An annual paper covers the full year’s syllabus, carries internal choice in the higher-mark sections, and produces a result that goes on a record and informs promotion. Its central difficulty is weighting: representing a whole year fairly when some of it was taught eight months ago.
We do not write the questions. TurtleDock holds no question bank: this gives you the paper’s structure and the marks arithmetic, and the items are set by the subject teacher against the syllabus actually taught.
An annual paper decides something. That makes it a different object from a unit test, and the difference is not size — it is that the result becomes permanent and gets read by people who were not in the room.
Which means two obligations. It has to be defensible: every question answerable from what was taught, every mark traceable to a scheme. And it has to be comparable, so this year’s result means roughly what last year’s did.
The default drift is towards the last term, because it is fresh for whoever is setting the paper. It is also the material students have had least time to consolidate, so a paper weighted that way measures recency rather than learning.
Choice is where annual papers become unfair, and almost always in the same way: two alternatives that are not of equal difficulty. The student who picks honestly loses to the student who picks well.
A school that cannot compare this year’s results with last year’s cannot tell whether anything it changed worked. That requires the blueprint to be stable and any change to it to be recorded.
It also requires the grading scheme to be versioned by academic year, so reprinting an earlier result reproduces what was actually issued rather than recomputing it under today’s rules.
Phrased the way they arrive, answered so each one stands on its own.
In proportion to the syllabus, not to how recent it is. If the final term is a third of the content it is roughly a third of the paper, regardless of it being freshest in the setter’s mind.
Someone other than the person who set them, before printing. The setter knows what they meant, which is exactly why they are the wrong person to judge whether two alternatives are equally hard.
Grading schemes are versioned per academic year, so a result reprinted from three years ago comes out under that year’s scheme and format rather than being recomputed under the current one.
Checked by the Assessment desk on 8 September 2026. Corrections go in the page, dated, rather than quietly.
Pricing is on the page, in the open, with the counting rule written next to it. Nobody has to book a call to find out what it costs.
See pricingForty minutes with an implementation engineer — your classes, your fee heads, your grading scheme, entered as we go. No slides, no sample school.