Versant Test Questions 2026: Every Task Explained

🦈 Every Versant task rewards the same thing: saying a full English sentence cleanly, at conversational pace, without stalling. SpeakShark builds exactly that with daily AI conversation, phoneme-level pronunciation feedback while you speak, and a teacher who remembers the mistakes you made last session. Free, three sessions a day, no card. Start a free speaking session →

The speaking-only Versant test asks six question types in a fixed order: Reading (read numbered sentences aloud), Repeat (say back a sentence you hear), Short Answer Questions (one-word answers), Sentence Builds (rearrange three scrambled phrases), Story Retelling (retell a short story), and Open Questions (give an opinion). That is roughly 63 items in 17 minutes, and no human ever speaks to you. If your employer uses one of the written Versant tests instead, you also get Sentence Completion (the "fill in the blanks" part), Dictation, Conversations and Passage Reconstruction. Below, every task is explained with original example items we wrote ourselves — the real item pool is randomised and confidential, so anyone selling you "the actual questions" is selling you nothing.

In this guide: questions at a glance · which test you'll sit · the six speaking tasks · fill in the blanks and dictation · what each task measures · how we wrote these examples · official demo · honest limits · one-week plan · FAQ

Versant test questions at a glance

There is no single "Versant test". Pearson sells a family of them, and the question types you face depend entirely on which one your employer or school bought. Here is the whole family as published in Pearson's official test guides, checked July 2026:

Sources: the Versant English Test official guide, the English Speaking and Listening Test guide (2024), the 4 Skills Essential guide (July 2024) and the Placement Test guide. Item counts for the speaking test are confirmed in Pearson's test description and validation report, which states the system "presents approximately 63 items in six separate sections".

Three structural facts apply to every test in the table, and they matter more than any individual question:

• You cannot go back. Every guide repeats the same line: "You can only move forward — it is not possible to go back to a previous question or re-record an answer."
• Silence ends the item. Start speaking within 6 seconds on most tasks (8 seconds on Sentence Builds and Conversations) or the system moves on and scores the silence.
• A partial answer still counts. "If you do not complete an answer in the allowed time, your partial response will be saved and scored as it is." Half a sentence beats no sentence.

Which Versant test will you actually sit in 2026?

Worth checking before you prepare, because 2026 is a transition year. Pearson's current test catalogue lists the classic English Speaking Test as "Retiring soon, replaced by English Speaking and Listening Test".

The replacement keeps the recognisable tasks — repeats, retellings, opinions — but reorganises the scoring. The old test reported Sentence Mastery, Vocabulary, Fluency and Pronunciation. The 2024 Speaking and Listening test reports Speaking, Listening and Manner of Speaking instead, where "Manner of Speaking" covers pronouncing English sounds close to typical norms and speaking "at a normal conversational pace, without too much hesitation or unusual pauses". It also adds a genuinely new task: listening to a passage and answering three comprehension questions about it, rather than only retelling it.

Practically, ask your recruiter two questions: which Versant test, and is it proctored with a webcam? BPO and IT-services hiring in 2026 mostly uses either the 17-minute speaking test or the 30-minute 4 Skills Essential Test — and the second one requires a camera, Chrome, and typing. Employer-specific patterns are covered in our guides to the Versant test at Accenture, Concentrix, Teleperformance, TCS and Cognizant, Deloitte USI and Amazon.

The six speaking task types, with original examples

Every example below is written by us to match the published format. None is a real item.

Part A: Reading (8 items, 15 seconds each)

You see a screen of numbered sentences and an examiner voice tells you which one to read: "Now, please read sentence 7." The official guide notes the sentences come in groups of four sequentially coherent lines, so the group gives you context for how each sentence should sound.

Our example set:

1. The bus to the airport leaves from the north entrance.
2. It runs every twenty minutes until midnight.
3. If you miss it, a taxi costs about the same for two people.
4. Just remember the driver only takes cards, not cash.

This part exists partly to warm you up — Pearson says reading aloud "presents a familiar task and is a comfortable introduction". It still scores Pronunciation and Fluency, so read at a normal pace, stress the content words ("north", "twenty", "cards"), and pause at the commas. Do not correct yourself: the guide explicitly warns "you should not correct yourself", because a restart wrecks the fluency measurement more than the original slip did.

Part B: Repeat (16 items, 15 seconds each)

You hear one sentence and say it back exactly. Sentences run from three words to about fifteen and get harder as the section goes on.

Our example ladder:

• "Close the door." (3 words)
• "I left my keys upstairs." (5)
• "We can meet after your shift ends." (7)
• "The train was delayed because of the storm last night." (10)
• "She told me the office would be closed until the following Monday." (12)
• "My brother said he would rather pay more and get the faster delivery." (13)

This is the single most misunderstood task on the test. It is not a memory test. Pearson addresses the accusation head-on in the validation report: verbal memory for digit strings "is distinct from the cognitive resources used to process and comprehend sentences", and if memory drove the score, native speakers would show wide score variation — they do not. What you can repeat is bounded by the size of the chunk you process automatically. Someone who hears "the really big apple tree" as one unit can repeat fifteen words; someone decoding word by word stalls at seven.

That is why grammar drills move this score so slowly and daily talking moves it faster: the skill being measured is automaticity, not recall.

Part C: Short Answer Questions (24 items, 15 seconds each)

You hear a short question once and answer with one word or a short phrase. Pearson designs these to need no cultural knowledge — they should be answerable by "a typical 12-year-old native speaker of English and an adult who has never lived in an English-speaking country".

Our examples:

• "If a shop opens at nine and closes at five, how many hours is it open?" → Eight.
• "What do you call the meal people usually eat in the morning?" → Breakfast.
• "Do you wear gloves on your hands or on your feet?" → Hands.
• "If today is Wednesday, what day was it two days ago?" → Monday.
• "Which is colder, ice or steam?" → Ice.

The trap is over-answering. This section scores Vocabulary, and the guide's own tip is "Do not give a long or detailed reply." "Eight" scores; "I think it would probably be around eight hours or so, maybe" wastes the clock and adds nothing.

Part D: Sentence Builds (10 items, 15 seconds each)

You hear three short phrases in random order and say one grammatical sentence containing exactly those three phrases.

Our examples:

• "in the drawer / I put / your passport" → I put your passport in the drawer.
• "by Friday / the report / needs to be finished" → The report needs to be finished by Friday.
• "who fixed / she asked me / the printer" → She asked me who fixed the printer.
• "before the store closes / we should go / if you need milk" → If you need milk, we should go before the store closes.

Say the whole sentence in one flow. Candidates lose points by thinking aloud — "in the drawer… I put… no wait" — because Fluency measures response latency, speaking rate and pause placement. Decide silently, then deliver.

Part E: Story Retelling (3 items, 30 seconds each)

You hear a story of three to six sentences (30 to 90 words) and retell it in your own words.

Our example story: A young man was cycling home from work when his chain snapped near a quiet park. He started pushing the bike, expecting a long walk. A woman jogging past stopped, opened a small tool kit in her backpack, and fixed the chain in under five minutes. She refused any money and jogged away before he could ask his rescuer's name.

A strong retelling covers who, what, where and how it ended, in complete sentences, in your own words. Pearson's tip is blunt: "Use your own words — do not repeat exactly what you heard." This is the first task that samples real connected speech, and it feeds all four subscores at once.

Part F: Open Questions (2 items, 40 seconds each)

You are asked something about family life or personal choices and you argue a position.

Our examples:

• "Would you rather work from home or in an office? Please explain."
• "Do you think children should learn a second language at primary school? Why or why not?"

Here is the fact almost no coaching site mentions: this section is not scored. The official guide states the responses are "not scored but will be available to authorized Pearson listeners if test scores need to be reviewed", and the validation report adds that they "are not considered test items". It is a human-audit sample. Answer it properly anyway — a coherent 40 seconds is exactly what a reviewer wants to hear if your file is ever pulled — but do not panic if the topic bores you.

Forty seconds of unbroken opinion is also the hardest thing to fake on test day. It is the same muscle you use in a job interview role play, and the only way to build it is to talk daily. Practise a 40-second answer free →

The written tasks: fill in the blanks, dictation and passage reconstruction

If your test runs 30 or 50 minutes, you are sitting a four-skills version and typing is part of your score. These are the tasks people search for as "Versant fill in the blanks" and "Versant dictation".

Sentence Completion, the fill in the blanks task (18 items on 4SE, 20 on VEPT, 25 seconds each)

You see a sentence with one word missing and type exactly one word. It scores Reading.

Our examples:

• "The flight was delayed for three hours, so we ______ in the terminal until midnight." → waited
• "She apologised ______ arriving late to the interview." → for
• "Because the main road was closed, the driver took a longer ______ to the warehouse." → route
• "There is not ______ time left to finish the report before the deadline." → enough

Two of those four are pure collocation and preposition knowledge, which is exactly the point. Pearson's tip: "Look at the words either side of the gap", "Pay attention to spelling and capitalization", and "Only type one word." Typing two words scores zero even if both are right.

Dictation (14 items on 4SE, 16 on VEPT, 25 seconds each)

You hear a sentence once and type it word for word. On the Placement Test it scores Listening; on the 4 Skills Essential Test it scores Listening and Writing, because "appropriate grammar, spelling, punctuation, and capitalization" are assessed.

Our examples:

• "The meeting has been moved to Thursday at ten o'clock."
• "Please send the invoice to the address on the second page."
• "Nobody answered the phone, so I left a short message."

The official tip is counter-intuitive and worth obeying: listen to the whole sentence before you start typing. Typing while listening splits your attention and you lose the ending. Hold the meaning, then type fast.

Conversations (12 items on 4SE, 12 on VEPT)

You hear two people talk, then answer one question with a word or short phrase.

Our example: A: "Did you book the meeting room for two or three o'clock?" B: "Three — two was already taken by the finance team." Question: What time is the room booked? → Three o'clock.

Distractors are deliberate. Both times get said; only one is the booking.

Passage Reconstruction (2 items on 4SE, 3 on VEPT, 90 seconds each)

You read a short passage, it disappears after 30 seconds, and you rewrite it in your own words.

Our example passage: Maria forgot her umbrella at the office. It started raining hard on her way to the station. A stranger walking beside her offered to share his umbrella until they reached the platform. Maria thanked him and only afterwards realised she had never asked his name.

Pearson's instruction is precise: "Do not write a quick summary — include all the details you can." Detail density scores; elegance does not.

The Placement Test adds two tasks nobody else has: a 60-second Typing passage (unscored, but administrators see it, and you need above 12 words per minute with 90% accuracy for a valid Writing score) and an 18-minute Summary and Opinion essay pair, 25 to 50 words of summary plus at least 50 words of your own argument.

What each task actually measures

The score report is not a single number, and knowing which task feeds which subscore tells you where to spend practice time.

Two numbers from Pearson's test description and validation report reframe the whole test. First, content accuracy counts for 50% of the Overall score and manner of speaking for the other 50% — "in the Versant English Test scoring logic, content and manner (i.e. accuracy and control) are weighted equally". Getting the words right while sounding halting caps your score just as hard as fluent nonsense does.

Second, the same validation report puts the Standard Error of Measurement on the Overall score at 2.8 points on the Versant 20–80 scale. A three-point gap between two attempts is measurement noise, not progress. What the bands mean in practice is covered in our guide to good Versant scores and levels.

How we wrote these example questions

A methodology note, because this topic attracts a lot of invented content:

• Every task description, item count, timing and tip above comes from a Pearson document we opened and read in July 2026 — the four official test guides linked in the table, plus the Versant English Test description and validation report. Where the guides and the validation report disagree in wording, we used the validation report, which is the more technical document.
• Every example item is original. We wrote each one to match the published constraints (Repeat sentences of 3 to 15 words; Story Retelling passages of 30 to 90 words in three to six sentences; Short Answer questions answerable without cultural knowledge). We deliberately did not reuse the sample items printed in the guides, and we do not publish real items, because the item bank is confidential and reproducing it would help nobody.
• We flag what is unknown. Pass marks are set by employers, not Pearson: the validation report says "score users must decide what Versant English Test score is considered a minimum requirement in their context". Any site quoting a company's exact cutoff as fact is guessing unless it links an official source.
• No affiliate relationship. We sell an English speaking app. We do not sell Versant tests, practice tests or scores, and nothing on this page earns us a commission.

SpeakShark is a speaking-improvement tool, not an exam-prep provider, and is not affiliated with or endorsed by Pearson, any exam board, or any employer named on this page.

Versant test demo: what is official and what is not

Both official guides end the same way: request a free unscored sample test, and optionally buy a scored practice test that returns a real score report. Access usually runs through your test administrator, so ask them for the demo link before test day — it is the only way to see the actual interface, timing bars and examiner voice.

What is not official: the flood of "Versant test questions and answers 2026" PDFs, YouTube dumps and Telegram files. They fail for a structural reason, not a moral one. Pearson draws items "at random from a large item pool", so "most or all items are different from one test administration to the next", and proprietary algorithms pick each item on difficulty and similarity. Memorising a leaked list buys you nothing, and unscored trial items are seeded into live tests specifically to keep the bank refreshed.

The one thing worth rehearsing from a demo is the interface: the countdown, the beep before your recording starts, the impossibility of going back. Combine that with a proper equipment check — covered in our guide to Versant test setup, camera and headphones — and you remove most of the avoidable failure modes.

Where practice helps, and where it honestly does not

This is the section most Versant guides skip, and it comes straight from Pearson's own data.

Format familiarity alone barely moves your score. In the test-retest study reported in Pearson's validation report (n=140), correlations between administrations were 0.97 with a practice test, 0.97 without one, and 0.97 across the repetition-effects comparison. Mean scores across three sittings differed by less than one point — "increasing familiarity with the tasks does not result in any consistent change in Versant English overall scores". Translation: taking six practice tests this weekend will not add points. What raises a Versant score is a genuine increase in how automatically you process and produce spoken English.

When something else beats a speaking app:

• Your test includes typing. No conversation app improves your words-per-minute or your comma placement. If you are sitting the 4 Skills Essential or Placement Test, spend part of your week on typing drills and written accuracy.
• You have a specific grammar gap. If you cannot form past-perfect clauses at all, conversation surfaces the gap but a grammar course closes it faster.
• You need an official score prediction. Only Pearson's paid practice test gives you a real score report. No third-party tool, ours included, can output a Versant score.
• Your bottleneck is nerves, not English. Then the fix is exposure, not content — start with our guide to overcoming the fear of speaking English.

Where daily speaking practice genuinely does help: the automaticity the test is built to measure. Repeat length, Sentence Builds latency, retelling density and pause placement all improve with volume of real talking plus feedback on the specific sounds you distort. That is a months-long process, not a weekend one, which is exactly why it beats cramming.

A one-week practice plan by task type

Twenty minutes a day, mapped to the tasks:

• Monday — Repeat. Read a news paragraph aloud, then close it and say each sentence from memory. Push your chunk size from 7 words to 12.
• Tuesday — Short Answer. Have someone (or an AI teacher) fire 20 one-word-answer questions at you. Answer in under two seconds. No hedging.
• Wednesday — Sentence Builds. Take any sentence you read, split it into three phrases, scramble them, and rebuild aloud in one breath.
• Thursday — Story Retelling. Listen to a 60-second podcast clip and retell it in 30 seconds, in your own words, timed.
• Friday — Open Questions. Answer one opinion prompt for 40 seconds with two supporting reasons. Record it. Listen back once.
• Saturday — full mock. Run the six tasks back to back under timing, with a headset, in a quiet room.
• Sunday — sounds. Take the two or three phonemes that keep getting flagged and drill them inside real sentences, not word lists.

The compounding problem is what makes most self-study stall: on day 12 you have forgotten which sounds day 3 flagged. SpeakShark is built around that gap — the AI teacher carries cross-session memory of your recurring mistakes, your level and your past topics, so Thursday's session already knows Monday's weak vowel and pushes you there. Feedback lands at the phoneme level during natural conversation, not in isolated drills, which is the same integrated listen-then-speak condition Versant measures. Free tier: three 5-minute sessions a day with teacher Sarah, 64 topics, no card. Premium is $10/month or $69/year (about $5.75/month) for unlimited 10-minute sessions with unlimited turns, four accent teachers (American, British, Australian, Canadian) and 320 topics.

For structured test-week preparation, see our Versant preparation guide; for the BPO context specifically, our guide to English speaking practice for call center agents covers the same skills in job form, and talking on the phone in English drills the audio-only condition Versant recreates. If you are not sure where you stand today, start with a free English speaking level check.

FAQ

Every Versant task is a variation on one question: can you understand English at conversational speed and answer in a clean full sentence without stalling? Nothing memorised gets you there. Talking every day, with feedback that remembers what you got wrong yesterday, does. Start your free SpeakShark session →

All task types, item counts, timings and quoted tips verified July 2026 from Pearson's official test guides for the Versant English Test, English Speaking and Listening Test, 4 Skills Essential Test and English Placement Test, plus the test description and validation report and Pearson's current test catalogue. Test formats change; confirm with your test administrator. All example questions on this page are original and written by SpeakShark. SpeakShark is a speaking-improvement tool, not an exam-prep provider, and is not affiliated with Pearson or any employer named.