FlexiLingo12 CELPIP Writing Mistakes That Cap You at CLB 7 (and How to Fix Them)
Most test-takers stuck at CLB 7 in Writing aren't short on English — they're repeating a handful of fixable habits. Here are the 12 score-capping mistakes, each with a concrete fix.
1Why capable test-takers plateau at CLB 7
If you keep landing on CLB 7 in Writing while your Reading and Listening sit higher, you are not the exception — you are the rule. CLB 7 is where strong, conversational English stalls. It is the band the test gives a response that communicates clearly but does not yet feel controlled, precise, and fully on-task. The gap to CLB 9 is rarely "more English." It is a short list of repeatable habits, and habits can be retrained in weeks.
CELPIP Writing has two tasks. Task 1 is an email of roughly 150–200 words in about 27 minutes, written to a specific reader for a specific purpose. Task 2 is a survey response of roughly 150–200 words in about 26 minutes, where you pick one of two options and justify it. Both are scored on four dimensions: Content/Coherence, Vocabulary, Readability, and Task Fulfillment. Every mistake in this guide pulls down at least one of those four — which is exactly why fixing them moves your band.
Notice what is not on that list: long words, rare idioms, or complex syntax for its own sake. Raters reward a message that is complete, organised, accurate, and appropriate for its reader. The test-taker who writes plainly but covers everything, in clean paragraphs, with the right tone, beats the one who reaches for impressive vocabulary and trips over it.
CLB 7 in Writing is almost never a vocabulary ceiling. It is the score for clear-but-uncontrolled writing. The 12 mistakes below each cost a specific dimension — Content/Coherence, Vocabulary, Readability, or Task Fulfillment — so fixing them is the fastest route to CLB 9.
2Mistake: your tone doesn't match the reader
CELPIP tells you exactly who you are writing to — a friend, a manager, a landlord, a customer-service department, a city official. Many test-takers write every email in the same neutral, semi-formal register, which reads wrong to whoever is actually receiving it. Writing "I am writing to inform you" to a close friend, or "Hey, just wondering" to a hiring manager, signals you have not adjusted to the reader. That costs you on Content/Coherence and Task Fulfillment, because matching the reader is part of the task.
Before you write a single line, name the relationship and pick a register: formal for officials and strangers, warm and relaxed for friends and family, polite-professional for colleagues and businesses. Then hold that register from greeting to sign-off — the opening, the requests, and the closing all have to agree.
Too formal for a friend: "Dear Sam, I am writing to inform you of my forthcoming visit."
Right for a friend: "Hi Sam, great news — I'm finally coming to visit next month!"
Fix: In the first 30 seconds, write the reader and the register at the top of your scratch space (e.g. "landlord → formal, polite"). Choose the greeting and sign-off to match before you draft, and the body will fall into the right tone.
3Mistake: leaving a bullet point unaddressed
Both CELPIP Writing tasks hand you specific points to cover — usually three. Task 1 emails list what to include (explain the problem, say what you want, propose a solution). Task 2 expects reasons that genuinely support your choice. The single most common reason a capable writer is capped is simple: they answer two of the three points fully and let the third slip, or they mention it in half a sentence. Task Fulfillment drops immediately, no matter how polished the rest reads.
Treat the prompt as a checklist, not a suggestion. Each required point deserves its own developed sentence or short paragraph — a claim plus a little detail — not a passing mention. Coverage beats elegance: a plain sentence that fully addresses point three scores higher than a beautiful paragraph that quietly skips it.
Skipped point: "I'd like to report the broken heater and ask when it can be fixed." (No proposed solution or deadline, though the prompt asked for one.)
Fully covered: "I'd like to report the broken heater, ask that a technician come this week, and request a partial rent credit for the cold days."
Fix: Number the required points before you write, and tick each one off as you cover it. Re-read the prompt in your final minute and confirm every numbered point has a developed sentence — not just a mention.
4Mistake: your word count is off
Both tasks target roughly 150–200 words. Coming in well under — say 110 words — almost guarantees a point was underdeveloped, so Content and Task Fulfillment suffer. Overshooting to 280 words is just as risky: you run out of time to proofread, you pad with repetition, and the extra length usually adds errors rather than substance. CLB 7 writers tend to live at one extreme or the other.
The sweet spot is a complete, developed response that lands in range without padding. Three covered points, each with a sentence of support, plus a greeting and a close, naturally lands near 170–190 words. If you are far short, you have under-developed a point; if you are far over, you are repeating yourself or over-explaining.
Padding to hit a count: "I really, really think this is a very, very important issue that is extremely important to me."
Developed instead: "This issue matters to me because the broken lock leaves my apartment unsecured overnight."
Fix: Aim for a developed 170–190 words and stop. Don't pad — if you're short, develop a point with a concrete detail; if you're long, cut repetition. Leave two minutes to proofread rather than spending them adding words.
5Mistake: weak, repetitive vocabulary
The Vocabulary dimension rewards range and precision, not rare words. CLB 7 responses often lean on the same handful of vague words — good, bad, thing, very, nice, a lot — and repeat the prompt's exact phrasing instead of rewording it. This reads as a limited range even when the writer knows plenty more. The fix is not to hunt for fancy synonyms (which often misfire); it is to choose the precise everyday word.
Precision is the goal. "The service was bad" tells the reader little; "the delivery arrived two days late and the box was damaged" is specific and naturally varied. Swap vague intensifiers for concrete detail, and vary how you refer to the same idea instead of repeating one noun five times.
Vague and repetitive: "It was a very bad experience and a very bad service and I am very unhappy."
Precise and varied: "The late delivery and the damaged box left me frustrated, and I'd like a refund."
Fix: Hunt down vague words — good, bad, thing, very, nice — and replace each with a precise detail rather than a fancier synonym. Reword the prompt's phrasing in your own words instead of echoing it; that alone widens your range.
6Mistake: recurring grammar slips that hurt readability
It is rarely one big grammar mistake that caps a band — it is the same small error repeated, which the rater notices and counts. The usual culprits are articles (a / an / the), subject–verb agreement, and inconsistent tense. A single slip is human; the same slip eight times signals an uncontrolled pattern and pulls down Readability across the whole response.
You do not need to master every grammar rule — you need to find your two or three recurring errors and watch for exactly those. Most CLB 7 writers can name them once they look: "I drop the before singular nouns," or "I switch from past to present mid-email." Knowing your pattern turns proofreading from vague re-reading into a targeted hunt.
Recurring slips: "I send the email yesterday because manager ask for report and it are urgent."
Controlled: "I sent the email yesterday because the manager asked for the report and it was urgent."
Fix: Identify your top two or three recurring errors (articles, agreement, tense are the usual three) from past practice, and in your final read, scan only for those. Targeted proofreading catches far more than a vague re-read.
7Mistake: one block of text, no paragraphs
A response written as a single dense block is hard to read, and Readability is one of the four dimensions. When everything runs together, the rater can't see your structure, and your three covered points blur into one. Worse, writing in a block makes it harder for you to check that each point got its own development — they hide inside the wall of text.
Use a clear, conventional shape. For an email: a greeting, a short opening line stating your purpose, one paragraph per main point, and a brief closing with a sign-off. For the survey task: a sentence stating your choice, a paragraph or two giving your supported reasons, and a short concluding line. White space between paragraphs is not decoration — it is part of how the response reads as organised.
One block: "Dear manager I want time off and the reason is family and I also need the dates approved soon thanks."
Paragraphed: "Dear Mr. Lee,\n\nI'm writing to request three days of leave next month for a family event.\n\nCould you confirm the dates by Friday so I can book travel?\n\nThank you,\nMaria"
Fix: Decide your paragraphs before you draft — greeting, one paragraph per point, close. Hitting Enter between each point forces you to develop them separately and instantly lifts Readability.
8Mistake: over- or under-using connectors
Connectors — however, therefore, in addition, on the other hand — guide the reader through your logic, and Coherence depends on them. Two failures cap writers here. Under-using them produces a list of disconnected sentences that the reader has to stitch together. Over-using them — starting nearly every sentence with Moreover, Furthermore, Additionally — reads as mechanical and padded, and often the connector doesn't even match the logic it claims.
Aim for natural, accurate signposting. Use a connector when the relationship between two ideas genuinely needs flagging — a contrast, a result, an addition — and let plain sentences carry the rest. Crucially, the connector must match the logic: "however" signals contrast, "therefore" signals result. A mismatched connector confuses the reader more than no connector at all.
Over-used and mismatched: "I like the gym. Moreover, it is close. Furthermore, it is cheap. Therefore, it has a pool."
Natural: "I prefer the gym because it's close and affordable, and it has a pool I'd actually use."
Fix: Use a connector only when two ideas genuinely contrast, build, or cause each other — and double-check it matches that logic. A handful of well-placed connectors beats one at the start of every sentence.
9Mistake: claims with no concrete support
This is the mistake that most often separates CLB 7 from CLB 9, especially in Task 2. A capped response states opinions and stops: "I prefer option A because it is better and more convenient." That is a claim with no support — better how, convenient for whom? Content/Coherence rewards developed ideas, and an unsupported claim is an undeveloped one. The reader is left with assertions, not reasons.
Back every main point with a specific reason, example, or consequence. The pattern is simple: state your point, then answer "why" or "for example" with something concrete. One vivid, specific detail develops a point far more than three more adjectives. This applies to Task 1 too — "the noise is a problem" becomes real with "the construction starts at 6 a.m. and wakes my children."
Unsupported: "I prefer working from home because it is better and more convenient."
Supported: "I prefer working from home because it saves me a 90-minute commute, which I use for focused work before meetings start."
Fix: After every claim, force yourself to add one concrete reason, example, or consequence — answer "why?" or "for example?" A single specific detail develops a point more than any number of extra adjectives.
10Mistake: a memorised template that doesn't fit
Memorising chunks of text — a stock opening, a generic body, a canned closing — feels like insurance, but it backfires. Raters are trained to spot prefabricated language, and a template that doesn't quite fit the prompt reads as off-task. Worse, leaning on a memorised frame makes you address the template's points instead of the prompt's, which is the fastest way to lose Task Fulfillment. The hours you spend memorising are better spent practising real responses.
Templates aren't all bad — a light, flexible structure (greeting → purpose → points → close) is fine and even helpful. The danger is memorised content: whole sentences you drop in regardless of the question. Learn the shape, not the script. Then fill that shape with language that responds to this specific prompt and this specific reader.
Memorised filler: "In today's modern world, this is a very important issue that affects many people in society."
Prompt-specific: "Our building's shared laundry has been out of service for three weeks, and residents are paying for laundromats."
Fix: Memorise a flexible structure, never fixed sentences. If a phrase could be pasted into any prompt, cut it and write something that names the actual situation, reader, and request in front of you.
11Mistake: skipping or misplacing the greeting and sign-off
Task 1 is a real email, and real emails have a greeting and a sign-off — "Dear Mr. Lee," ... "Thank you, Maria." Some test-takers drop straight into the body with no greeting at all, or skip the sign-off to save time. Others carry the email's greeting-and-sign-off habit into Task 2, where the response is a survey answer, not a letter — adding "Dear Sir/Madam" or "Best regards" there looks like you've misread the task. Either way, it's a format the rater has to mentally correct, which costs Task Fulfillment.
Match the shape to the task. Task 1 always gets a greeting that fits the reader and a sign-off that matches the register. Task 2 is a direct response to a survey question — no greeting, no sign-off, just your position and your reasons from the first sentence.
Task 1 with no greeting: "I want to report a problem with my internet connection." (Straight into the body, no opening.)
Task 1 with a greeting: "Dear Customer Service, I'm writing to report a problem with my internet connection."
Fix: Before you draft, confirm which task you're in. Task 1 always opens with a greeting and closes with a sign-off matched to the reader. Task 2 never does — it opens directly with your position.
12Mistake: burying your position in Task 2
Task 2 asks you to choose one of two options and justify it. CLB 7 responses often describe both options at length before finally revealing a preference in the last sentence, or never state a clear choice at all. Raters need to identify your position immediately to judge whether your reasons actually support it — a late or unclear stance costs Content/Coherence and Task Fulfillment even when the reasoning that follows is solid.
State your choice in the first sentence, then spend the rest of the response supporting it. You don't need to explain in detail why you rejected the other option — one brief acknowledgment is enough. The bulk of your words should develop reasons for the option you picked.
Buried position: "Both options have advantages. Working from home saves commute time, but the office offers more collaboration. In the end, I suppose home might be slightly better for me."
Clear position: "I would choose working from home, mainly because it saves me a 90-minute daily commute that I can put toward focused work instead."
Fix: Write your one-sentence answer first — "I prefer X because Y" — before you draft anything else. Everything that follows should support that opening sentence, not build up to it.
13Mistake: long, run-on sentences that lose control
Trying to sound advanced, some test-takers write single sentences that stack three or four clauses together with commas and connectors, reaching for a complexity their control doesn't quite support. Partway through, subject-verb agreement slips or the sentence loses track of its own structure. A grammatically shaky 35-word sentence scores worse on Readability than two clean 15-word sentences saying the same thing.
Complexity is not the goal — control is. Native-sounding writing mixes some longer sentences with plenty of short, clear ones. If you're not confident a sentence is fully correct once you've written it, split it in two. A rater can follow two accurate short sentences far more easily than one impressive-looking sentence that breaks down grammatically in the middle.
Run-on and shaky: "Because the noise from the construction site which starts every morning at six a.m. keep waking up my children and it are affecting their school performance so I am asking you to help."
Clean and controlled: "The construction site starts work at 6 a.m. every morning. The noise wakes my children and is affecting their performance at school."
Fix: If a sentence needs more than one comma to hold together, read it back out loud. If you stumble, split it into two shorter sentences — you'll rarely lose points for simplicity, but you will for a sentence that breaks down.
14How FlexiLingo helps you fix these
Every mistake above is a habit, and habits change through targeted, repeated practice on real tasks — not through reading more rules. FlexiLingo is built to give you that practice on authentic Canadian content and CELPIP-style tasks, so you train the exact dimensions raters score.
Write Task 1 emails and Task 2 survey responses against realistic prompts and the real time limits, so register, coverage, and word count become automatic under pressure.
Practice oriented around the four scored dimensions — Content/Coherence, Vocabulary, Readability, and Task Fulfillment — so you can see which one is capping you and aim your effort there.
Track the small grammar slips you repeat — articles, agreement, tense — and turn proofreading into a targeted hunt for your own top two or three patterns.
Mine precise, natural words from authentic Canadian reading and listening and review them with spaced repetition, so your range grows without resorting to risky rare synonyms.
Frequently Asked Questions
Because CLB 7 is the band for writing that communicates clearly but isn't yet controlled, precise, and fully on-task. Reading and Listening are recognition skills; Writing is production, where every gap shows — a skipped point, a tone mismatch, a repeated grammar slip. The fix is rarely more English. Find which of the four dimensions (Content/Coherence, Vocabulary, Readability, Task Fulfillment) keeps costing you and drill that one on real timed tasks.
Neither in isolation — both are scored, but as control and precision, not difficulty. For grammar, you don't need rare structures; you need to stop repeating two or three small errors (articles, agreement, tense) that the rater notices. For vocabulary, you don't need fancy words; you need precise everyday ones and a bit of range instead of the same vague words repeated. If you must pick one to start, fix your recurring grammar errors — they hurt Readability across the entire response.
Important, but as a symptom more than a rule. Both tasks target roughly 150–200 words. Coming in far under almost always means a required point was underdeveloped, which costs Content and Task Fulfillment. Going far over usually means padding and leaves no time to proofread, which adds errors. Aim for a developed 170–190 words — if you're short, develop a point with a concrete detail rather than padding; if you're long, cut repetition.
Yes, when it's memorised content rather than a flexible shape. Raters are trained to recognise prefabricated language, and a canned opening or generic body that doesn't fit the prompt reads as off-task — and it pulls you toward addressing the template's points instead of the prompt's, which is the quickest way to lose Task Fulfillment. A light structure (greeting → purpose → points → close) is fine; fixed sentences you'd paste into any prompt are not. Learn the shape, not the script.
Fully address every required point. More test-takers are capped by quietly skipping or half-mentioning one of the three required points than by any grammar or vocabulary issue. Number the points before you write, give each its own developed sentence with a concrete reason or example, and re-read the prompt in your final minute to confirm none slipped. It costs no extra English and directly lifts Content/Coherence and Task Fulfillment — the two dimensions that move CLB 7 toward CLB 9.
Train the writing habits that lift your band
Practise CELPIP-style Task 1 emails and Task 2 survey responses on FlexiLingo — under real time limits, focused on the four dimensions raters actually score.