TOEIC Link Speaking Strategy — Question Types, Rubric, and Score-Band Priorities
TOEIC Link Speaking is a skill that L&R does not test: read aloud, describe a picture, respond to questions, and express an opinion are recorded back to back in 15–20 minutes. CEFR is reported on six bands (Pre-A1 to C1), and L&R high scorers often stall on the first attempt because the standards for "amount of speech" and "fluency" are unfamiliar. This page lays out the structure and rubric of Link Speaking and works backwards from each score band to the highest-leverage training target for the next CEFR step.
How Link Speaking is structured — four families across 15–20 minutes
Link Speaking has four families: read aloud → describe a picture → respond to questions → express an opinion, with one to three items per family. Total items run 8–12 in 15–20 minutes including preparation. Recording is done through a microphone and headset; answers submit automatically.
The first wall L&R takers hit is how to use preparation time. Each family gives 15 to 45 seconds of prep, and trying to write a full script eats time you needed for the spoken response. The standard playbook is to leave the prep time at 3–5 keywords plus a logical outline, and generate full sentences during the spoken delivery.
- Family 1: Read aloud — 30 sec prep, 45 sec delivery on a 40–60 word passage
- Family 2: Describe a picture — 30 sec prep, 45 sec delivery on one image
- Family 3: Respond to questions — three short prompts, 15 sec prep + 15–30 sec each
- Family 4: Express an opinion — 30–45 sec prep, 60 sec delivery on a given topic
- CEFR output: Speaking reported separately on Pre-A1 to C1
- Total time: 15–20 minutes including prep, recorded format
Rubric — pronunciation, grammar, vocabulary, fluency, task completion
Speaking is scored on five dimensions (pronunciation / grammar / vocabulary / fluency / task completion) combined into an overall judgment. A weakness on one dimension is recoverable through the others, and a CEFR step is achievable through compensation. L&R takers already have a strong base in grammar and vocabulary, so allocating at least half of training time to pronunciation and fluency is the leverage point.
Task completion is the often-missed dimension: describing a picture out of order (general → focal → background), or stating an opinion without a conclusion, drops the CEFR by a band even if the language itself is fine. Keep four logical-structure templates ready, one per family — that is the foundation for Link Speaking.
- Pronunciation: vowel/consonant distinction + stress placement + word-final reductions
- Grammar: tense agreement + subject-verb agreement + relatives + passive voice
- Vocabulary: abstract word selection + collocations + paraphrasing power
- Fluency: speaking rate (90–140 wpm) + frequency of fillers (uh / um)
- Task completion: directly answering the prompt + completing the logical structure
- Weighting: pronunciation + fluency 50% / grammar + vocabulary 30% / task completion 20%
Score-band priorities — what to train next
The training that yields the biggest jump depends on your current Speaking CEFR. Pre-A1 / A1 takers gain most from reading aloud fixed phrases plus instant generation of basic sentence patterns; A2 → B1 from picture description structure plus quick question response; B1 → B2 from opinion-development logic plus abstract vocabulary. Aim just one family above your current band — that is the shortest path past each CEFR wall.
One caveat: pronunciation correction takes eight to twelve weeks to show effect. If you need to move a CEFR band in less than four weeks, invest in logical-structure templates and fluency rather than pronunciation. Run pronunciation correction in parallel as a longer-term, six-month project.
- Pre-A1 → A1: read aloud high-frequency vocabulary + verbal SVO/SVC sentence generation
- A1 → A2: 3-sentence picture description (general + focal + background) + Yes/No instant response
- A2 → B1: 5W1H instant response in Q&A + conclusion-first opinion templates
- B1 → B2: opinion development (claim → 2 reasons → conclusion) + abstract vocabulary
- B2 → C1: counter-rebuttal + extemporaneous speech in domain (legal / finance / medical)
- Pronunciation: an 8–12 week project, run separately from short-term goals
A 12-week practice menu — five hours per week
Five hours a week for twelve weeks is sixty hours. Whether that moves you a CEFR band depends on the individual, but a defensible split is 1h read-aloud + 1h picture description + 1h Q&A + 1h opinion + 1h pronunciation. Recording is mandatory: save every session on a phone voice memo and listen back for 30 minutes on the weekend. Practice without listening back grows at less than half the rate — that is the durable empirical rule of Speaking training.
For visibility, run a 15-minute mini-mock every weekend with recording, and log time, filler frequency, and task completion in a spreadsheet. If speaking rate and filler frequency stay flat for three months, the bottleneck is usually not vocabulary but automaticity of speech, so increase the shadowing portion before changing anything else.
- Mon: read-aloud, 60 min (3 sets × 20 min, recording mandatory)
- Tue: picture description, 60 min (5 images × 12 min, 3-sentence template)
- Wed: Q&A, 60 min (10 prompts × 6 min, prioritize speed of response)
- Thu: opinion, 60 min (3 topics × 20 min, conclusion-first practice)
- Fri: pronunciation, 60 min (30 min shadowing + 30 min individual sounds)
- Weekend: 15 min mini-mock + 30 min listen-back + 15 min weekly review
Speaking family × score band — training priority matrix
| Band | Read aloud | Picture | Q&A | Opinion |
|---|---|---|---|---|
| Pre-A1 | ★★★ | ★★ | ★ | ★ |
| A1 | ★★★ | ★★★ | ★★ | ★ |
| A2 | ★★ | ★★★ | ★★★ | ★★ |
| B1 | ★★ | ★★ | ★★★ | ★★★ |
| B2 | ★ | ★★ | ★★ | ★★★ |
| C1 | ★ | ★ | ★★ | ★★★ |
* Stars indicate weekly study weight (★★★ = primary, ★ = maintenance only). Pre-A1 to A1 emphasizes read-aloud and picture description templates; B1+ emphasizes opinion development.
Three checks when Speaking stalls
- Are you trying to write a full script in prep time? (Limit to 3–5 keywords plus a logical outline)
- Are you listening back to your recordings at least once a week? (Practice without listen-back grows at less than half the rate)
- Do you have one logical-structure template ready for each of the four families?
Frequently Asked Questions
Related articles
- TOEIC Link Listening — Question-Response strategyA deep dive on the partner skill referenced in this article when discussing speech automaticity and shadowing. Covers Q&A pacing, response speed, and rubric so Listening and Speaking can be planned in parallel.
- TOEIC Link Reading strategyThe same Pre-A1–C1 score-band framework this article uses, applied on the Reading side. Speaking and Reading share a vocabulary base, so this is the companion read for keeping skill weights balanced.
- TOEIC Link CEFR B1 guideThe starting point for the wall this article describes ("B1+ rarely reachable from L&R alone"). Builds the logical-structure templates and conclusion-first delivery needed to take B1 on Speaking.
- TOEIC Link CEFR B2 guideB2 — the typical English-interview pass mark referenced as "opinion development with abstract vocabulary" — broken down per skill. Covers the common stalls Speaking-strong takers hit at B2 and the path to C1.
- TOEIC Link vs TOEIC L&RBackground for the opening premise of this article ("L&R does not test Speaking"). A four-axis side-by-side of design intent, scoring, and taker base that explains why Speaking needs a Link-specific training plan.
TOEIC® and TOEIC Link™ are registered trademarks of ETS. EnglishBlitz is not affiliated with, endorsed by, or associated with ETS. Time, question families, and rubric notes above reflect publicly available information and aggregated taker reports as of April 2026 — confirm current specifications on the IIBC and ETS official sites.