How to revise for exams when the lecture slides are all you have

The person who wrote your lecture slides is usually the person who wrote your exam. That makes the deck something no textbook can be: a weighted map of what will be assessed. Three lectures on market failure and half a slide on behavioural economics tells you exactly where the marks live. The topics get examined roughly in proportion to the time they got — and the slides record that proportion perfectly.

So before revising anything, spend 20 minutes turning the module into a one-page map: every lecture title, one line on what it covers, and a rough weight (how many lectures, how emphasised). This map decides where your hours go. Revising without it is how people spend three days on the topic they like and 30 minutes on the one worth a third of the paper.

The weakness of slides-as-course is that slides assert; they rarely explain. The explanation was spoken, and it's gone. Highlighting an unexplained bullet point doesn't change that — you'd just be memorising the assertion, which collapses the moment an exam question comes at it from an angle.

So pass one of revision is re-attaching an explanation to every slide that needs one: what does this mean in plain words, why is it true, why is it here? Do this with AI against your actual deck (fast), a coursemate who understood it (slower), or the recording's relevant two minutes (last resort). The output you want per slide is one sentence you wrote yourself.

Here's the pivot that separates strong revision from busy revision: stop treating slides as something to read and start treating them as something to be questioned on. Every slide or slide-cluster becomes one or more questions:

• A definition slide → “Define X. When does it break down?”
• A formula slide → “State the formula, explain each term, and compute a small example.”
• A diagram slide → “Redraw this from memory and label it.” (Brutal and brilliant for anatomy, mechanisms, supply-and-demand shifts.)
• A comparison slide → “Argue each side, then say which the lecturer favoured and why.”
• An example slide → “Solve it cold, then invent a variation and solve that.”

AI-generated quiz questions get you 80% of this bank for free; add your own questions for anything the lecturer verbally flagged. From here on, “revising topic X” means answering topic X's questions from memory, notes closed, and patching what fails. Reading is demoted to a repair activity.

Two findings from memory research should shape your calendar. First, the testing effect: retrieving from memory strengthens it far more than re-exposure. Second, spacing: three 30-minute retrieval sessions on a topic across a week beat one 90-minute session, at identical total cost. Practically:

1. 01Split topics across days so everything gets hit at least three times before the exam, with gaps between hits.
2. 02Interleave — each session mixes two or three topics rather than perfecting one. It feels messier and scores better.
3. 03Keep a “missed list”: every question you fail goes on it and gets re-asked next session until it dies. The missed list is your revision priority queue, updated by evidence instead of vibes.

Past papers do a different job from your question bank: they calibrate you to the examiner's style — how questions are phrased, how marks are distributed, what “discuss” means in this department. Bring them in once the slides make sense and your question bank mostly passes; used too early they just teach you to reproduce confident nonsense under time pressure.

Do at least one paper fully timed. Mark it against the scheme if one exists; where none does, check your answers against the slides — the marking points are almost always phrases from the deck. Anything you miss goes on the missed list, which by exam week should be the only thing you're studying from.

Days-not-weeks from the exam? Compress this whole system with the triage protocol in our dense-content guide. And if the problem is lectures you never saw at all, start with catching up fast.