CELPIP Writing Tasks, Timings And How Raters Actually Score You

Last Updated November 15, 2026
shane

Written by Shane Jordan

Summarize with AI

On this page

CELPIP Writing is 53 minutes, split unevenly. Task 1, the email, gets 27 minutes. Task 2, the survey response, gets 26. Both ask for 150 to 200 words, and the test owner’s own 2026 study pack puts those four numbers on a single page.

Candidates rarely lose this component on timing, though. They lose it on Task 2, by writing the balanced, two-sided answer that IELTS Task 2 trains them to produce, when CELPIP asks them to choose one option and defend it.

CELPIP Writing Runs 53 Minutes Across Two Tasks

The CELPIP-General Overview Study Pack sets out both tasks on page 15. Read on 12 November 2026:

“Task 1: 27 minutes to complete. Write 150–200 words. Write an email about an everyday matter, in which you’ll usually be asked to do around 3 things.”

“Task 2: 26 minutes to complete. Write 150–200 words. Respond to an opinion survey, choose between two options, and explain your choice (why is it better?). No need to use email format here.”

Note the asymmetry. Task 1 has one extra minute, and it is the task with the formal conventions to satisfy. Note also the three things: an email prompt usually sets out three bullet points, and answering two of them caps your Task Fulfillment rating no matter how well the prose reads.

So the component is not two 26-and-a-half-minute halves. If you rehearse both tasks on the same clock, you are rehearsing Task 1 a minute short and Task 2 a minute long.

The Screen Gives You A Word Counter And A Spellchecker

The study pack lists what the interface provides: “Background information is on the left side. Specific instructions are on the right side. Basic editing tools. Automatic word count. Spellcheck!”

So the word count is not something you estimate. It is on screen, under your response. The test owner’s tips and resources page confirms both tools and adds a caution about the second one: the spellchecker “is sometimes difficult to choose the right word when the spelling checker offers you a choice.”

That matters more than it sounds. A spellchecker that offers you a plausible wrong word turns a spelling error into a word-choice error, which is scored under Vocabulary rather than Readability.

Going Slightly Over Or Under Is Not An Automatic Penalty

Google’s summary for this topic says the limit is “strictly tracked” and that “penalties apply if you go under or over.” Several ranking pages build a whole strategy around hitting an exact number, and one invents a 170-to-190-word target that appears nowhere official.

What the publisher actually says is softer: candidates have “a little bit of room to be under or over the suggested word count limits on the CELPIP Writing Test tasks.”

Word count is one of four factors inside Task Fulfillment, alongside relevance, completeness and tone. It is not a gate. A 210-word response that answers all three bullets is in better shape than a 175-word response that answers two.

Four Dimensions Decide Your Writing Level

The dimensions are named on the test owner’s results page, with their sub-factors:

DimensionWhat sits inside it
Content/CoherenceNumber of ideas, quality of ideas, organization of ideas, examples and supporting details
VocabularyWord choice, suitable use of words and phrases, range of words and phrases, precision and accuracy
ReadabilityFormat and paragraphing, connectors and transitions, grammar and sentence structure, spelling and punctuation
Task FulfillmentRelevance, completeness, tone, word count

Two things follow. Spelling and punctuation are pooled with grammar and paragraphing under Readability, so a handful of typos competes for attention with sentence structure rather than being counted separately. And tone is scored, which is why the study pack’s instruction that the email be “neither too formal nor too informal” is a scoring instruction and not a stylistic note.

A lot of published advice renames these: “Coherence and Structure”, “Grammar & Mechanics”, “Readability/Grammar”. Google’s summary for one query renames two of the four. If you are self-assessing against a rubric, use the four names above, because those are the categories the raters work in.

Four To Six Raters Score Your Writing, And Nobody Averages Anything

Here is the claim to drop. A widely repeated model says each dimension is scored 0 to 5 and the four are averaged to produce your level. One page ranking for writing-sample queries puts it in two sentences: “Each criterion is scored on a scale of 0 to 5, with 5 being the highest. Your final score is the average of the scores across all four criteria.” Google’s summary repeats the same idea as “Scoring Dimensions (0–5 / CLB conversion).”

The publisher does not publish a 0-to-5 scale, and it does not publish an average. What it publishes is this:

“Each dimension is divided into five performance levels.”

“The Speaking and Writing component scores are derived from the dimensional ratings assigned by the raters.”

Those ratings are then “transformed into a CELPIP level” by rules “established by English language experts who participated in a standard setting exercise.”

Five performance levels per dimension is not a 0-to-5 score, and a standard-setting transformation is not an average. The same page gives the rater count: “each CELPIP writing performance is rated by four to six writing raters.” They work independently, with no knowledge of each other’s ratings, and a benchmark rater is assigned automatically if they disagree.

The practical consequence is the opposite of what the averaging model implies. You cannot trade a weak dimension against a strong one by arithmetic, because no arithmetic is published. What you can do is stop leaving a bullet point unanswered.

What Each Writing Level Asks For, From 3 To 12

The test owner publishes a two-page level descriptor chart dated May 2026, with Speaking on page one and Writing on page two. Each level carries a label:

LevelPublished label
12Expert proficiency in high-stakes social, educational, or workplace contexts
11Advanced proficiency in high-stakes social, educational, or workplace contexts
10Highly effective proficiency in some high-stakes social, educational, or workplace contexts
9Effective proficiency in some high-stakes social, educational, or workplace contexts
8Good proficiency in more demanding social, educational, or workplace contexts
7Adequate proficiency in somewhat demanding social, educational, or workplace contexts
6Developing proficiency in everyday social, educational, or workplace contexts
5Acquiring proficiency in everyday social, educational, or workplace contexts
4Adequate proficiency for daily life activities
3Some proficiency in limited contexts of personal relevance

Read the gap between 8 and 9, because that is the gap most permanent residence candidates are trying to cross. At level 8 a candidate can “develop a main idea with supporting details” and write “with good control of complex grammatical structures.” At level 9 the same candidate must “support key ideas with relevant facts, descriptions, details, or quotations” and choose “words and phrases to provide accurate details, descriptions, and comparisons.”

The difference is specificity, not complexity. Level 9 wants particular facts and precise comparisons. It does not want longer sentences. Our page on the CELPIP score chart sets out how these levels convert to Canadian Language Benchmarks.

Writing Band 8 Is Where Most Candidates Land

The test owner publishes score distributions annually, and almost nobody cites them. From the 2025 test taker data report, Table 5a, the percentage of CELPIP-General candidates at each Writing band:

BandShare of candidates
125.5%
113.3%
105.9%
99.4%
836.1%
723.6%
69.9%
54.4%
41.4%
30.5%
0 to 20.2%

Add the top four rows: 24.1% of candidates reach Writing 9 or above. Nearly three in five land on 7 or 8. If you are targeting Canadian Language Benchmark 9 across all four skills, Writing is the component where the published data says you are most likely to come up one level short, and band 8 is the specific trap, because 36.1% of all candidates sit there.

The report also breaks results down by declared nation. India’s Writing mean is 7.90 with a standard deviation of 1.39, the tightest spread of any of the top ten nations. Indian candidates cluster hard around 8. That is a coaching problem with a precise shape: the distance from 8 to 9 is small, common, and specific.

One caveat the report invites. It publishes no sample size and no reporting dates beyond the year, so treat the percentages as shape rather than precision.

Re-evaluation Refunds Its Own Fee If Your Level Rises

If you believe a Writing level is wrong, the policies page sets out the route. You may apply within six months of the test date, results arrive in roughly one to two weeks, and there is “a limit of one re-evaluation for any particular component.”

The incentive is unusual. If the level increases on Speaking or Writing after re-evaluation, the fee paid on that component is refunded. On the same site the condition is worded differently in one place, as the level changing rather than increasing, so do not plan around the looser version.

Two further points. Requests are final sale and cannot be cancelled. And the fee amount is not published anywhere public: you see it inside your account at the point of application. The cancellation and transfer charges that sit alongside it are covered in what CELPIP costs in India.

There Is No Single-Skill Retake, So A Weak Task Costs The Whole Fee

IELTS offers a One Skill Retake. CELPIP publishes nothing of the kind, and the only retake rule on its policies page is about frequency:

“You may register for no more than one Test Session within five (5) calendar days.”

So a candidate who hits 9, 9, 9 and 8 cannot resit the writing alone. The whole test goes again at the full fee. It is also worth knowing that Immigration, Refugees and Citizenship Canada does not accept the IELTS One Skill Retake for Express Entry either, so the comparison does not help you in that direction.

That raises what a single Task 2 misjudgement is worth. Preparation for this component is cheap next to a second sitting, which is the case for working with a rater’s criteria rather than against a word count. Our CELPIP coaching is built around the four dimensions above, and the CELPIP Speaking component carries the same scoring logic with a different fourth dimension.

Key Facts

  • Writing is 53 minutes: 27 minutes for Task 1, the email, and 26 for Task 2, the survey response.
  • Both tasks ask for 150 to 200 words, with an on-screen automatic word count and a spellchecker.
  • Task 1 usually sets three things to do. Answering two of them limits Task Fulfillment.
  • Task 2 requires you to choose one of two options and say why it is better. No email format, and no balanced two-sided answer.
  • Four dimensions are scored: Content/Coherence, Vocabulary, Readability and Task Fulfillment, each divided into five performance levels.
  • No 0-to-5 scale and no averaging is published. Dimensional ratings are transformed into a level by standard-setting rules.
  • Four to six raters score each writing performance, independently, with a benchmark rater on disagreement.
  • 24.1% of candidates reach Writing 9 or above, and 36.1% land on exactly 8.
  • India’s Writing mean is 7.90, with the tightest standard deviation of the top ten nations at 1.39.
  • Re-evaluation runs six months from the test date, takes one to two weeks, and refunds the component fee if the level rises.
  • No single-skill retake exists, and registrations are limited to one test session in five calendar days.

Questions Candidates Ask Before Booking

How many minutes do you get for CELPIP Writing Task 1 and Task 2?

Task 1 gets 27 minutes and Task 2 gets 26 minutes, for a 53-minute component. The test owner’s 2026 CELPIP-General Overview Study Pack publishes both figures on the same page. Several widely read guides state 27 minutes for Task 2, which is wrong by one minute in the harder direction.

How many words should you write for CELPIP Writing?

150 to 200 words for each task. The screen displays an automatic word count under your response. The publisher says candidates have a little bit of room to be under or over, so the range is guidance scored inside Task Fulfillment rather than a hard cut-off, and targets such as 170 to 190 words appear in no official document.

Is CELPIP Writing scored out of 5 per criterion and then averaged?

No. The test owner publishes neither a 0-to-5 scale nor an averaging formula. It states that each of the four dimensions is divided into five performance levels, and that the dimensional ratings assigned by raters are transformed into a CELPIP level by rules established through a standard setting exercise. The averaging model is a third-party invention repeated widely, including in Google’s AI summary.

What are the four CELPIP Writing scoring criteria?

Content/Coherence, Vocabulary, Readability and Task Fulfillment. Content/Coherence covers the number, quality and organization of ideas plus supporting details. Vocabulary covers word choice, range and precision. Readability covers format and paragraphing, connectors, grammar and sentence structure, spelling and punctuation. Task Fulfillment covers relevance, completeness, tone and word count.

Does CELPIP Writing have a spell checker?

Yes. The interface includes an automatic spellchecker, basic editing tools and an automatic word count. The publisher adds a caution: it is sometimes difficult to choose the right word when the spellchecker offers a choice. A wrong selection moves the error from spelling, under Readability, to word choice, under Vocabulary.

Is CELPIP Writing marked by a computer or a human?

By humans. Each writing performance is rated by four to six writing raters who work independently and have no knowledge of one another’s ratings, with a benchmark rater assigned automatically if they disagree. Listening and Reading are computer rated, which is why the publisher says re-evaluating those two is unlikely to change a result.

How hard is it to get 9 in CELPIP Writing?

Harder than most candidates assume. The publisher’s 2025 data report shows 24.1% of CELPIP-General candidates reaching Writing 9 or above, and 36.1% landing on exactly 8. India’s Writing mean is 7.90 with a standard deviation of 1.39, so Indian candidates cluster tightly just below 9.

Can you retake only the CELPIP Writing component?

No. CELPIP publishes no single-skill retake. A candidate who needs one more level in Writing has to sit the full test again at the full fee. The only published retake rule limits you to one test session within five calendar days. Immigration, Refugees and Citizenship Canada also does not accept the IELTS One Skill Retake for Express Entry.

How long does CELPIP re-evaluation take and what does it cost?

Roughly one to two weeks from submitting the application and paying the fee, and you may apply within six months of the test date. There is a limit of one re-evaluation per component, and requests cannot be cancelled. The fee amount is not published publicly: it appears inside your CELPIP account and depends on which components you choose.

What is the difference between CELPIP Writing Task 1 and Task 2?

Task 1 is an email about an everyday matter, usually with around three things to accomplish, in a register that is neither too formal nor too informal. Task 2 responds to an opinion survey by choosing between two options and explaining why the chosen one is better, with no email format required. Task 1 is the task with conventions to satisfy; Task 2 is the task that punishes a balanced answer.

About the Author
shane

Shane Jordan

Head of Learning and Development

An experienced English Language Teaching (ELT) Specialist and Director of Learning & Development at InSync, with over 25 years of global teaching and training experience. As India’s First OET Teacher Trainer and a former British Council IELTS Examiner with 40,000+ assessments, he specializes in OET/IELTS preparation, teacher training, corporate communication, healthcare English, and learner-centered program development, working with leading organizations, universities, schools, and healthcare institutions across India.
OET Coaching in Chennai

Learn from British Council (Former) IELTS Examiner Shane Jordan

Related Reading