The 30-minute order: read your last recap (3 min) → write the opening scene (5) → list three to five likely scenes (8) → name everyone (4) → build the one encounter you are confident about (7) → pick one reward worth finding (3).
If you only have fifteen minutes, do the first four and stop. If you have an hour, add a map, a handout, or a second encounter, in that order.
Most prep advice gives you a list grouped by category: NPCs here, locations there, encounters at the bottom. That is a tidy way to organise a list and a useless way to spend a limited half hour, because it tells you nothing about what to drop when you look up and there are eight minutes left.
What follows is the same material in a different order, sorted by how reliably each piece survives contact with the table.
Minutes 0–3: read your last recap
Before you write anything, re-read what happened last time. Not your plan for last time. What actually happened, including the bits you improvised.
This is first because it is the only step that changes all the others. Prepping without it means prepping for the campaign you thought you were running rather than the one that exists, and it is the single most common reason a carefully-prepared session lands flat.
Write down anything left hanging: a promise a player made, a door nobody opened, an NPC who walked off mid-conversation. Those are next session's scenes and they are free.
Minutes 3–8: write the opening scene
One paragraph. Where is everyone, and what is already going wrong?
This has the best return of any prep you will ever do, because you control the start of the session and nothing else. The players decide everything after minute one; they do not decide minute one. A prepared opening also gets you past the worst part of running a game: the ten seconds of silence at the start where everyone is still finding their dice.
Start in motion. Not "you arrive at the town" but "the gate guard is already arguing with someone about your cart."
Minutes 8–16: three to five likely scenes
Where will they probably go, and what is happening there when they arrive?
Note the word likely. You are not mapping every branch. You are writing down the two or three obvious next moves and giving each one a situation. A scene here is a sentence, not a page: "The chapel: the priest is packing to leave and will not say why."
Three is genuinely enough. Prep decays: scene one is near-certain, scene two is likely, scene three is a guess, and scene four is fiction written against events that have not happened. Everything past that is where prep time goes to die.
Minutes 16–20: name everyone
A flat list of names with one trait each. Ten of them, more than you need.
This step looks trivial and is not. Stalling on a name is the most reliable way to break the atmosphere at a table, and it happens because names are the one thing that is genuinely hard to improvise well under pressure. Four minutes here buys you a whole session of never saying "uhh."
One trait, not a personality: "Hessa, talks too fast when lying." That is enough to play a full scene from.
Minutes 20–27: the one encounter you are confident about
Not three encounters for three possible routes. One, for the thing that is most likely to happen.
Have the stat blocks to hand, and write two extra notes: what the enemies want, and when they break. Monsters that only fight feel flat, and a fight with no retreat condition is the thing most likely to eat the back half of your evening.
If a fight happens somewhere you did not expect, you can reskin this one. That is much faster than building a new one at the table, and nobody has ever noticed.
Minutes 27–30: one reward worth finding
A magic item, a secret, a favour owed, a name that unlocks something later. One thing that makes the session feel like it moved.
Sessions without a reward can be perfectly good and still feel like nothing happened. This is cheap insurance against that.
This as a checklist you can tick Pick your session type and how long you have: combat, intrigue, dungeon, session one, finale. It cuts from the bottom. Free, no account. →What to add if you have an hour
- A map, but only if a fight or chase is likely somewhere specific.
- A handout or prop. High effort, high impact when it lands, first thing to cut when it does not fit.
- A second encounter, for the second-most-likely route.
Notice what is not on that list: backstory, written dialogue, and the second half of the session. Those are the three biggest consumers of prep time in the hobby and the three least likely to reach the table.
Why this order and not another
Because the order is the method. Any list of prep tasks is roughly the same list; what makes a time-boxed prep work is knowing which end to cut from, so that running out of time is a normal outcome rather than a failure.
Your own hit rates will differ from these. A heavy intrigue campaign uses NPC prep far more than a dungeon crawl does. If you want the version of this order that is about your table specifically, measure a few sessions and find out which category you personally waste.
How much D&D prep actually gets used? Which categories survive contact with the table, which get thrown away, and how to find your own numbers. →The step this whole method depends on
Everything above is built on minute zero: having a recap of last session to read. Without it, step one becomes twenty minutes of trying to remember, and the half-hour is gone before you write anything.
Most DMs know this and still do not write recaps, because writing one up after a four-hour session is the last thing anybody wants to do. Which is a problem worth solving separately from prep.
Scrivorn Record the session and it writes the recap for you, keeps your campaign wiki current, and gives your players a portal to read between games. →