How to design a subject curriculum

A subject curriculum is the multi-year plan that tells your department and your students where the discipline is heading. It sits above the scheme of work and below the whole-school vision. At secondary, it usually spans KS3 to KS4, and increasingly into KS5.

A good subject curriculum is not just a list of units in teaching order. It is an argument about what matters in the discipline, what students should be able to think and do by the end, and how each year builds for the next. Designing one forces a department to articulate things that often stay implicit, and that articulation is usually where the biggest gains live.

This guide walks through the decisions involved in designing a subject curriculum from scratch, or in reviewing one that has drifted. It is aimed at heads of department and middle leaders, but ECTs leading a strand will find the same logic applies.

Before you sketch any units, write down what the subject is for in your school. Not a slogan, but a paragraph that a Year 10 parent could read and understand. What kind of thinking does this subject develop? What body of knowledge should a student leave with? What does it mean to be good at this subject?

Christine Counsell's work on curriculum design returns repeatedly to this point: A curriculum is a body of knowledge selected and sequenced with purpose. Subject teachers, particularly at HoD level, are the disciplinary experts, and the curriculum should reflect that expertise rather than borrowing wholesale from the specification.

This is uncomfortable for some teams because it means making choices. You cannot teach everything, and the specification will not tell you what is most important. The act of choosing (and being able to defend the choices) is the first and most useful piece of work.

Every well-designed curriculum has a small number of big ideas that recur, deepen, and connect across the years. In history, these might be: Cause and consequence, change and continuity, evidence, interpretation, significance. In science, perhaps energy, particles, forces, evolution, systems. In English literature, narrative voice, context, characterisation, language and form.

The value of naming the big ideas explicitly is that it gives you a coherent thread through the curriculum. When a Year 7 history unit on the Roman invasion teaches cause and consequence, students should encounter cause and consequence again in Year 8 (the English Reformation), Year 9 (the slave trade), Year 10 (the Cold War), and at increasing levels of sophistication each time. By Year 11, they are not learning the concept for the first time; they are deploying a tool they have been refining for five years.

Mary Myatt's principle of fewer things in greater depth, set out in books such as The Curriculum: Gallimaufry to Coherence (2018) and Back on Track (2020), applies here too. As a working rule of thumb, five or six big ideas tend to be more useful than fifteen. If you cannot name your big ideas without consulting a document, the list is probably too long.

Vertical progression is what makes a curriculum coherent across years. It means knowing exactly how a concept introduced in Year 7 grows into the more sophisticated version a student needs at GCSE, and what the intervening steps are.

This is where curriculum design tends to get hard. Many specifications list end points (the knowledge required for GCSE), and most schemes of work cover the next term, but the bit in the middle (how a thirteen-year-old's grasp of a concept becomes a sixteen-year-old's) is often left implicit. Making it explicit is one of the highest-leverage things a department can do.

A progression map for one big idea, across five years, makes the abstract concrete. The table below shows a worked example for the concept of 'argument' in KS3 to KS4 history.

Repeat this exercise for each big idea. The product is a set of progression maps that, taken together, tell you what your curriculum is doing across the years. The maps will also expose the gaps in the current scheme of work: Where a concept is introduced and then abandoned, where a Year 10 unit assumes a level of fluency the Year 9 units never built, where two big ideas have been collapsed into one and need separating out.

Once you have the big ideas and progression maps, sequencing the units is partly mechanical. Each unit serves one or more big ideas, sits at a particular level of sophistication, and depends on the units that came before. You are essentially designing a directed graph and then flattening it into a teaching order.

A few sequencing principles tend to help. First, build a strong KS3 foundation in the foundational knowledge and disciplinary moves the subject relies on. Students who arrive at KS4 fluent in these basics learn the GCSE specification faster and remember more of it. Second, return to the same ideas at increasing depth rather than treating them as one-shot topics. Third, leave room in the timetable for the things that will inevitably overrun, especially in Year 7 where new students often need longer than expected on foundational material.

It also pays to think about the order in which difficult content lands relative to the school year. New, hard ideas tend to land badly in the last fortnight of summer term, when student attention is at its weakest. Heavy assessment in late November tends to coincide with the highest staff absence. These are not iron rules, but factoring them in produces a more humane plan.

Assessment is part of the curriculum, not a separate thing. Designing it alongside the curriculum (rather than after it) keeps everyone honest about what is actually being taught and what students are actually being held to.

A workable assessment architecture has at least three layers: Low-stakes retrieval embedded in lessons (the bread and butter of memory work), end-of-unit assessments that test the unit's specific end points, and synoptic assessments at the end of each year that test progression against the big ideas. The synoptic pieces are what tell you whether the curriculum is doing what it claims to do; without them, you have unit-level data and very little else.

The most useful kind of assessment data, for a HoD, is not the headline percentage. It is the pattern of strengths and weaknesses across the cohort, and whether those patterns track with the curriculum's stated progression. If Year 9 students are still struggling with the level of argument the Year 8 map said they had reached, the issue is upstream: Either the Year 8 units were not building it, or the Year 9 units assumed too much.

Cognitive science is most useful at curriculum level when it is built into the structure rather than left to individual teachers. The EEF's guidance on cognitive science in the classroom highlights spaced practice and retrieval practice as two of the most robust strategies, and both work better at the curriculum scale than at the lesson scale.

At curriculum scale, spacing means deliberately revisiting earlier content in later units. A Year 9 chemistry unit on chemical reactions should not be the last time a student writes a balanced equation; that idea should turn up in Year 10 (rates, equilibria) and Year 11 (analysis, organic chemistry) so it is retrieved and refined over years rather than parked. Building these revisits into the scheme of work, rather than hoping teachers improvise them, is what turns a sequence of units into a curriculum.

Retrieval practice scales similarly. A department-wide approach to the do-now or starter, with content drawn from a rolling pool of past topics, produces compound effects across the years that no individual teacher's effort can match. The cost is the upfront work of building the question bank; the payoff is years of consistent retention.

A curriculum that only works when delivered by the HoD is not a curriculum. It is a personal scheme of work. A genuine subject curriculum has to be teachable by the whole department, including ECTs and non-specialists where they exist, and that puts real constraints on the design.

This usually means investing in shared resources: Knowledge organisers, slide decks, model answers, mark schemes, misconceptions documents. It also means being clear about what the curriculum requires of the teacher and what it leaves to professional judgement. Both extremes (a fully scripted curriculum that allows no flex, and a thin outline that demands every teacher to invent the lesson) tend to underperform. The middle path (a clear core, with deliberate space for adaptation) is usually where strong departments land.

It also means planning CPD against the curriculum. If Year 9 includes a tricky disciplinary move (analysing historical interpretations, modelling chemical bonding, comparing unseen poems), the department's CPD calendar should have a session on it before the unit lands. CPD that is generic and disconnected from what teachers will actually teach next month tends to have low return.

A subject curriculum is a living document. The first version is rarely the keeper, and even mature curricula benefit from annual review. The risk is the opposite extreme: Reviewing so often that no version is taught twice in the same form and no useful data accumulates.

A sensible cadence is to lock the curriculum for the year, run it as designed, and review it formally at the end with the assessment data and the department's experience to hand. Identify two or three changes for the following year, document the reasoning, and run again. Over three or four cycles, the curriculum will mature into something that genuinely reflects the department's collective expertise.