8-week OET Speaking plan for working nurses

Most published "OET preparation timelines" you will find on the open web assume you are a full-time student with 25–30 hours of study time per week. That is not the candidate profile we actually meet on this site.

The typical OET Live user is a registered nurse already working three or four shifts a week in their home country, often with a family at home, preparing for AHPRA or NMC registration. Their realistic practice budget is four to five hours per week — and most of it has to happen on a phone, in 20-minute pockets between other obligations.

This roadmap is built for that reality. Total time investment: about 40 hours over 8 weeks, with a clear weekly objective and an honest assessment of what each block is trying to fix.

If you have 4 weeks instead of 8, or 12 weeks instead of 8, scroll to the compressed and extended variants further down. The order of the blocks matters more than the duration of each block.

Before you read this: if you have not yet taken any OET Speaking sub-test, even a mock one, do the Week 1 diagnostic first. The plan will be different depending on what your baseline actually is, and the diagnostic only takes 20 minutes.

The principle behind the order

OET Speaking is scored on nine independent criteria. Four are linguistic (intelligibility, fluency, appropriateness of language, resources of grammar and expression) and five are clinical-communication (empathy, patient perspective, organisation of talk, clinical accuracy of information, vocabulary).

Different criteria respond to practice at different rates:

• Pronunciation and intelligibility improve slowly. Two weeks of daily 10-minute drilling beats one weekend of intensive practice.
• Fluency and filler reduction plateau quickly. You can move a 2 to a 4 in a week if you are deliberate; getting from a 4 to a 5 takes months.
• Clinical communication criteria improve quickly under structured feedback. Most candidates jump 5–10 points on patient perspective and empathy in a single week, once they know what the rubric is actually looking for.
• Grammar under exam stress is largely a function of how much you have rehearsed common clinical phrases. It improves through repetition, not through grammar-book study.

The roadmap below sequences these by improvement curve: slow-moving things first, fast-moving things last. That way, by the time you are doing full role-play mocks (week 7), your pronunciation and fluency baselines have had six weeks to consolidate.

A roadmap that does role-play mocks every week from week one — which is what most casual "OET preparation" advice tells you to do — wastes the early weeks. You score badly, get feedback on the same pronunciation and fluency issues you would have anyway, and never actually fix them because you keep moving on.

Week 1: diagnostic + baseline

Time: 4 hours. Goal: figure out your starting band, identify your two weakest criteria, write them down.

You cannot plan a route without knowing where you are starting.

Day 1: take one full OET Speaking mock (45 min)

Record yourself doing two role-plays cold, with no preparation, using any practice case you can find. The OET Live sample cases work; so do the official samples in the OET preparation pack. The point is to get a recording.

Day 2–3: self-score against the rubric (90 min)

Listen back. Score yourself, criterion by criterion, against the nine-criteria rubric. Be harsh — most candidates over-score themselves by one band.

If you have access to an AI scoring tool (we built one, but any reasonably calibrated one works), run the recordings through it and compare. Where the AI scores you lower than you scored yourself is usually where you have a blind spot.

Day 4: identify your two weakest criteria (30 min)

Write them on paper. Tape it to your study station. Every week of this plan is calibrated to either drill those two, or shore up the criteria that support them.

The most common "weakest two" combinations we see:

• Intelligibility + fluency: classic non-native-speaker pattern, especially for candidates from countries where English education emphasised reading and writing over speaking. Most leverage: deliberate pronunciation drilling.
• Patient perspective + empathy: clinically skilled candidates who treat the OET role-play like a clinical interview. Most leverage: rephrasing exercises that practise acknowledging before informing.
• Organisation of talk + clinical accuracy: candidates who know the medicine but lose track of the role-card tasks. Most leverage: pre-rehearsed transition phrases.
• Vocabulary + resources of grammar: candidates whose English education was clinical but rigid. Most leverage: paraphrasing drills.

The rest of this roadmap assumes you have done this diagnostic and know which two criteria you are targeting. If you skipped it, go back. The plan does not work without that anchor.

Week 2: intelligibility — the highest-leverage week

Time: 5 hours. Goal: nail one consonant cluster, one vowel pair, and three multi-syllable clinical words you keep mis-stressing.

Pronunciation is the criterion that takes the longest to improve. Starting here gives the change six weeks to bed in before the exam.

What to drill

Do not try to "fix your accent". The OET scoring rubric explicitly says any accent is fine if a non-native English listener can follow you without effort. What loses you marks is specific failures:

• Sound substitutions that change meaning (a Tagalog-speaker substituting "p" for "f" in "fever"; a Mandarin-speaker substituting "s" for "th" in "think"). L1-specific drill posts: Tagalog speakers, Mandarin speakers
• Stress on the wrong syllable in multi-syllable words ("hyperTENsion" instead of "hyperTENsion" — yes, that is the same; the wrong version is "HYpertension" or "hypertenSION")
• Running word boundaries together in ways that change meaning ("I'm" + "sorry" merged into "imsorry" with no word boundary)

The drill (do this every day, 30 minutes)

1. Pick one consonant pair from your own recording where you noticed slips. Common ones: l/r, b/v, p/f, th/s.
2. Find a clinical word list with that pair (any pronunciation app will do; we have one built in to OET Live but the Cambridge Dictionary audio works too).
3. Record yourself saying ten words slowly, then at normal pace, then in a sentence. Listen back. Repeat until your version sounds the same as the reference.
4. Then do the same for ten clinical phrases that contain the sound — "no fever or fatigue", "rapid pulse rate", "blood pressure reading".

How to know if it is working

By Friday of week 2, the words you drilled on Monday should come out automatically when you use them in a sentence — without you thinking about them. If they do not, slow down. It is better to nail three words this week than to skim across thirty.

Week 3: fluency and filler reduction

Time: 5 hours. Goal: cut your filler rate (uh, um, you know, basically) by 50%, and stop pausing mid-clause.

Fluency is the second-easiest criterion to move quickly. Most non-native speakers have a filler rate of 4–8 per minute on a clinical topic. The Band A range is 1–2.

The two patterns that kill fluency

Pattern 1: mid-clause pauses. You start a sentence, get halfway, freeze while you search for a word, then either restart or use a filler. Examiners notice.

Pattern 2: clause-end fillers. Every sentence ends with "okay?" or "right?" or "you know". One or two of these is fine — they are valid hedge markers in clinical conversation. Ten in five minutes is not.

The drill

1. Record yourself talking for two minutes on any clinical topic ("explain what a fever is", "why a flu shot matters"). No script.
2. Listen back with a notepad. Tally every filler. Tally every pause longer than two seconds.
3. Re-record the same two-minute monologue, with the goal of cutting the tally in half. You will sound stiff at first. That is fine — that is your nervous system installing the new habit.
4. Do this for a different topic every day.

By end of week 3, your filler rate on a fresh topic (one you have not rehearsed) should be roughly half what it was at the diagnostic.

If your week 1 diagnostic showed fluency was already a strength, swap this week for an extra week on whichever criterion in your "weakest two" you have not addressed yet.

Week 4: empathy and patient perspective

Time: 5 hours. Goal: develop a stock of three acknowledgement phrases and use them automatically whenever a (real or AI) patient expresses worry.

This is the week where clinical candidates often make the biggest single-week jump. The criterion is not "be a nice person to your patient". It is "acknowledge the patient's emotional state before you give clinical information". Two different things.

What examiners are scoring

When the role-card patient says "I'm worried about losing my job because of this", a Band C response is:

"Don't worry, this is a common condition and you'll be fine. Let me explain the treatment plan…"

A Band A response is:

"I can hear that the impact on your work is what's worrying you most — that makes sense, this kind of injury is scary precisely because you don't know how long you'll be out. Let me walk through what we know about recovery times, and then we can talk specifically about your work."

Same clinical content. The Band A version acknowledges the emotion, names it, validates that it is reasonable to feel, then moves to the clinical answer.

The phrase bank

Memorise three acknowledgement phrases. Three is the right number — fewer feels canned, more is hard to deploy under pressure.

Some that work:

• "I can hear that ___ is the part that's worrying you most."
• "That makes sense — most people feel ___ when they're going through this."
• "Before I talk about the medical side, can I just check that I've understood what's most important to you?"

The drill

Do five role-plays this week (about 20 minutes each, plus 10 minutes of listen-back). After each one, count how many times you used an acknowledgement phrase. Aim for at least one per role-play minute.

If your week 1 diagnostic showed empathy was already a strength, do this week anyway — but focus on patient perspective, which is the related criterion that asks whether you elicited the patient's concerns, preferences, and constraints (not just whether you responded warmly).

Week 5: organisation of talk and information giving

Time: 5 hours. Goal: learn three transition phrases that signal structure, and rehearse the four-step "diagnose → explain → plan → check" arc until it is automatic.

Most OET Speaking role-cards have four to five tasks. Examiners are scoring whether you covered all of them, in a logical order, with clear transitions.

The transition phrases that buy you marks

• "Let me walk you through this in three parts. First, what's going on. Second, what we're going to do about it. Third, what you can do at home."
• "Before we move on to the plan, let me just check — what questions do you have about the diagnosis?"
• "I want to make sure we cover one more thing before we run out of time…"

The arc

Almost every OET Speaking case can be structured as:

1. Open the conversation, acknowledge the patient, take history (or review what's already known)
2. Explain the diagnosis or situation in plain English
3. Present the plan — what you are recommending and why
4. Check understanding — invite questions, confirm the patient knows what to do next

This is not a script. It is a scaffold. You will deviate based on the role-card. But if you finish a five-minute role-play and you missed step 2 or step 4, you will lose marks on organisation regardless of how well you did the other steps.

The drill

Do five role-plays this week. After each one, write down which of the four arc steps you covered, and which you skipped. If you skipped step 2 ("explain the diagnosis") three times in a row, the issue is rehearsal: pre-script how you would explain three common conditions in 90 seconds each, and practise them out loud.

Week 6: vocabulary and grammar — paraphrasing under pressure

Time: 5 hours. Goal: paraphrase three clinical terms into plain-English versions automatically, and stop using passive voice in places where active sounds better.

By week 6 your pronunciation, fluency, and clinical-communication framework should be in place. This week is about language texture — the surface of how you sound.

The two micro-skills

Skill 1: clinical → plain English on demand. When you say "hypertension" to a patient and they look blank, you should automatically say "high blood pressure" without breaking stride. Practise this with twenty common clinical terms. Pick a partner (or use an AI tutor) and have them ask "what does that mean?" at random points.

Skill 2: passive → active for warmth. "You will be given a prescription" is bureaucratic. "I'll give you a prescription" is warm. Both are grammatical. The second scores higher on appropriateness and empathy. Hunt your own recordings for passive constructions and rewrite them.

The drill

Take three of your week 5 role-play recordings. Go through them and rewrite three sentences in each: one paraphrase from clinical to plain English, one passive-to-active rewrite, one "I/we" insertion to make the sentence more present.

Then redo the role-play with those rewrites in mind. The point is not memorising the specific rewrites — it is training the underlying instinct.

Week 7: full mocks with feedback

Time: 6 hours. Goal: do five timed, scored, full-length mocks. Use them to find your remaining weaknesses.

This is the only week in the plan that looks like the way most OET prep guides start. We saved it for week 7 because by this point, doing a mock actually reveals real issues — not the same intelligibility and fluency issues you would have found in week 1.

How to run a mock

1. Pick a case you have not seen before. Use the OET Live case bank, the official sample pack, or whatever you have access to.
2. Set a 5-minute timer. Speak as if to a real patient. Record.
3. Use AI scoring (or self-score against the rubric) to get a per-criterion score.
4. Identify the single weakest criterion on this mock.
5. Do one focused 20-minute drill on that criterion before the next mock.

The trap most candidates fall into in week 7 is doing five mocks in a row without targeted drilling between them. The mock surfaces a weakness; the drill addresses it; the next mock confirms whether the drill worked.

Scenario variety

Pick mocks across at least three scenario types:

• Initial assessment (history-taking, diagnosis)
• Discharge / treatment plan explanation
• Emotional or difficult conversation (bad news, non-compliance, anxious patient)

The exam can give you any of these. If you have only practised one type, you are gambling.

Week 8: taper and exam-day plan

Time: 4 hours. Goal: arrive at the exam having done two role-plays in the last three days, with your weakest criterion firmly under control.

Resist the urge to cram in week 8. Five hours of new practice in the final week buys you very little. You are not learning new skills — you are consolidating.

The taper

• Day 1–2: one mock, one drill on the weakest criterion from that mock.
• Day 3–4: one mock with a different scenario type, one drill.
• Day 5–6: rest, or one 20-minute light role-play if you genuinely feel out of practice.
• Day 7 (exam day): warm-up. Read a clinical article aloud for five minutes. Do not do a role-play. Save your nervous system for the actual exam.

What to bring mentally

Most candidates lose marks on exam day not because they did not prepare, but because they drop something they had been doing well in practice. The top three drops:

1. Acknowledgement phrases stop coming out under nerves. Solution: have one phrase memorised so cold you could say it in your sleep. ("I can hear that ___ is the part that's worrying you.") Use it in the first 30 seconds of the role-play.
2. Filler rate spikes. Solution: be aware that this happens, and consciously slow your tempo for the first minute. The slower start anchors you for the rest of the role-play.
3. Patient perspective gets skipped. Solution: in the prep time before the role-play starts, write "ASK CONCERNS" at the top of the candidate card. Every examiner I have spoken to says this is the single most-missed task.

If your exam is in 4 weeks

Compress to:

• Week 1: diagnostic + intelligibility (combine weeks 1 + 2 above)
• Week 2: fluency + empathy (combine weeks 3 + 4)
• Week 3: organisation + vocabulary (combine weeks 5 + 6)
• Week 4: mocks + taper (combine weeks 7 + 8)

Total time: ~25 hours. You will not improve as much as the 8-week version, but you will arrive prepared and structured rather than panicked.

The compromise: you lose the deep pronunciation work. If your week 1 diagnostic showed intelligibility as one of your two weakest criteria, you should consider postponing the exam — pronunciation does not move in 4 weeks the way other criteria do.

If you have 12 weeks or more

Insert two extra deep-dive weeks between weeks 2 and 7:

• Week 2.5: a second week on intelligibility, but with a different sound pair than week 2 — drill the second-most-difficult one.
• Week 5.5: a deep dive on clinical accuracy of information — practise explaining 10 common diagnoses in plain English, in 90 seconds each.

The risk of a longer plan is loss of focus. Many candidates with 12+ weeks end up doing 6 weeks of useful work, then losing momentum. If you have a long runway, build in two scheduled rest weeks (week 4 and week 9), where you do no formal practice. Your skills will consolidate during the rest. Coming back fresh after a planned break almost always produces a score jump on the next mock.

Common ways this plan fails

• Skipping week 1 because you "already know your weaknesses". Almost no candidates accurately know their weaknesses without a recorded baseline. Do the diagnostic.
• Doing all the drills but no mocks until week 7. The plan compresses but does not eliminate role-play practice in the earlier weeks — each week has 1–2 short role-plays integrated into the drills. Do not save mocks for week 7 only. Save timed, full-length mocks for week 7.
• Doing mocks every week and skipping the drills. The opposite failure. Mocks without targeted drills between them is the most common ineffective approach.
• Practising the same scenario type over and over. Vary across initial assessment, discharge, and difficult-conversation. The exam will pick one of these — you do not get to choose.

How OET Live fits this plan

You do not need OET Live to follow the roadmap. Any source of practice role-plays plus any rubric-aligned scoring works.

What we do is collapse the practice-plus-feedback loop. Each AI patient role-play in OET Live runs 5–10 minutes, and the scoring report comes back in under a minute with per-criterion feedback tied to specific moments in your transcript. That changes the cost of doing a 20-minute weekly drill into the cost of doing five 20-minute drills, which is roughly what this plan needs you to do.

If you want to try one before deciding whether the workflow fits, start with our nursing case bank — there is no signup gate on the first practice.

---

This roadmap is a starting point. If your weak criteria do not match the most common combinations described above, the week ordering may need to shift — pronunciation should still come first, mocks should still come last, but the middle weeks can rearrange. The principle is what matters: slow-moving skills first, fast-moving skills last, mocks only after the foundation is in place.