I have more or less always run my campaigns as very continuous affairs, in which each game session picks up, in game, only moments or hours after the previous session ended. If the party ends play in the field or in the dungeon, we pick up from that spot.

More and more, as my play groups are harder to assemble regularly in their entirety, it becomes clear that things would be much easier if I were running self-contained adventures with an explicit "return to base" at the end of each session. As I've never worried about forcing any such thing, I'm at a loss as to how to make it happen reliably, although I realize that a good first step is to make it clear to the players that it will help make the game more coherent over time.

Nonetheless, I'm looking for a set of rules to follow that will help me produce appropriate-length sessions without the need for me to end with "…and then you all manage to escape and get back to the safe house." I am especially interested in this in the context of a target of exploration that is clearly too large to explore in one session, such as a large ruin or megadungeon.

  • \$\begingroup\$ Do you have a particular system in mind? I have a feeling answers could vary hugely, particularly given that an 'appropriate length' depends a lot on the complexity of what you are playing. \$\endgroup\$
    – Wibbs
    Apr 6, 2014 at 21:07
  • 1
    \$\begingroup\$ Right now I largely run Basic D&D, but would like to run, in the future, at least FAE and Eclipse Phase. I've run a fairly broad spectrum of systems in the past (dozens, I guess), and am comfortable with the idea of using a different system if I get significant benefits. \$\endgroup\$
    – rjbs
    Apr 6, 2014 at 23:40
  • 2
    \$\begingroup\$ I've had this problem too when I attempted episodic campaigns (in BD&D even). You might want to clarify two points: that a session may have different players/PCs present than the last, and how/why "guaranteed safe retreat" is a problem. I think answers might be missing those issues, which if I understand the question right are fairly core to the problem in need of solving. \$\endgroup\$ Apr 7, 2014 at 6:25
  • 1
    \$\begingroup\$ @Flamma: Megadungeons don't conflict with self-contained sessions if you approach them from a standpoint of "we're going in to accomplish X and then get out" instead of trying to clear or "solve" them. \$\endgroup\$ Apr 7, 2014 at 10:08
  • 6
    \$\begingroup\$ @Flamma Quite the opposite, in fact—Gygax's own D&D campaign was built around a megadungeon that was never fully explored. Think of it as an underground wilderness: do you have to explore the whole wilderness in one session for it to be not a waste of work creating it? \$\endgroup\$ Apr 7, 2014 at 15:22

9 Answers 9


One simple way for sandboxy games: Explain to the players that they need to get back to base by the end of the session. If they don’t, then you’ll roll on a table you’ve made up to determine what happens on their way back to base.

  • 1
    \$\begingroup\$ I recall one of the many OSR blogs I've read had an article and table for just this. I've never tried it, but it did work well for them. Their players knew that lingering in the dungeon, or exploring into someplace without ensuring their escape route was secure, meant they were risking injury, equipment loss, or even death at the whim of one roll. They didn't let the table roll happen if they could help it, and arranged to be home safe at the end of each session. \$\endgroup\$ Apr 7, 2014 at 15:34
  • \$\begingroup\$ I’ve been trying to find an example to link. I thought either Ben Robbins did this in his West Marches campaign (it appears not) or Jeff Reints did in his Wessex campaign (but, if so, I haven’t found the info). \$\endgroup\$ Apr 7, 2014 at 19:31
  • 1
    \$\begingroup\$ I did the same (I thought it was Ben as well), but it must have been someone else. It was a while ago, and I've read a lot of OSR blogs... Edit: Found a post by Jeff Rients on such a table. Not the one I was looking for, but it's the same idea and provides the same motive for the players to not dally in the dungeon near session's end. \$\endgroup\$ Apr 7, 2014 at 19:36
  • 2
    \$\begingroup\$ Here's another example of such a table: thealexandrian.net/wordpress/2149/roleplaying-games/… \$\endgroup\$ Apr 8, 2014 at 10:21
  • \$\begingroup\$ @DaveSherohman That's the one I was remembering! Thanks for the link. \$\endgroup\$ Apr 9, 2014 at 19:00

Use a dynamic amount of content.

You can't know how long that 3 page long dungeon will take to play. Instead prep the games in small chunks. Add more chunks as necessary during the game session. When you're approaching the end point, run the final segment of the game.

As requested, here are some examples. I held off from posting them when I first wrote this up because they're tricks I've used for different purposes. But I think they'd also work well here.

In social or urban games I like to plan ways to interrupt the party. Maybe it'll be an NPC seeking a PC, or a messenger raven, a world event, whatever. When I do these I also make a note of which PC an interruption will address. A PC's best friend, mentor, apprentice, etc does not interrupt the party, he interrupts that PC. The reason I do this is so that if I see anyone losing interest in the game, I'm immediately ready to reel them back in.

I don't think that last half applies to you (although I do think it's a good idea for anyone.) The part I think you can use is to simply have these mini hooks in your pocket. Deploy them as needed.

My other example is actually another implementation of this same idea but I didn't realize it till now because I came about it for a different reason.

I don't like making maps when I prep. Many of the decisions when I map are totally arbitrary, but I spend a ton of effort making sure I transcribe the map perfectly. I prefer to decide on the map at game time.

I still plan for map features in advance. I do this by making a list of the things I want to see in the dungeon. My list will look something like this: 3 kobolds, 1 goblin shaman; 2 kobolds, 4 hidden kobold archers; femur full of teeth marks, pit trap, pit trap w/ trapped kobolds, scrap of paper with elven scribbling, etc, with book references for anything I may need to look up.

As we play I improvise some paths between the rooms. When the players reach a room I look through my list and pick something that feels appropriate. As I said I do this because it lets me improvise the part of the dungeon I don't like to prep while keeping the dungeon from feeling like I'm totally winging it.

I think you could apply it to stretch or compress the length of your games. Make a list of 15 things to put in your dungeon. Expect 10 of them to happen and don't feel obliged to run them all. If the session gets you through only five of them, so be it, the sixth item will be the conclusion to the dungeon, allowing the players to return to town.


Be willing to cut things short

It's okay to end the game early if you're ending on a cliffhanger. Ending with "cliffhanger - reveal - 15 minutes of setup for the next week" isn't going to satisfy anybody.

I'm being pretty loose with the term cliffhanger by the way. What I mean is if you reach a logical stopping point, stop. If the party finds its way back to town, does some shopping, rests up, and then resumes their journey, you're not automatically obliged to keep running. You may have some play time left, but if that play time is going to dig the players into a hole somewhere, you're probably better off stopping in town where it makes sense to end.

  • 5
    \$\begingroup\$ The problem with cliffhangers is that they really don't work when next session doesn't have the same set of players. Changing parties is the situation I understand from the question, and why an episodic structure is being considered. I've had the problem too in a game I intended to be episodic, and solving it would be very useful to make such campaigns less prone to fail. \$\endgroup\$ Apr 7, 2014 at 6:15
  • 3
    \$\begingroup\$ As someone facing the same situation as the asker, the cutting things short suggestion isn't helpful, exactly as SSD describes. The dynamic content is, however. Can you expand on that further? \$\endgroup\$ Apr 7, 2014 at 8:14
  • \$\begingroup\$ Examples provided in edit. I also commented on what I meant by cliffhangers (hint: I didn't really mean cliffhanger.) \$\endgroup\$
    – valadil
    Apr 8, 2014 at 0:55

First of all, you need to choose a game that allows you to do that, which comes with the usual implications of making your group buy in.
If you already play one of those, no problem. If you don't and your group still wants to use that system, then maybe someone else will write an answer about how to shorten combat encounters and create short missions.

This of course also depends on the length of your gaming sessions, but some games where combat takes really long are not really suited to episodic games. I've had three hour World of Darkness games where we did lots of things including taking choices and fighting a demon and a bunch of demoniac knights in Joan of Arc's France and I've had a D&D 3.5e fight where we made one and a half turn of combat in the same time.

Then, you talk to your players and have them know that your objective is to create short missions that they can and should end in a single session. This way, they can all help you bring the session to an end in time.

Last but not least, you should prepare missions that can be brought to an end in time and keep track of the passing of time during the session, making things happen to keep the rhythm up.

As an alternative, you could instead play a game whose structure naturally breaks down into scenes that are considerably shorter than one session. This way you can have a longer overarching narrative, with pieces of the cast going in and out the active scene. Choose a game where the characters aren't supposed to go around in a party and can be effective on their own. More than that, choose one where being there for a session less doesn't hamper you, like not getting the XP for attending the session in nWoD or getting less XP in D&D. Less XP in, say, Apocalypse World is not a problem because of how characters are balanced.

EDIT: Here are some examples of games I organized where episodic structure is enforced:

  • Dogs in the Vineyard has each session (or two, which might be a problem) set in a different branch (a town of the Mormon-like community the game is set in).
  • Trollbabe has the players each chose where on the map they want to start their game, and separate adventures for each player (unless some want to gather on the same place and step on each other toes). Each session presents them with a problem that NPCs urge them to solve and ends with the characters walking away from the place.
  • D&D 4thcore is some sort of D&D 4e challenge that the players need to overcome in a fixed time - it's more a tabletop game than a RPG, though, because every thing you as a player discover in the high mortality dungeon can be used to avoid deaths on the following runs. This might be more a string of one-shots than a real episodic structure.
  • 2
    \$\begingroup\$ Good answer. But IMO the questioner would need some pieces to be better explained. '"you need to choose a game that allows you to do that"' What are those games and what makes them episodic friendly (apart from an agile combat system)? \$\endgroup\$
    – Flamma
    Apr 7, 2014 at 6:53

So, to add to some already great content, here are my ideas about this topic:

Find a reason for them to go back home

You should have a very clear reason to send them home at the end of the session, a reason that you should be so familiar with that you'll be able to adjust it to whatever the characters did to your dungeon in each and every session. It may be a guardian, or a curse, or the common knowledge that in nights bad things are moving through the dungeon and destroy and kill and maim whatever they see. Then let the PCs see what this thing does so they'll know for themselves why they shouldn't bump in the dungeon for the night. This way, it will feel less forced. They'll also still have the option to stay, which means that they chose to stay and thus they carry the consequences.

Always come with a clear ending

You should always know, even before the character sheets are presented on the table, even before the recap begins, how the session should end. The reason for that is quite simple: If you know how it should end, you'll be able to both drive the story to that end while also knowing how to bring them back from there to their home base. It doesn't mean that you should railroad your players and their characters up to that ending, but that if time is short, change a little bit the dungeon so the last room will be this one. Until then, their free to do whatever they want, but when time is short you have to finish it well and believably.

Sometimes they'll suffer from a resources shortage

Resources are a really important thing, and sometimes those will end beforehand. They won't have enough for everything they'll want, and thus will have to come back home. It doesn't have to be food or water, either. Sometimes, they'll have to get something from the city to open that door or to solve that riddle. When done sporadically enough, they won't feel cheated. After all, it is part of the genre.


Every once in a while, let them find a person that they need to bring back to their home base, or a person who they need to turn in to the authorities. When finding those characters, it should be both clear and rewarded that they'll bring them back home as early as possible (and as such is very connected to "know thy ending").


If you want a return to base type thing, I suggest giving the players a keep to come back too. Managing the keep will be a nice way to fade the episodes in and out.

For your mega-dungeon: suppose the place has some sort of uber-guardian spirit that manifests on intruders at random times. To aviod this, one of the players develops a spell that auto-teleports them out of dungeon and back to their keep. Whenever you want to end the session, the thing pops out, the players leave and spend a few minutes playing in their castle, and go back next session.


One idea that occurs to me is that the PCs are staging from some sort of mobile base - one that for some reason must move on at irregular intervals with it's occupants. It does not necessarily have to move far each time it moves, allowing for repeated access to some feature of the world the players/characters are interested in exploring, but the characters will find it difficult or time-consuming to locate the base again if it moves off without them, and will suffer some negative consequences if they are left behind, such as having to go into hiding to avoid the reason the base is moving in the first place at the very least.

Since, from long experience, I know that game-time and real-time do not run concurrently (It is amazing how a few minutes of combat can take hours to play out), it is difficult to put a hard in-game time limit on the characters' activities. This means that real-world time limits must be imposed on the characters as an in-game time limit in some way that is not predictable by the characters until some time after they have departed their base on their adventure.

To maintain the in-character reality while enforcing the real-world time limit, you could have the characters be in possession of some sort of means of communication with their base - it need only be a simple one-way "It's time to return to base" signaling device, but could be more capable if you wanted. When real-world time constraints make it apparent that the session needs to be wrapped up and the characters must return to base, the "Return to base" signal can go off, and any characters who don't make it back by the end of the session could suffer some fate the players would rather have them avoid, and the characters may or may not have found their way back by the next session (depending on if the player(s) show up).

This way, the episodic nature of the campaign will be both enforceable and believable, and it allows for real-world time constraints. It also allows for players who never return - once a player makes it clear that they aren't coming back, even some time later, the possibility exists for the remaining PCs to find the departed player's character's corpse - thoroughly looted and dismembered beyond the possibility of resurrection if the GM so requires. Alternatively, if another player decides to take on the departed player's character, the character could (by GM fiat) have just escaped the "enemy" and returned, possibly with psychic scars that would explain the change in personality that comes with being played by a different player.

  • 1
    \$\begingroup\$ Although I can't apply this to my current campaign easily, it is quite similar to something I've been thinking about doing in a new one. I sometimes think of the silly "timer" from the TV show Sliders. :) \$\endgroup\$
    – rjbs
    Apr 16, 2014 at 2:12

Here is a simple set of rules/guidelines that I follow to make sure my adventures can be played within a time-limit of four to six hours.

Divide the adventure into the following three distinct parts


This is the first part. It is where the players familiarize themselves with each other and their surroundings. It ends with the Presentation of Conflict. For established players in an established world this can be made shorter, but I often allocate about an hour of gameplay to this. It gives me time to show the players the world around them and let them know what is expected of them. An hour and a half can be acceptable if you need to establish a lot, but two hours is almost always too long.


This is the main body of the adventure and also where you can vary the length of the session the most. First you must prioritize. Decide on what info you absolutely need to get to the players and what scenes are absolutely crucial. Plan approximate times for those scenes. Then do the same for non-crucial scenes.

Now, if the players move slower than expected you can always skip scenes that are non-crucial to the plot. They can always be reused in a subsequent adventure. This way you can easily control the time of gameplay and make sure the players always finish in time.

A common rule of thumb is to allocate about one hour per scene, but I personally have trouble with that. It never seems to be correct for me so I just try to make sure that all the crucial stuff can be done within two to three hours and then have a lot of extra padding to add to make the adventure longer.


The End Game (or Climax, or Final Boss) should be short and focused and the aftermath should be efficient. Make sure the ending is clearly an end to the adventure and that you tie up any loose strings that are not part of the continuing plot. Also make sure to mention that the players return to their base afterwards.

If the end of the session is nearing but the players are far from the end you can always speed up things. Perhaps an NPC can give them a copy of the key so that they don't need to explore another dungeon. Perhaps an NPC (that the players trust) can tell them that the key is not really needed, or that the lock can be easily picked, or that there is an alternative way in. If you need to cut the adventure short, keep track of the time and do it before the climax begins. That way it won't feel cut short.

Downtime is playtime too

Let the downtime last for a week or two of in-game time and ask the players what they do during that time. This can help develop the characters as actual characters and not just stats on a paper. If a player is absent during a gaming session, treat that as extra downtime for that character and award a few XP to make it clear that it's OK to miss a session. Not too many though, or your players might get lazy.

Don't lock yourself in too much

Sometimes stuff happens that can't be foreseen. Maybe you need to break up early because someones mom has ended up in the hospital. It may not feel right to just boot that player out of the adventure so it's quite alright to just quit there. Also, it's often better to make the adventures shorter and then talk about what just happened or what happens in downtime or which direction the group is currently headed.

Also, just like the PC:s, you'll gain XP over time and learn how to control the flow of the adventure, so if things don't work out right away you should ask yourself (and your players) why and then try again.

Reuse all the stuff

The great thing about episodic gameplay is the possibility to add reoccurring villains and other NPC:s. If you need to cut a climax short you can always let the players get the upper hand and then let the villain escape. It's a kind of cliffhanger, but it's not dependent on the same players continuing from the exact same spot next week.


I find when it comes to giving sessions a time limit, nothing beats an actual literal time limit.

Each session, set out in that night's story a clear reason for having limited time, set a timer, and run with it. Why the time limit exists should change as often as possible and make sense for the genre. Eclipse Phase: "We have two hours before the life support systems fail and anyone with a biomorph dies." "Whoever did this stole the password hashes and is no doubt running a brute force attack. If we don't catch them in two hours, they'll have the logins to the entire station." Standard Fantasy: "You can see the lights from the ritual! We have two hours to stop them before they raise an army of the dead!" Dresden Files: "I will shoot a hostage every hour until my demands are met..." "The cops will be here in an hour. Once that happens, we want to be gone." "Sundown's in an hour. If we're out after sundown, we're dead meat." I usually set the timer to go off fifteen minutes or so before we need to wrap up- enough for a kind of epilogue of how they did, and a bit of kibitzing before everyone heads out.

When setting up each session's story, make it self contained as possible. The status quo is your friend. Watch some TV to see how this works- If your absent players know the general state of affairs, you can skip any kind of recap of last session. Open each session with the mission of the day and the timer, then let them go at it. One important thing to keep in mind is what I call the failure state- Namely, what happens if they don't accomplish the goal in the set amount of time. I generally prefer to have the failure state not be deadly to characters. (Well, unless death isn't that big a deal. Killing someone's sleeve in Eclipse Phase is fine, killing someone in Dresden Files is not.) If it's a dangerous area (Like being out after sundown) assume that everyone retreats to safety when the clock runs out. I like sacrifices in my games, so I also include (if narratively possible) the option for one last ditch run at the goal before pulling out, a the potential cost of the character. It's been worth it to the player's just the once, but it was awesome when it happened.

One thing you may run into is system dependent, and that's how your system handles time. In FATE, where most of my experience with this is, each turn takes up some vague amount of time that I can wiggle around easily. In D&D 3.x, a round is exactly six seconds, you can set your watch by it. For this to work, you kind of have to fudge the time of turns. I can't think of a situation where this would present a problem, but be aware of it if time is tracked exactly in your system. Otherwise, you may run into a situation where five minutes on the clock gets taken up with twelve seconds of combat.

  • 2
    \$\begingroup\$ Trying to sinchronize the in-game timer with the out-of-game timer like "we only have 6 seconds to play each player's turn" seems unreasonable to me. While the idea is good if you just run with it and ignore the realism implications (the timer ends when game session is over), trying to tie the timer to the real time is, IMHO, recipe for disaster, in every system where you need to think about what your characters need to do. \$\endgroup\$
    – Zachiel
    Apr 7, 2014 at 16:11

This sort of structure lends itself well to a "Western Marches" style game.


The basic idea is that you have a safe base and then the players are venturing out into the wilderness. Because it's unsafe out there and there are no shops the players want to return to town on a regular basis.

You then make sure you structure your encounters so that they can complete on time.

For example you might find that you have time to run 3 normal encounters or 2 long ones in an average session so structure your game about that. Give the players a goal and an incentive to get out, complete the goal, get back.

In the game I'm running at the moment I'm using "bounties" to give them a target. They choose their mission from a list on the board, they go out, complete the mission, come home. Over time the idea is that they will transition to setting their own targets and the bounties will become less prominent but for now it gives them real starting points to get a feel for the world. The world outside the home base is a dangerous place though so they always want to return back at the end of the session to collect their bounty, rest up, and get new supplies.


Feel free to use any of the bounty ideas for your own games and add your own. The idea is to give the players something they can achieve in one session but leave them wanting more.

For example so far they've found a sealed tower, a "door of doom", a mysterious plateau, and much more while completing their missions. At some point they are going to want to go back and investigate all of those in more detail.

  • 1
    \$\begingroup\$ I'm a big fan of the West Marches, but what is not clear from your post or any of the many posts I've read is (a) how to motivate the players to get back to base and (b) what to do if they don't. (These are, obviously, quite likely to relate to one another.) Jeff Rients posted a random table of consequences, which is the closest I've seen. \$\endgroup\$
    – rjbs
    Apr 7, 2014 at 21:58
  • 1
    \$\begingroup\$ This sounds quite unlike the West Marches as described in the original posts—goals and encounters aren't the domain of the GM in a WM game and can't be pre-planned to fit the session. Further, WM used a fixed calendar and a stable of characters per player, so that if a group of PCs were still "in the wild" at the end of a session, they were stuck there until such a time as that same set of players arranged to play again to extract their PCs; and if they got "ahead" of other player groups in time, they had to wait until time "caught up". Those are possible motivators, but aren't mentioned here. \$\endgroup\$ Apr 8, 2014 at 0:23
  • \$\begingroup\$ @SevenSidedDie As I see it the key premise of WM is that the players drive the decisions of where to go and are based from a single location with each game starting and ending there. Each gm should tailor that to make the game they and their players want to play. So far as I know the original WM game did not have a "stable" and certainly neither the one I've played in or the one I'm running does and both those are working well, Take the ideas that fit with what you are trying to achieve rather than completely changing a campaign overnight. \$\endgroup\$
    – Tim B
    Apr 8, 2014 at 6:45
  • \$\begingroup\$ @rjbs I prefer carrot to stick. They want to get back to town to claim their bounty, go shopping, rest up, etc. They had a mission goal and that goal has been achieved...and it's dangerous out there. A couple of times we've had to finish up on the forums but they always complete the objective and head home inside the time limit. The one time they only got half way through a scouting trip and I said if we get the same people for next time you can continue from here, if not you head back to town and set out again and no-one argued. The key thing is to get the players on the same page with you. \$\endgroup\$
    – Tim B
    Apr 8, 2014 at 6:58
  • \$\begingroup\$ @rjbs It can be hard to add into an existing campaign though as you need a suitable base with enough to do around it and incentives to return. \$\endgroup\$
    – Tim B
    Apr 8, 2014 at 7:00

You must log in to answer this question.

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