How to supercharge your learning with Mark
A practical study system built on what Saanora actually does — projects that hold your syllabus, explanations you can drag, quizzing that fights the fluency illusion, and a Mark that tells you where you're wrong.
There is a specific trap in studying with an AI, and it has almost nothing to do with cheating.
It is this: a good explanation feels like understanding. You read it, every sentence lands, you nod along, and you close the tab convinced you know the material. Then something asks you to produce it from nothing, and it is not there.
Psychologists call it the fluency illusion. It is not new — it is why re-reading your notes feels productive and mostly is not — but a fluent assistant makes it much easier to fall into. Everything below is a way of using Mark that works against it rather than with it.
1. Start with a project, not a chat
The single biggest upgrade is not a prompt. It is putting the subject in a project before you ask anything.
Attach the syllabus and the lecture slides to a Saanora project, then keep every chat for that subject inside it. Mark reads those files once, at upload, and every conversation in the project starts already knowing them — including conversations you have not started yet, in December, about a chapter you have not reached.
This changes the answers, not just the filing. “Explain diffusion” gets you the internet's idea of diffusion. The same question inside a project with your syllabus attached gets you your course's scope, your lecturer's notation, and the depth your paper actually examines. Add a standing instruction once — “third-year syllabus, worked examples before theory, show every step” — and it applies to all of them.
2. Ask Mark to show you, not tell you
When something is spatial, dynamic, or structural, ask for it in that form. Mark can build the explanation rather than narrate it:
- “Show me this as a 3D model I can rotate” — for molecules, anatomy, geometry, orbits.
- “Make it a simulation with sliders” — for anything where you need to see how the system responds to a change.
- “Draw the relationship as a diagram” — for processes, hierarchies, and proofs where the connections are the point.
- “Walk me through it one step at a time, and stop at each step” — for derivations you keep losing halfway through.
A wall of text puts the work of finding the structure back on you. A picture you can turn hands you the structure directly, and the version you build by dragging something tends to survive the walk to the exam hall.
3. Make Mark quiz you before you feel ready
This is the step most people skip, and it is the one that does the work.
The difference between reading an explanation and knowing something is whether you can retrieve it with nothing in front of you. So stop asking Mark to explain, and start asking it to test you. “Quiz me on chapter four.” “Make flashcards from these slides.” “Ask me ten questions and don't give me the answers until I've tried.”
Inside a project this gets sharper, because the questions come from your material rather than the general topic — and getting one wrong and finding out why is worth more than three correct explanations you merely read.
4. Explain it back and ask where you're wrong
Read the explanation, close it, then write the idea out in your own words and paste that back with: “where am I wrong?”
Saying something out loud — or writing it cold — exposes the parts you were skating over. It only works if the thing marking you is willing to disagree, which is why Mark is built to say plainly when you have got something wrong rather than finding the generous reading of it. An assistant that agrees with everything is a very pleasant way to learn something incorrectly.
The gap between what you wrote and what was actually there is your real revision list. Everything else is re-reading.
5. Check what Mark actually read
If you are working from slides, a scanned chapter, or a spreadsheet of data, check what got through before you trust the answer.
Open the file in the canvas beside the chat and use “See what Mark read”. It is a receipt, not a preview: if a section is missing there, Mark could not see it either. A slide whose entire content is a screenshot arrives as an empty slide, and that is exactly the kind of thing you want to discover now rather than the night before.
6. Turn each session into something you keep
At the end of a good session, ask for the artefact: “turn this into a one-page revision sheet as a PDF”, “make a deck of the six things I kept getting wrong”, “put the formulas in a spreadsheet”.
Mark builds real files — PDFs, Word documents, decks, sheets — not text you have to reformat at midnight. They land in your Library alongside everything else, so the work compounds across a term instead of dissolving into scrollback.
7. Let memory carry the boring part
You should not have to re-establish who you are every week. Mark remembers what you are studying, where you keep going wrong, and how you like things explained — across chats, and across the whole term.
That means the fourth session can open with the weak spots from the first, and a fact you mentioned in September is still known in December. It is also fully readable: Settings → Memory shows every line, and you can edit or delete any of it.
What a week actually looks like
Put together, a week in one subject looks like this — and none of it takes longer than what most people already do:
- Once, at the start of term: make the project, attach the syllabus and slides, write one standing instruction.
- After each lecture: one chat in that project. Ask for the hard idea as a diagram or a model, not a paragraph.
- Same session, five minutes: close it and write the idea from memory. Paste it back and ask where you are wrong.
- End of the week: “quiz me on everything from this week, hardest first.” Get things wrong on purpose, early.
- End of the month: “build a revision sheet from the things I kept missing” — and keep the PDF.
The one habit that matters most
If you take one thing from this: the moment an explanation clicks, immediately try to state the idea without looking.
If you can, the understanding is real and you have just strengthened it. If you cannot — and the first few times, you often cannot — you have caught the illusion in the act, which is precisely what you want happening at your desk rather than in an exam hall.
That single check is the difference between an assistant that explains things at you and one you actually learn with.
One honest note to finish on. None of this is an argument that Mark does the learning. Used well it is the most patient tutor you will ever have — it will explain the same idea six different ways at two in the morning without once sighing — but it cannot do the retrieval for you, and the retrieval is the part that turns into knowledge.
What it can do is make every part around that faster: the context you stop re-explaining, the explanation built in the right shape, the questions that find your gaps, and the record that survives the term.