How Writing Maps to CLB 1 to 12
Bands are easier to understand from examples than from descriptors. Here is one bullet answered at four levels, with the specific thing holding each version down.
How the bands are built
Each of the four criteria is scored from 1 to 12, and the weighted total maps to a level: 10 to 12 is Highly Effective, 7 to 9 is Effective, 5 to 6 is Developing, 3 to 4 is Limited, and 1 to 2 is Level M.
The descriptors behind those numbers are worth stating plainly. At 11 to 12 every bullet is addressed accurately using the scenario's own facts, the vocabulary is sophisticated and correct, and the writing is essentially error-free. At 10, everything is addressed with minor lapses in precision or range. At 9, the same, with slightly less control. At 7 to 8, the response works but has clear gaps: a bullet addressed thinly, a supplied detail altered or dropped, functional rather than precise vocabulary, or noticeable errors. At 5 to 6, several bullets are weak or missing and the errors interfere.
Two consequences follow. The first is that 11 and 12 are genuinely rare and mean near-native control, so they are the wrong target for almost everyone. The second is that the jump from 7 to 9 is not made by writing more impressively. It is made by closing the specific gaps the 7 to 8 descriptor names.
One bullet, four versions
The scenario: a moving company cracked your glass cabinet and lost a box of kitchen items. The bullet asks you to describe the damage and how you discovered it. Here is the same bullet answered at four levels. All four are invented for this lesson to isolate the differences.
Developing, around 5 to 6. When I open the boxes I see my cabinet is broken and one box is not there. It was very bad and I am very upset about this situation because it is my things.
Effective, around 7 to 8. When I unpacked the boxes, I found that my cabinet was damaged. Also, one of the boxes was missing. This was very disappointing because these items are important to me.
Effective upper, around 9. When I unpacked the following morning, I found that the glass door of my display cabinet had cracked across its full width. A box of kitchen items was also missing from the delivery.
Highly Effective, around 11. When I began unpacking the following morning, I discovered that the glass door of my display cabinet had cracked across its full width, and that the box containing my kitchen items was not among those delivered. Your crew leader had assured me that both were loaded last and would be handled separately.
What separates 5 to 6 from 7 to 8
The Developing version has three problems and only one of them is vocabulary.
The tenses do not hold. When I open the boxes I see my cabinet is broken narrates a past event in the present, and the reader has to place it in time themselves. Errors that make a reader stop are exactly what the Readability criterion weighs.
The content is emotion rather than fact. It was very bad and I am very upset tells the reader how to feel about an event they have not been shown. Content and Coherence rewards relevant, supported ideas, and an intensifier is not support.
The phrasing breaks down at the end. because it is my things is a grammar failure that also leaves the sentence meaning almost nothing.
The Effective version at 7 to 8 fixes all three. The tenses are consistent, the sentences are complete, and the reader can follow it without effort. That is what moves a response out of Developing: control, not range.
What separates 7 to 8 from 9
This is the jump most candidates actually need, and the two versions above differ by almost nothing that looks like English ability.
Precision of the noun and verb. My cabinet was damaged becomes the glass door of my display cabinet had cracked across its full width. No rarer word appears. A category word has been replaced by the event, which is what Vocabulary means by accurate word choice.
Fidelity to the scenario's own facts. The 9 version names the box of kitchen items, which is what the prompt supplied. The 7 to 8 version says one of the boxes, which drops a supplied detail. Dropping or altering a given fact is precisely what the 7 to 8 descriptor calls out.
No filler connector. Also, one of the boxes was missing uses an additive signal to attach a second fact. The 9 version attaches it with also inside the clause and keeps the sentence moving.
Emotion removed, consequence kept. This was very disappointing because these items are important to me is nineteen words of feeling. The 9 version spends those words on the second fact instead.
The lesson is worth stating bluntly: the distance from 7 to 9 is specificity and fidelity, not sophistication. Both versions could have been written by the same candidate on the same day.
To move from 7 to 9, replace category words with the event and check every supplied fact appears in your own words. That gap is specificity and fidelity, not sophistication.
What separates 9 from 11
The Highly Effective version adds two things, and both are structural rather than lexical.
The first is that a single sentence now carries two facts in a controlled parallel structure: that the glass door had cracked ... and that the box was not among those delivered. Sentence variety and control are named in Readability, and a sentence that holds two clauses cleanly demonstrates both.
The second is that the final sentence connects the discovery back to something in the scenario, the crew leader. That is coherence in its strongest form: the paragraph is not a list of facts but a small argument in which the damage matters because of an assurance that was given.
What the 11 version does not do is reach for rare vocabulary. Discovered, assured, handled separately are ordinary words used exactly. The band descriptor requires sophisticated vocabulary used correctly, and correctly is doing most of the work in that phrase.
It is also worth saying what 11 costs. That version is longer, and a full response at that density will use most of your 200 words. Bands 11 and 12 are awarded only where no substantive weakness can be listed, which is why aiming for a solid 9 across all four criteria is a better strategy for almost every candidate than aiming for 11 on one.
Using the bands on your own writing
Take a practice answer and score each criterion separately, out of 12, before looking at any total. Scoring the whole thing at once produces a general impression, and a general impression cannot tell you what to change.
For each criterion, write the one weakness that is holding it down. If you cannot name a weakness, the score should be 10 or above, and if you have named one, the score cannot be 11 or 12. That constraint is what keeps self-assessment honest, and it is the same consistency rule the assessment applies.
Then do the arithmetic: content times 0.30, vocabulary times 0.20, readability times 0.25, fulfilment times 0.25. Seeing that a two-point gain in Vocabulary moves the total by 0.4 while the same gain in Content and Coherence moves it by 0.6 tends to change what people practise next.
Finally, rewrite one paragraph at the next band up. Not the whole answer: one paragraph. The 7-to-9 move demonstrated above is roughly fifteen words of change, and doing it by hand a few times is how it becomes automatic under time pressure.
When you self-assess, name one weakness per criterion before writing a number. A criterion with a named weakness cannot honestly be 11 or 12, and that single constraint stops self-scoring drifting upward.
Practise it
Five judgements on what separates one band from the next, using versions of the same bullet.