AI Skill Report Card

Generating Schemes of Work

A-80·Oct 6, 2026·Source: Extension-page
Markdown
--- name: generating-schemes-of-work description: Produces sequenced, spec-aligned schemes of work (topics, lessons, assessment points, homework) for a half-term, term, year or multi-year period across GCSE, A-Level and IB subjects. Use when a teacher or department needs to plan a coherent teaching sequence quickly, standardise planning across colleagues, or build a shared baseline for ECTs/trainees without starting from a blank document. ---
13 / 15

Given a subject, exam board, level/tier and time frame, produce a sequenced scheme of work:

Input: AQA GCSE Combined Science (Higher), Year 11, 3 lessons/week, full year, exam in May
Output: A term-by-term sequence of topics → sub-topics → lesson objectives,
         each tagged to a spec point and assessment objective, with
         interleaving built in and assessment points placed at natural
         checkpoints (end of topic, before mocks, pre-exam revision block).

No preamble — go straight to clarifying the 4-5 inputs needed (subject, exam board, tier/level, time frame, lessons-per-week + non-teaching weeks), then generate.

Recommendation▾
Add an example showing a bad/generic output contrasted with a good spec-tagged output to reinforce the pitfalls section
14 / 15

Progress:

  • Step 1: Gather parameters — subject, exam board (AQA/Edexcel/OCR/Cambridge IGCSE/IB), tier/level, time frame (half-term/term/year/multi-year), lessons per week, known non-teaching periods (holidays, exams, trips)
  • Step 2: Map the live specification — list all spec points/topics required for the chosen board and tier, in the order the board publishes them
  • Step 3: Sequence for coherence — order topics so prior knowledge is built on before it's needed; identify natural interleaving pairs (e.g. revisit algebra skills inside a physics topic)
  • Step 4: Allocate lessons — break each topic into lesson-sized chunks with a single clear learning objective per lesson, tied to one or more spec points
  • Step 5: Weave in assessment — place low-stakes checks (exit tickets, short quizzes) within topics and higher-stakes assessment points (end-of-topic tests, mocks) at logical boundaries, not just tacked on at the end
  • Step 6: Add homework — attach a homework task per lesson or per week that reinforces or pre-teaches, consistent with the assessment calendar
  • Step 7: Tag everything — every lesson row carries: spec point reference, assessment objective, resource/lesson-generator hook
  • Step 8: Output as editable table — topics → lessons → objectives → spec point → AO → assessment → homework, so any row can be regenerated or edited without rebuilding the whole scheme
Recommendation▾
Include a minimal table template/skeleton (columns) in the Quick Start so output structure is unambiguous before reading the full workflow
17 / 20

Example 1: Input: Edexcel A-Level Economics, Year 13, half-term (6 weeks, 4 lessons/week), focus on microeconomics market failure Output: A 6-week table: Week 1-2 "Externalities and market failure" (3 lessons: negative externalities, positive externalities, government intervention), spec refs 3.4.1-3.4.3, AO1/AO2 tagged; Week 3 assessment point (12-mark essay under timed conditions); Week 4-5 "Public goods and information gaps" interleaving a prior-knowledge recap of PED/PES from Year 12; Week 6 end-of-topic test mapped to AO1-AO3, homework set each week alternating data-response practice and essay planning.

Example 2: Input: IB Biology SL, Year 1, full year, 3 lessons/week, two mock exam windows already fixed Output: Full-year scheme split into 6 units matching IB syllabus topics 1-6, sequenced so Topic 2 (Cell Biology) precedes Topic 1 depth content requiring it, assessment points placed immediately before each fixed mock window as revision-and-test blocks, homework tagged to IA skill-building in the terms where practical work occurs.

Recommendation▾
Address edge cases like mid-year transfers, spec changes, or combined/split classes to round out completeness
  • Always tie every lesson to a specific spec point — this is the non-negotiable differentiator from a generic plan
  • Build assessment into the sequence logic, not as an afterthought at the end of each unit
  • Respect fixed non-negotiables first (mocks, holidays, exam dates) before sequencing content around them
  • Default to interleaving related topics rather than teaching strictly linearly, unless the spec has hard prerequisite chains
  • Keep output granular enough to hand a single lesson row to a lesson generator and get a teachable lesson back
  • Make every row independently editable/regenerable — never force a full rebuild for one changed lesson
  • Preserve house style/tone when regenerating for an existing department scheme — ask before overriding established conventions
  • Don't sequence by textbook chapter order if it ignores prior-knowledge dependencies in the actual spec
  • Don't bunch all assessment at the end of term — distribute checkpoints throughout
  • Don't produce a scheme with vague objectives ("learn about forces") — every lesson needs a specific, spec-tagged objective
  • Don't ignore stated non-teaching periods when calculating lesson allocation — it breaks the pacing immediately
  • Don't regenerate the entire document for a small edit request — isolate and regenerate only the affected section
  • Don't omit homework or treat it as generic filler — tie it to reinforcement or pre-teaching of the next lesson
0
Grade A-AI Skill Framework
Scorecard
Criteria Breakdown
Quick Start
13/15
Workflow
14/15
Examples
17/20
Completeness
17/20
Format
15/15
Conciseness
13/15