So I'm starting a new campaign with a group this week, I'm the DM with 4 PC's (2 are long-time RPers and 2 are fairly new to D&D).

I've got notes made for the first session, how the party meets, a structure of sorts, encounters, etc, but I'm not really sure how to follow it up from there.

It's a sandbox style game, so I know it does depend on the players' actions, but I'm just wondering how to lead one session in to the next. What do I need to prepare, how much should I have planned out, how do I build up the story in the first few sessions?



2 Answers 2



Every DM is different. If you are good on your feet and take naturally to story creation, minimal prep is necessary and you pull almost everything out of the air. I learned I need prep, but too much prep for me leads to me wanting to railroad. So, sketches of what might be rather than what the players will do.

For instance, when I was brand new to DMing, first game I ran, I didn't prepare enough and I had a session of climbing over the mountains with every encounter the party ran into was just a bear. It was horrifically boring.

The next time we played, seeing the error of my way, I built a great story thread. It was detailed, it had intrigue it has hooks for both players. The problem was, players don't do what you expect. The first thing the players did was sneak up on the quest giver and mug him. I hadn't planned on that. I didn't know how to adapt. Sadly, more bears ensued.

Some good things you might want to prepare are:

A Main Arc

Even in a sandbox game, you'll want to define overarching goals. It isn't I want them to do A, then B, then C, etc. But it is more along the lines of, "At some point they will run into the NPC Zed. He'll tell them a rumor about a cave that has spatula of power. They will need the spatula of power if they want to defeat Argust the Grand." Now, they can choose to pass all of that up for large swaths of time, but have a main arc. The players can choose to ignore the main arc (like you can do in Skyrim), and a new main story line may emerge but have a arc in case.

Side Quests

Also, build a few one shot like story lines to throw at them in the case of a "stall" in the game. The party has completed a arc or part of an arc, and doesn't want to move on to the next big thing yet, but also seems to be floundering around town fighting over what to do next, and there is a lot of time left in the session. You grab them with an immediate event that requires quick action and leads to a short dungeon. "While your arguing you hear shouts from the wall of the city, guards are lining up on the wall and firing down at something. You hear a loud roar as flame comes licking up the wall and takes down a guard." Or "while you are arguing at the pub, about what to do next, a man at the next table approaches and asks if you can help him with a problem..."

Some of these are just an encounter, others can be a single dungeon, others might be the beginning of longer story lines. And if you don't use them, they aren't wasted, keep them in your back pocket for another session or another game.

Personal Quests

If you have had a session 0, the players emailed you a backstory, you've done character creation individually with the players, or the like... You can begin to weave their stories into your world. One fun thing to do is to tie thier back stories together. Maybe the guy who killed the rogue's parents came from the town that wizard is from. If there aren't backstories yet, you don't need them.


Also, in a sandbox game, it is helpful to have some (but certianly not all) NPCs in the town fairly flushed out even if you don't expect the players to interact with them. It is useful for when a player (who has no goat) randomly says, "I want to buy a harness for a goat." Or, "Is there anywhere to bet on fights in this town?" Certainly, there will be lots of times when you have to create an NPC on the spot, but it is nice to have some interesting and developed characters for the characters to stumble into.

First Session

A lot of first sessions are one shot "warm ups" to let the players interact, learn the team's dynamic, etc. They also tend to be less open ended than later sessions, just so that everyone gets used to DM style and the game -- especially if there are new players.

  • 1
    \$\begingroup\$ Coming back to this after a while, I think your last point is probably one of the most valuable and I ended up doing exactly that - the first session became a small one-shot type session, with a bit of intrigue, some social encounters and some combat encounters, leading into a small mini-boss. It allowed me to gauge the players and their characters and also allowed them to see my DMing style! \$\endgroup\$ Aug 16, 2016 at 12:33

At the Story level:

A model I have learned from other GMs is to have a general notion of, "What happens if the PCs do nothing?" Generally, you want this to be both bad, and interesting. Bad, because you want to provoke the players into opposition, and interesting because until your players start taking the bait, you at least want to have noticeable and interesting situations to describe.

I chose the phrasing here carefully, as did the people I stole it from: "What happens if the PCs do nothing," automatically raises the considerations of what can they do that affects the outcome, and so acts naturally to keep the PCs at center stage of their activity, as agents.

You can apply this line of thought at more than one level: At the highest level, it might be, "Big Evil Dude wins and everyone dies."

But at the session to session level it might be, "Orcs overrun the watch tower," or "Mayor's son gets kidnapped," or some smaller level bad thing. Or, if that does not fit your sandbox aesthetic, but the players have their own goals, then those goals should be steadily receding in the face of player inactivity.

At the Mechanical or Table Level:

As ongoing activities on your part, though, what you should do between sessions involves:

  • Taking stock of what actions the characters did to foil their opponents and achieve their own goals (ideally you should do this while events are fresh in your mind-- write them down)
  • Go through that cycle of how the world will react without the players, in terms of what they did and in terms of all the higher level stuff. This, ultimately, is what you need to prepare for: The things that will happen if the players do nothing, because you basically want the players to oppose that, and you need to know the details and/or mechanics of it.
  • At the start of the next session, remind them of what happened last time and make sure you still understand their goals.

A technique I find very helpful in all of this, if I have co-operative players, is getting one or more players to help with the first and third step. Meaning, if I can get a player to scribe the events of the game, provide them to me, and lead the other players through the recap, that yields all sorts of benefits. (Chiefly, but not limited to, increased player involvement, and greater insight into the player experience and their evolving goals.)

Significant edit/update: If your game is on the extreme end of sandbox, entirely player driven with little or no existential threat working in the background, it is also okay (and in fact, wise) to simply ask your players, "Okay, what sorts of things are you planning to do next session?" Then figure out what will oppose them, and prepare for that.


Not the answer you're looking for? Browse other questions tagged .