your receiving organisation's own page before you set targets.
What is not official here
Be clear about what is method and what is rule. Official: the averaging and rounding rules, the Listening/Reading anchor marks with their "vary slightly" caveat, and the four criteria names for Writing and Speaking. Not official: everything structural about this plan — week zero, the allocation logic, the re-measure cadence. That is our coaching, and we label it as such. No official source publishes a raw-to-band table for Writing or Speaking, and none publishes a "study X hours per band" figure we could verify — any page giving you either is inventing it. And no plan, this one included, can promise you a band.
What actually costs you
- Scheduling content ("week 1: grammar") before measuring which skill is weak.
- Studying the skill you enjoy instead of the skill that is furthest from target.
- Treating Writing and Speaking like Listening and Reading — they have no raw score, only four criteria each, so "more practice essays" without a criterion diagnosis repeats the same errors.
- Ignoring the rounding boundaries — an average ending in .25 rounds up, so half a band in one skill can be the whole difference.
- Never re-measuring, so the plan drifts back to habit by week three.
Try one
Week zero starts with a measurement, and the measurement is a full mock.
[Try a full mock test →] — one full task, marked, no card required.