What a mistake tracker has to do
Whatever tool you pick, a study mistake tracker earns its keep by doing four jobs. It has to be fast to write to (under a minute per mistake, or you'll stop), searchable (find every past slip on one topic before a test), analysable (show you which category of mistake is actually costing marks), and it has to bring old mistakes back for re-testing, because a log you never re-attempt trains nothing — that's the core of the error log method.
The three common tools each nail some jobs and fail others.
Paper notebook
Where it wins: zero setup, zero friction to start, and writing by hand is genuinely good for maths and science — you can copy the actual working, draw the diagram, annotate the exact step that went wrong. Many HSC and A-Level students start here, often because a teacher told them to keep a "corrections book."
Where it breaks: everything after the writing. You can't search it. You can't filter it by topic or mistake type. Counting patterns means flipping pages and tallying by hand, so nobody does it. And copying a full question out by hand is slow enough that entries quietly stop after a few weeks. Paper logs almost always die as write-only archives: honest record in, nothing ever comes back out.
Spreadsheet
Where it wins: structure and analysis. Columns force a consistent format, filters give you per-topic views, and a pivot table will tell you in seconds that 40% of your physics errors are unit conversions. If you run the log manually, a spreadsheet is the strongest option — our free error log template is exactly this, with the columns pre-argued.
Where it breaks: capture. A spreadsheet is miserable at holding the mistake itself — handwritten working, a geometry diagram, a screenshot of a question bank answer. You end up writing prose descriptions of visual working, or pasting links to screenshots that live somewhere else. Each entry takes 3–5 minutes of clerical work, and that cost is paid at the worst moment: right after getting something wrong, when motivation to do admin is at its lowest. Spreadsheets also don't resurface anything; the re-test discipline is entirely on you.
App
Where it wins: capture speed and the re-test loop. A purpose-built tracker takes the mistake as it exists — a screenshot of the question, a photo of paper working, handwriting on a tablet — so an entry costs seconds, not minutes. Because entries are data, the analysis is automatic (categories, per-topic counts, weak-spot trends), and old mistakes can be scheduled back at you for re-testing without you remembering to do it.
Where it breaks: a generic notes app gives you none of that — Notion or Apple Notes is just a slower notebook. The benefits only show up when the app is actually built around mistakes: diagnosis, categories, and resurfacing. And any app is another thing to open; if it isn't where you already study, it adds its own friction.
Side by side
| Notebook | Spreadsheet | Purpose-built app | |
|---|---|---|---|
| Time per entry | 2–4 min (hand-copying) | 3–5 min (typing + screenshots) | Seconds (capture in place) |
| Maths & diagrams | Excellent | Poor | Good (photo / ink) |
| Search & filter | None | Good | Good |
| Pattern analysis | Manual tallying | Good, if you build it | Automatic |
| Brings mistakes back to re-test | No | No | Yes |
| Still alive in week 3? | Usually not | Sometimes | Depends on the app |
Honest recommendation
If you want to stay manual: spreadsheet, using our template, with photos of working stored alongside and a fixed weekly review slot. It's the best manual trade-off, and the discipline it demands is survivable if your exam is close enough to keep you motivated.
If you'd rather the clerical work didn't exist: that's the gap errorboard was built for. It captures mistakes where you already study — the Windows app watches your screen on demand, the notebook marks handwriting as you write on iPad, the studio takes a photo of paper working — then writes the diagnosis, categorises it, files it into a per-topic error log and quizzes you on old entries until they're fixed. Individual students start free.
But pick something. The students losing the fewest repeat marks aren't the ones with the best tool — they're the ones whose tracker was still alive in week ten. Choose the option whose friction you'll actually tolerate, and if you're not sure why this habit is worth the effort at all, start with the error log study method.