How OET Writing is scored: the 6 criteria

When you submit an OET Writing letter, it is not read once and given a gut-feel grade. Trained assessors score it independently against a fixed rubric of six criteria. Five of those criteria are scored 0–7; one is scored 0–3. Those raw marks are combined — using a scaling OET keeps proprietary — into the 0–500 numeric score and the band letter that lands on your Statement of Results.

The single most useful thing to understand about this rubric is that the criteria are graded separately. There is no "overall impression" mark that lets a beautifully written letter paper over a missing purpose, and no single slip that tanks an otherwise strong letter. You can lose a band on one criterion while scoring near the top on the other five. Knowing which criterion is leaking marks is the whole game.

This post walks through all six criteria in plain English — what each one actually rewards, how the 0–7 and 0–3 scales behave, and the three places we most often see candidates quietly lose a band. If you have not yet read the OET Writing format guide, start there; this post assumes you know what a referral letter is and how the 45-minute sub-test is structured.

The shape of the score

Here is the rubric at a glance:

| Criterion | Scale | What it asks | |---|---|---| | Purpose | 0–3 | Is it immediately clear why this letter was sent? | | Content | 0–7 | Is the right clinical information there, accurately? | | Conciseness & Clarity | 0–7 | Is every sentence relevant and well-summarised? | | Genre & Style | 0–7 | Does it read like a clinical letter to this reader? | | Organisation & Layout | 0–7 | Does it flow, paragraph, and lay out like a letter? | | Language | 0–7 | Grammar, vocabulary, spelling, punctuation, cohesion |

Five criteria sit on a 0–7 band; Purpose sits on a tighter 0–3 band because it is a narrower, more binary judgement — either the reader knows why they received the letter within the first sentence or two, or they do not.

Notice what is not on this list: there is no "length" criterion, no "word count" criterion, and no penalty band for going over or under a target. This is the most misunderstood part of OET Writing, so it gets its own section below.

Purpose (0–3): say why you are writing, first

Purpose is the criterion candidates most often assume they have nailed and most often have not. It asks one question: does the opening of the letter make the reason for writing immediately clear, and is the right reader being addressed?

A high Purpose score comes from an opening sentence that states, in one breath, why the letter exists and what the recipient is being asked to do or be aware of. "I am writing to refer Mr Daniel Okafor, a 58-year-old man with newly diagnosed type 2 diabetes, for ongoing management and dietary education." The reader knows, before they have read anything else, what this letter is for.

A low Purpose score comes from burying the reason. Candidates open with the patient's full history, or a date of admission, or a paragraph of background, and only arrive at the point of the letter halfway down. The clinical content might be perfect — but the recipient had to dig for the purpose, and the criterion penalises that. On a 0–3 scale, the gap between a 3 and a 1 is often a single sentence in the wrong place.

The fix is mechanical: the patient's age, history, and dates belong in sentence two onward. Sentence one is reserved for the purpose and the action. Compare these two openings for the same case:

Buried purpose: "Mrs Hannah Reed is a 72-year-old woman who was admitted on 3 March with a fractured neck of femur. She underwent a hemiarthroplasty on 4 March and her post-operative recovery was uncomplicated. She has a history of osteoporosis and hypertension…"

Clear purpose: "I am writing to refer Mrs Hannah Reed, a 72-year-old woman recovering from a hemiarthroplasty, for community physiotherapy and fall-prevention support following her discharge home today. She was admitted on 3 March with a fractured neck of femur…"

Both letters might end up containing identical clinical detail. The second one scores higher on Purpose because the reader knows, at the end of the first sentence, who the patient is, why they are being written about, and what is being asked. The first makes them wait.

Content (0–7): the right facts, accurate, relevant

Content asks whether you have included the clinical information the reader needs — no more, no less — and whether it is accurate. This is the criterion most directly tied to the case notes. The assessor is checking your letter against the source material for two failure modes: omission (you left out something the recipient needs) and error (you stated something the notes do not support, or stated it wrong).

A strong Content score selects. It pulls the details that matter for this reader and this purpose — current medications and doses, relevant results, the functional state at discharge, the specific concern being referred — and leaves out the rest. A weak Content score either transcribes the entire case file regardless of relevance, or misses a detail that changes the recipient's decision (the allergy, the red-flag result, the follow-up that has already been booked).

Content is also where factual fidelity is judged. Inventing a value, mis-transcribing a dose, or stating a diagnosis the notes only raise as a query will cost marks here even if the letter reads beautifully. When in doubt, stay faithful to the case notes — the assessor has them open beside your letter.

Conciseness & Clarity (0–7): why length is never the penalty

This is the criterion every candidate worries about for the wrong reason. OET does not penalise letter length on its own. There is no word-count band, no "you wrote 230 words so you lose a mark" deduction. A long letter that is entirely relevant and well-summarised can still score 7 on this criterion. A short letter that is vague or omits the summarising work can score low.

What Conciseness & Clarity actually rewards is two things:

1. Relevance — does every sentence earn its place, or are there red herrings, repeated facts, and background the reader does not need? Length only becomes a problem here when it carries irrelevance.
2. Summarising — have you condensed and synthesised the case notes into clinical prose, or have you copied them across as a list? Good summarising turns five lines of observations into one sentence that captures the trend.

So the honest guidance is: do not pad, and do not write to hit a number. Most strong OET letters land around 180–200 words simply because that is what it takes to cover the content concisely — but that is a consequence of good summarising, not a target you are scored against. Write what the purpose needs, summarise hard, cut the red herrings, and this criterion takes care of itself. (Our Writing scorer encodes exactly this rule: it judges length only through relevance and summarising quality, never with a mechanical word-count deduction.)

Genre & Style (0–7): does it read like a clinical letter?

Genre & Style asks whether the letter reads like a piece of professional clinical correspondence written to this specific reader. It is the criterion that separates a clinical handover from a general essay about a patient.

The reader matters. A letter to a specialist you are referring to is pitched differently from a letter to a community nurse continuing care, which is different again from a letter to a patient's GP. Register, the level of explanation, and how much you assume the reader already knows are all part of this criterion. Explaining what an ECG is to a cardiologist reads wrong; omitting context a non-specialist reader needs reads wrong too.

Genre & Style also covers the conventions that mark the letter as a letter: an appropriate salutation, a professional tone throughout (not chatty, not telegraphic), and a correct sign-off. "Dear Dr Patel" with "Yours sincerely"; "Dear Doctor" — when you do not have a name — with "Yours faithfully." These feel small, but they are scored consistently, and getting them wrong signals to the assessor that the genre has not been internalised.

Organisation & Layout (0–7): structure the reader can follow

Organisation & Layout is about flow and presentation. Does the letter move in a logical order — purpose, then relevant background, then current status, then the request or action — or does it jump around? Are related ideas grouped into paragraphs, or is it one undifferentiated block? Do the connections between paragraphs make sense, so the reader is carried from one idea to the next?

A strong letter here has a clear beginning, middle, and end, with paragraphing that mirrors the clinical logic. The opening states purpose; the body builds the picture the reader needs in a sensible sequence; the close states the action and signs off. A weak letter buries the request in the middle, mixes background with current status, or runs everything together so the reader has to reconstruct the structure themselves.

This criterion overlaps with the four structural patterns we describe in OET Writing for nurses: 4 referral patterns — handover, discharge home, GP follow-up, and specialist referral. Recognising which pattern a case calls for is, in effect, pre-loading a strong Organisation & Layout score, because each pattern is a tested paragraph order.

Language (0–7): grammar, vocabulary, and the things candidates over-index on

Language is grammar, vocabulary, spelling, punctuation, and cohesion. It is the criterion most candidates assume is the most important — and it is genuinely important — but it is one of six, not the whole score.

A strong Language score shows control: accurate tenses (past for what was done, present for ongoing, future only for the recipient's actions), precise clinical vocabulary, correct spelling of drug names and conditions, and sentences that connect with appropriate linking. A weaker score shows recurring grammar slips, imprecise word choice, or punctuation that makes the meaning work to extract.

Here is the part worth internalising: a letter with a couple of minor grammar slips that nails Purpose, Content, and Genre will usually outscore a grammatically flawless letter that buries its purpose or copies the case notes wholesale. Language matters, but candidates who pour all their preparation into grammar drills and none into purpose and relevance are optimising the wrong criterion. The same lesson holds on the Speaking side — see how OET Speaking is actually scored for the parallel.

The band letter is indicative, the criteria are real

OET reports a band letter (A to E) and a 0–500 numeric score for Writing, the same as the other sub-tests. The exact arithmetic that turns six criterion marks into that band is OET's own and is not published. That means any tool — ours included — that shows you a band letter is giving you an indicative band, calibrated against the public rubric, not a reconstruction of OET's internal scaling.

What this means in practice: do not chase the band letter. Chase the criteria. The band is downstream of six independent judgements, and the only way to move it is to find the one or two criteria that are leaking marks and fix those. A candidate stuck at Band C who discovers their Purpose and Conciseness scores are dragging two strong-Language letters down can lift a whole band by changing where the purpose sentence sits and cutting the red herrings — without touching their grammar at all.

The three places candidates quietly lose a band

Across the OET Writing letters we score, the same three leaks come up again and again:

1. The purpose is buried. The single most common avoidable loss. The clinical content is there, the grammar is fine, but the first sentence describes the patient instead of stating why the letter was sent. Purpose is only 0–3, but on a sub-test where every criterion counts, losing two of those three points pulls the whole score down. Fix: purpose in sentence one, history in sentence two.

2. Length is treated as the enemy — or the goal. Candidates either pad to hit an imagined word count (importing red herrings that hurt Conciseness & Clarity) or slash so hard they omit content the reader needs (hurting Content). Neither is what the rubric rewards. Fix: write what the purpose requires, summarise the case notes hard, and let the length be whatever relevance demands.

3. The letter ignores its reader. A discharge letter written as if to a specialist, or a specialist referral written as if to the patient, loses on Genre & Style and often on Content too (wrong details for the wrong reader). Fix: read the task instructions to identify the recipient before you start, and pitch register, explanation, and detail to that person.

How to use the rubric in practice

The rubric is a diagnostic tool, not just a grading sheet. The most effective Writing preparation we see is criterion-by-criterion: write a letter, get a per-criterion breakdown, find the lowest criterion, and do the next letter targeting that one specific thing. Two letters later you check whether that criterion moved, then attack the next-lowest. This is far faster than re-writing whole letters hoping the overall "feel" improves.

That per-criterion loop is exactly what the OET Live Writing scorer is built for: every letter you submit comes back marked on all six criteria, with inline annotations showing the phrases that earned or cost marks, and band-5 and band-7 model answers for the same task so you can see what the next band up looks like concretely. Pair that with the four referral patterns and you have both halves of the skill — the structure that pre-loads Organisation and Genre, and the per-attempt calibration that tells you which criterion to fix next.

The candidates who lift their Writing band fastest are not the ones who write the most letters. They are the ones who know which of the six criteria is costing them marks on each letter — and write the next one to fix it.