It’s not a subscription you’re missing. The IB chemistry HL questionbank problem is a functions problem: questions organized by topic, coverage that maps to the full specification, and solutions that show how marks are actually earned. The official IB Questionbank makes those functions explicit—customizable quizzes filterable by exam session, paper, level, time zone, and question type, all bundled with markschemes and subject reports.
Those three behaviors are the product. A self-built resource assembled from legitimate past papers and teacher materials can deliver all of them, but only if it’s designed around a structure that keeps questions findable, coverage measurable, and solutions instructive. That structure needs a foundation before a single question goes in.
Stage 1 — Build Your Topic Index from the 2025 Specification
Before touching a past paper, build a master topic index against the current Higher Level Chemistry guide. The 2025 specification organizes HL content under two main themes—Structure and Reactivity—with subtopics nested underneath; a practitioner-built resource mirroring this layout, with click-through subtopics and notes under each, shows how workable that structure is in practice. Use the official subject guide as your authority for names and boundaries, and let the structure/reactivity model tell you how granular your tags need to be.
Create a tagging sheet or table with at least two columns: Domain (Structure or Reactivity) and Subtopic, copied from the 2025 guide at enough resolution that any typical exam question clearly belongs to one or two entries. Add an ‘HL-only’ flag for subtopics that extend beyond SL content so those areas stay visible when you review coverage. This sheet is the backbone of your bank. Every question you extract will be tagged against it—and that’s what eventually lets you filter practice by concept instead of grinding through whole papers end to end.
Stages 2 and 3 — Extract Questions, Then Build Transparent Solutions
Most students extract questions. Very few standardize them—which is why most end up with a pile of PDFs they can open but can’t filter, sort by topic, or compare against specification coverage. The fix is a consistent per-question entry format and a repeatable build loop, applied at the moment of extraction, not retroactively.
- ID: [Year/Session]-[Paper]-[Q#] (so you can always re-open the original)
- Tags: Domain (Structure/Reactivity) → Subtopic (your Stage 1 list) → HL-only? (Y/N)
- Question type: calculation / short response / data-based / mechanism / spectroscopy / etc.
- Command-term cue (e.g., determine, explain, sketch) + key skill (e.g., ICE setup, NMR assignment)
- Status: Unattempted / Attempted / Correct / Wrong-by-concept / Wrong-by-method / Fixed
- Solution link: markscheme reference plus your worked-solution notes (method marks + follow-through notes)
- In a 20–40 minute sprint, pick 1 paper (or specimen set) and skim for 6–12 extractable questions.
- For each chosen question, create a bank entry immediately with all of the fields above filled in.
- Add the markscheme reference and write a short worked solution focused on marking logic, not just the final answer.
- Stop the sprint when solutions start slipping—quality of feedback beats raw volume.
- Keep original papers as read-only files and store bank entries separately so the ID alone is enough to reconnect any entry to its source later.
Consistent fields and short build sprints keep the bank searchable and spec-aligned. The system is the easy part. What fills it—and what gets left out—is the harder decision.
Focus Your Extraction Effort in HL Chemistry
Not all topics repay equal extraction effort. In Structure, organic mechanisms, HL-level spectroscopy interpretation (MS, IR, NMR), and extension thermodynamics (entropy, Gibbs free energy, Hess cycles) come up repeatedly. In Reactivity, prioritize kinetics (rate equations and mechanisms, including Arrhenius), equilibrium calculations with ICE tables, quantitative acid-base work (buffers and titration curves), and electrochemistry using cell potentials and Faraday calculations. Tagging these against your Structure/Reactivity index makes coverage gaps immediately visible.
For older or off-spec questions, the triage is fast. Keep a question when its objective matches a current subtopic and the markscheme method matches how you’re being taught. Adapt and label it ‘adapted’ when the skill is current but the wording, context, or data format is outdated; preserve the original ID. Park any question you can’t confidently tag in under a minute into a ‘Needs mapping’ queue—ambiguous classification mid-sprint compounds into lost sessions. Discard anything whose objective would require inventing a tag that isn’t in your 2025-based index, and for topics that are new or reshaped, lean on teacher-produced material rather than stretching misaligned past-paper questions. With that triage logic in place, the bank becomes a calibrated collection—and using it well is a different kind of discipline than building it.
How to Use the Bank You Built
Use the bank differently at different stages of preparation, rather than just pulling random questions. Early on, work by subtopic: filter for a single tag, attempt those questions without time pressure, then compare your work to the recorded solutions at the level of method and mark allocation—not just the final answer. In the middle phase, pull mixed sets combining several related subtopics within Structure or Reactivity so you practice switching between ideas the way higher papers demand. Closer to exams, reserve untouched questions to assemble timed sets that approximate whole papers. Classroom research on practice testing found that more tests were associated with higher exam performance and better judgment about one’s own understanding, and that corrective feedback improved learning further; the worked solutions and status tracking in your bank are what make it that kind of feedback-rich practice, rather than just more questions.
- After each set, log the subtopic tag, your score, and one error label: concept, method, careless, or time.
- Once a week, review the last 30–50 attempts and sort them by lowest accuracy and most frequent error label.
- If any subtopic has fewer than five questions in the bank, schedule an extraction sprint there before more drilling.
Done consistently, those logs convert scattered practice into a diagnostic—a clear record of where understanding breaks down rather than a vague sense of how revision is going.
Why Building the Bank Is Part of Your Preparation
Building and maintaining a question bank is not merely a workaround for not having a subscription; it’s a form of exam preparation in its own right. Classifying questions by specification subtopic, deciding whether older material aligns with the 2025 guide, and rewriting solutions to reflect markscheme logic all force you to think like an assessor rather than a guesser. That assessment literacy—knowing what’s being tested, how, and where every mark is available—is exactly what passive use of a ready-made platform tends to obscure; in the exam room, it’s not a brand that closes that gap.
