You’ve got the rubric open in one tab and your essay plan in the other, and they don’t seem to be talking to each other. ‘Sound structure.’ ‘Critically evaluate the evidence.’ ‘Appropriate use of sources.’ None of it tells you what actually separates a 58 from a 68 on the page in front of you.
A marking rubric university assessors hand out isn’t written to be decorative. It’s the closest thing you’ll get to the marker’s actual thought process while they’re reading your work, if you know how to read it back.
This guide covers the UK grade bands and what each one genuinely asks for, what command words like ‘analyse’ and ‘critically evaluate’ actually mean in practice, a real rubric mapped to an essay plan at three different levels, where the marks are weighted, and a five-minute self-check to run before you submit.
Marking Rubric University Grade Bands: 40s, 50s, 60s and 70+ Explained
UK universities set their own precise wording, but the underlying band structure comes from a shared national framework, and it’s remarkably consistent from institution to institution.
Below 40: fail. The work doesn’t meet the basic requirements of the task, whether that’s missing core content, fundamental misunderstanding, or not actually answering the question asked.
40 to 49: third class. The work shows basic understanding and covers the required content, but stays largely descriptive, restating what sources say rather than doing anything with that information.
50 to 59: 2:2 (lower second). Understanding is solid and some analysis is present, but it’s inconsistent, sound in places and thin in others, without a clear thread of argument running through the whole piece.
60 to 69: 2:1 (upper second). This is where independent analysis becomes the norm rather than the exception. The work engages with evidence critically, not just descriptively, and builds a coherent argument rather than a list of points.
70 and above: first. Everything a 2:1 does, plus genuine independent judgement, awareness of competing viewpoints, and often something that reads as the writer’s own synthesis rather than a well-organised summary of what they’ve read.
What actually separates a low 2:2 from a high 2:1 in practice usually isn’t effort. It’s the shift from accurately describing what sources say to weighing them against each other and reaching a defensible position of your own.
| Band | What the work does | What’s usually missing below this band |
|---|---|---|
| Below 40 (fail) | Attempts the task, but core requirements are missing or misunderstood | Basic understanding of what’s being asked |
| 40-49 (third) | Accurately covers required content | Any real analysis beyond description |
| 50-59 (2:2) | Some analysis, applied inconsistently | A consistent argument running through the whole piece |
| 60-69 (2:1) | Consistent independent analysis, coherent argument | Genuine critical evaluation of competing evidence |
| 70+ (first) | Critical evaluation, independent judgement, real synthesis | Nothing structurally, refinement is what separates a 70 from a 78 at this level |
Two things worth noticing in this table. First, each band doesn’t just add more of the same, it adds a genuinely different skill, which is why simply writing more at the 2:2 level rarely produces a 2:1 mark on its own. Second, the jump from 2:1 to first is smaller in kind than the jump from third to 2:2, which is why students sometimes find the last few marks the hardest to explain even though they feel closest within reach.
Exact percentage boundaries and the precise wording of each band are set by your own university’s regulations, not centrally, so treat this as the shared UK-wide structure and check your module handbook for your institution’s specific descriptors. The QAA’s Framework for Higher Education Qualifications is the national source this structure is built on, if you want to see where it actually comes from.
If you’re on a postgraduate taught programme, the bands you’ll see are usually different from the undergraduate scale above. Most UK postgraduate courses work on pass, 50 to 59, merit, 60 to 69, and distinction, 70 and above, with the fail threshold typically sitting at 50 rather than 40.
The underlying skills each band rewards are broadly the same shift from description towards independent critical judgement, the numbers have just moved up a bracket to reflect the higher baseline expected at master’s level. As with the undergraduate scale, check your own programme handbook for the exact figures your department uses, some postgraduate courses set their own variations on this pattern.
What ‘Describe’, ‘Analyse’, ‘Evaluate’ and ‘Critically Evaluate’ Actually Mean
These four words show up on almost every UK rubric, and they’re ranked by cognitive demand, not just by how formal they sound.
Describe means state what is. ‘Remote working increased during 2020 and 2021.’ True, useful as a foundation, and worth close to nothing on its own once you’re above third-class level.
Analyse means break something down and explain how or why it works the way it does. ‘Remote working increased because lockdown restrictions removed the option of commuting, and because employers who’d previously resisted it had no practical alternative.’ Now you’re explaining mechanism, not just stating fact.
Evaluate means judge strengths and weaknesses against stated criteria. ‘Remote working increased productivity for roles with clearly defined, individual tasks, but reduced it for roles depending on spontaneous collaboration, based on the studies reviewed here.’ You’re weighing evidence against a standard, not just describing what happened.
Critically evaluate means evaluate, then go further: weigh competing viewpoints against each other and justify your own position given the tension between them. ‘While Smith (2022) found productivity gains, this study measured only output volume, not quality or long-term team cohesion, which weakens the strength of that claim when set against Jones’s (2023) longitudinal findings.’ You’re not just judging the evidence, you’re judging the judgements other researchers have already made.
Two other command words worth knowing sit alongside these four. Discuss asks you to explore a topic from multiple angles without necessarily reaching a firm conclusion, closer to evaluate than to describe, but more exploratory. Justify asks you to defend a specific position with reasons and evidence, closer to critically evaluate, but focused on defending one conclusion rather than weighing several open-endedly.
The single most common way students lose marks here is writing description dressed up in analytical language. ‘This shows that remote working is significant’ sounds like analysis because of the phrase ‘this shows that’, but it isn’t one unless the sentence actually explains a mechanism, a weighing-up, or a judgement. University of Bath’s guide to essay instruction words covers the fuller glossary beyond these four, if your rubric uses others like ‘compare’ or ‘justify’.
Command words also drift slightly by subject, even where the underlying demand is the same. A humanities rubric might ask you to ‘critically evaluate’ a historical interpretation, while a science or engineering rubric asks you to ‘critically appraise’ a methodology or ‘critically discuss’ a set of findings. The wording changes, the actual skill being tested, weighing evidence and justifying a position, doesn’t.
If a command word on your own rubric isn’t one of the four covered here, look for it in your department’s own glossary first, then treat it as sitting somewhere on the same describe-to-critically-evaluate scale until you find the specific definition.
A Worked Example: One Rubric Mapped to an Essay Plan
Here’s how this plays out against a real question: ‘Assess the effectiveness of remote working policies for employee productivity.’ The rubric below is a realistic, generic structure you’ll recognise the shape of even if your own rubric’s exact wording differs.
Before mapping a rubric to a plan at all, it’s worth unpacking what the question itself is actually asking, since a plan built against the wrong reading of the question will score poorly on every row no matter how well each individual section is written. ‘Assess’ here is functioning close to ‘evaluate’, judge the effectiveness against stated criteria, in this case productivity, rather than simply describe what remote working policies involve. University of Hull’s guide to analysing essay questions is a useful second resource for this exact step, breaking a question down into its topic, its command word and its scope before you start planning against the rubric at all.
Structure. A 2:2-level plan lists points in the order they came to mind. A 2:1-level plan groups points thematically, for example by productivity type, then by moderating factor, so the argument builds. A first-level plan does the same, but the ordering itself makes an argument, moving from the strongest counter-evidence to the writer’s own synthesis, so the structure carries meaning rather than just holding content.
Evidence and research. A 2:2 cites a handful of sources that broadly support the same position. A 2:1 includes sources that disagree with each other and states that they disagree. A first goes further again, explaining why credible researchers reached different conclusions, different measures of productivity, different industries, different timeframes, rather than just noting that a disagreement exists.
Analysis. A 2:2 explains what each source found. A 2:1 explains why each source found what it found, and starts connecting findings across sources. A first builds an argument that couldn’t be reduced to any single source, genuinely the writer’s own synthesis of the material.
Critical engagement. This is where most students plateau, and it’s worth the most attention. A 2:1-level plan evaluates: it says which studies are more convincing and why. A first-level plan critically evaluates: it also addresses the strongest possible objection to its own position and explains why that position still holds, or honestly qualifies it where the objection has real force.
Concretely, for the remote-working example, a 2:1-level answer might conclude that remote working generally improves productivity for individual-task roles. A first-level answer would take that conclusion and then genuinely test it: what would have to be true for that conclusion to be wrong, is there a plausible confound in the studies being relied on, and does the answer still hold once that’s accounted for. That extra step, actively trying to break your own argument before someone else does, is what critical engagement looks like on the page rather than in the abstract.
Referencing. Consistent, accurate citation in the required style at every level from 2:2 upward is assumed rather than separately rewarded, but consistent errors here are one of the most common reasons a genuinely strong 2:1-level argument gets marked down.
Weighting: Where the Marks Actually Sit
Not every row on a rubric is worth the same share of the final mark, and treating them as equal is one of the more common ways students misallocate their revision time.
A realistic weighting split for an essay like the one above might run: structure 15%, evidence and research 25%, analysis 30%, critical engagement 20%, referencing and presentation 10%. Your own rubric’s actual split will differ, this is illustrative, not universal, but the shape, analysis and critical engagement carrying the most weight together, is a common pattern across UK humanities and social science essays specifically.
The practical takeaway: find the two or three highest-weighted rows on your actual rubric first, and check your draft against those before spending time polishing lower-weighted elements like formatting or a slightly awkward introduction. A beautifully formatted essay with thin analysis will still mark lower than a plainly formatted one with strong analysis, because that’s where the marks actually are.
Weighting also shifts by assignment type, not just by subject. A dissertation or major project rubric typically weights methodology and originality more heavily than a standard essay does, while a reflective piece or portfolio often weights personal critical reflection over external evidence. Always check the specific rubric for the specific piece of work in front of you rather than assuming last term’s weighting still applies.
Self-Checking a Draft Against the Criteria Before Submission
Run this five-minute pass before you submit anything, reading the rubric against your actual draft, not against the essay question.
Go row by row. For each one, ask the specific question that row is really testing, not ‘have I written about this topic’ but ‘does this row’s command word actually happen on the page, in a sentence a marker could point to.’
If the row says ‘critically evaluate’, find the actual sentence where you weigh one source against another and justify a position. If you can’t point to it, it isn’t there yet, no matter how much you’ve written around the topic.
The most common self-check failure is assuming a point is implicitly clear from context. A marker can only credit what’s explicitly on the page, not what you meant but didn’t quite write. If your referencing is one of the rows being assessed, this is also the moment to check it’s actually consistent throughout, not just correct in the bibliography, our Harvard referencing guide is worth a final pass against if that’s your style.
Once you’ve been through every row, ask one last question honestly: which row would a marker credit least, right now? That’s where your remaining time should go, not wherever feels most comfortable to keep polishing.
As a quick reference, the check for each row breaks down to three questions: does the command word genuinely happen in a specific sentence, is that sentence in one of the highest-weighted rows or a lower one, and would a marker who’d never spoken to you be able to see the point without any extra context in your head. If the honest answer to any of those three is no, that’s the sentence to rewrite before you submit, not the one to leave and hope for the best.
Before You Submit
The rubric isn’t a formality sitting above your essay, it’s a description of exactly what the marker is looking for while they read it. Treat the highest-weighted rows as your priority, make sure each command word is genuinely happening on the page rather than implied, and run the five-minute self-check before you submit.
None of this requires more work than you’re already doing. It just points the work you’re doing at what’s actually being marked, so the hours you’ve already put in show up where the marks actually are, rather than being spread evenly across rows that were never going to carry equal weight.
If you take one thing from this guide into your next submission, make it this: open the rubric and the question side by side before you write a word of your plan, not after you’ve finished a draft. Every example above, the grade bands, the command words, the weighting, assumes you’ve read the rubric as a set of instructions rather than a scoring afterthought, and that single habit change is usually worth more marks than any individual sentence-level fix.