The Five Writing Errors We See Most
Five errors account for most of the marks lost in this section, and none of them is a grammar problem. All five are decisions, which means all five are fixable this week.
Why these five
The errors that cost the most in CELPIP Writing are not the ones people prepare for. Candidates arrive worried about tenses and articles, and lose their marks to reading failures instead.
All five errors below share a shape. Each is a decision rather than a skill, each is made in the first four minutes, and each damages a heavy criterion. Three of the five hit Task Fulfilment at 25 per cent, and two of them hit Content and Coherence at 30 per cent as well.
That is good news, because a skill takes months and a decision takes one practice session to change. If you fix only these five, your band moves further than it would from a term of grammar study.
Error one: the bullet that is mentioned, not answered
This is the most common and the most expensive. The response touches all three bullets, so it feels complete, but one of them has a single vague clause and no specific detail.
It is almost always the administrative bullet: the one asking for a date, a reference number, an amount, a duration. The emotional bullet and the request bullet feel natural to write, so they get a paragraph each, and the third gets I paid quite a lot for the service recently.
That clause answers nothing. The prompt withheld those facts deliberately so that you would supply them, and supplying them is the bullet. My move took place on 14 March under booking number MR-40821, and I paid $1,340 for the full-service package answers it entirely, in one sentence, with facts invented on the spot.
The cost is double. Task Fulfilment falls because a bullet is not addressed, and Content and Coherence falls because an idea is missing from the response. Together those are 55 per cent of the score.
The fix is mechanical: before submitting, point at each bullet and find the sentence answering it. If the sentence contains no number, name, date or observable fact, the bullet is not answered yet.
Error two: changing a fact the prompt supplied
The prompt says a glass cabinet was cracked and a box of kitchen items is missing. The response describes a broken television and a missing suitcase. Everything else is fluent.
This is treated as a factual error, not a stylistic one, and fluent writing does not excuse it. One substituted or invented fact, or one bullet left unaddressed, caps Task Fulfilment at 8 and Content and Coherence at 9. Two or more such errors, or getting the central premise of the task wrong, caps Task Fulfilment at 6.
It usually happens for one of two reasons. Either the candidate read the scenario once, started planning, and wrote from memory; or they practised a similar prompt and their memory supplied the practised version's details.
Accurate paraphrase is not the problem. Writing the kitchen box for a box of kitchen items is fine, and so is the display cabinet for a glass cabinet. What is flagged is a genuine change of meaning: a different object, a different number, a different date, a different amount.
The fix takes fifteen seconds: read the scenario's nouns and numbers one last time before submitting, and confirm each one appears in your answer in some form.
Error three: the wrong or mixed register
Register is set by the scenario and it is scored under Task Fulfilment. Writing a formal email where an informal one was required, or the reverse, carries a clear penalty even when the grammar and vocabulary are fine.
The outright reversal is less common than the mixture. An email that opens Hi Ms Okafor, continues I am writing to formally express my profound dissatisfaction, includes my stuff got pretty wrecked and closes Talk soon is comprehensible throughout and still scores badly, because the tone changes four times and the criterion asks for an appropriate tone as one consistent thing.
Mixing usually comes from a practised formal template colliding with an informal prompt. The stored sentences arrive in formal register while the candidate's natural writing is casual, and the seam shows.
Two checks catch it in under twenty seconds. Sweep for contractions: if a formal email has one I'm, or an informal one has none at all, you have mixed. Then read the greeting and the sign-off as a pair, because Hi Sam against Sincerely, Daniel Reyes is instantly visible when the two lines sit together and invisible otherwise.
Point at each bullet and find the sentence that answers it with a number, a name, a date or an observable fact. A bullet with no specific in its sentence is mentioned rather than answered, and that is the most expensive error in the section.
Error four: the survey answer that never commits
Task 2 requires one option selected and supported. The criterion is explicit that arguing for an option other than the one selected, or selecting none, is penalised heavily.
Three versions of this recur. The balanced essay weighs both options for 180 words and resolves nothing. The drift opens with Option A and finishes praising Option B, usually because the concession paragraph was written without a reply. The mismatch argues one option in the text while the selection control says the other.
Behind all three is the same misunderstanding: that the assessment has an opinion about which option is better. It does not. Neither option is correct and the opinion itself is not scored, so hedging protects nothing and costs a quarter-weight criterion.
The fix is two sentences. Name the option by label and content in the first line, and name it again in the last. I would choose Option B, extending the library's evening hours and For these reasons, extending the library's hours is the better use of the budget. Anything between them can concede as generously as it likes, provided the concession is answered.
Error five: the padded opening
The response opens with thirty to forty-five words that contain no fact and no decision. I hope this email finds you well. My name is Daniel and I am a tenant in your building. I am writing to you today regarding a matter I would like to bring to your attention.
In a 180-word budget that is a quarter of the response spent before the reader learns anything. The words have to come from somewhere, and they come from the bullet that later gets one thin clause, which is how error five produces error one.
On Task 2 the same habit appears as a general statement about society, or as a restatement of the survey prompt. Both use twenty to thirty words to tell the reader something they already have.
The rule that removes it: the first sentence after the greeting must contain a fact from the scenario or a decision. I am writing about damage that occurred during my move on 14 March, under booking number MR-40821 is eighteen words and it has already answered part of a bullet.
If you notice padding only at the proofreading stage, cut it then. It is the one deletion that makes the response better as well as shorter, which is why it is the first thing to remove when you are over the word count.
Read the scenario's nouns one final time before submitting. Swapping a supplied fact for a similar one caps Task Fulfilment at 8 however fluent the writing is, and it is invisible unless you look for it deliberately.
Practise it
Five judgements on the errors that cost the most, and on what actually fixes them.