01The reading conditions.
Marking happens in blocks, because that is the only way it fits into a life. Which means your paper is not read in isolation; it is read against thirty others on the same prompt, and the marker's sense of what a good answer looks like has been built and revised across that stack.
It also means attention is finite and declining. The first pages of any paper get more of it than the last, and a marker several hours in is reading efficiently rather than luxuriously: looking for the thing each row of the sheet asks about, finding it or not, and moving.
None of that is a complaint about markers. It is the condition under which every piece of student work is assessed, and writing as though it were being read slowly and singly is the source of most avoidable disappointment.
02What this does to your writing.
Five consequences, all of them practical, none of them about being cleverer.
- Answer early. A paragraph that builds towards its point over half a page is a paragraph whose point may not be reached. The claim goes first and the support follows it.
- Make the structure visible. Headings that match the sheet's language mean a marker looking for one thing finds it immediately rather than hunting through prose about something else.
- Say what you are doing, briefly, where it might be unclear. One sentence signposting an argument costs almost nothing and prevents an entire section being misread.
- Front-load quality. If your best paragraph is on page eight it is being read by a more tired person than the one who read page one, which is worth knowing when deciding what goes where.
- And be different, deliberately, in the opening. By the thirty-first paper the standard opening has been read thirty times, and a marker will not consciously penalise it but will not be woken by it either.
03The pattern effect, in both directions.
Reading a stack does something to a marker that individual reading does not: it makes the common shape visible. When twenty-eight papers open the same way, the twenty-ninth that does not is noticeable, and noticeable is usually good.
It works against you too. A misunderstanding shared by half a cohort gets recognised instantly by the marker, who has met it repeatedly, and it will draw a shorter and blunter comment by the twentieth encounter than it did by the second. That is not unfairness; it is the sixth explanation of the same thing at nine at night.
The useful conclusion is that the ways your paper is like everybody else's are invisible and the ways it differs are what gets remembered, which is an argument for taking a real position rather than a safe one. This is the same phenomenon read from the other side in the issue about feedback.
04The mechanical things that cost more than they should.
A marker part-way through a stack has limited patience for friction, and friction is entirely within your control. A missing heading, a citation in the wrong shape, a table without a label, a document that opens in an awkward format: each is small and each spends a little of the attention that was going to your argument.
The cumulative effect is what matters. Nobody fails for formatting. But a paper that presents cleanly is read more generously than one that does not, because it is easier to read, and ease and generosity are closely related at hour four.
So the last pass before submitting is not proofreading. It is looking at the document as an object: does it open cleanly, is it obviously structured, can each thing the sheet asks about be found in five seconds. Ten minutes, and it is the cheapest improvement available, in the same way the checks at the essay desk are cheaper than a rewrite.
05What markers actually want to find.
Worth saying, because the picture above can read as cynical. Nobody marks a stack hoping to be unimpressed. The experience of reading a paper that has something in it, after twenty that do not, is the reason people teach.
What produces that is rarely brilliance and almost always specificity: a real example, a genuine complication acknowledged, a position taken rather than surveyed. Those things are available to any student and they cost nothing but the willingness to commit to something.
The paper that gets remembered is not the most sophisticated one in the stack. It is the one where somebody clearly decided what they thought.
Questions at this desk.
Does this mean markers are unfair?
No. It means they are people reading under real conditions, and that writing which ignores those conditions performs worse than writing which accounts for them. Every point here makes a paper easier to read, which is a legitimate improvement rather than a trick.
Should I use headings even where the assignment does not require them?
In most cases yes, unless a discipline or a brief specifically forbids them. Headings let a marker locate what each row of the sheet is asking about without searching, and being findable is worth more than the small stylistic cost in almost every kind of student writing.
Is a distinctive opening risky?
Less than restating the prompt, which is the genuinely risky choice because it is indistinguishable from thirty others. Distinctive here means concrete rather than clever: a specific example or a precise claim. Ornamental openings are a different thing and they do not help.
How much does presentation actually matter?
Directly, very little; indirectly, more than students expect. A clean document is read more generously than a messy one carrying the same argument, because it is easier to read late in a stack. The ten-minute check before submission is among the best-value time in the whole process.
Notes in the margin, from the people at the desks.
READ THE LETTER