Concurrent campaigns happen when two or more groups play in the same campagn settings AND the actions from both groups impact on the setting for both groups. While its common for a DM to run the same setting for many groups at the same time but with none or very limited crossaction, or to run the same continuity for years over many campaigns, running many groups at the same time is both taxing, due to the increased effort and problematic. The outcome is a much more vibrant campaign, with many more details growing organically from play.
Here is the core of concurrency: play the same setting with more than one group and let players change the setting in a way that other groups can perceive and interact with. If you run a sandbox it’s going to be easier than in other styles of campaign, but then again, sandboxes are easier to play with. Easy peasy. You will soon realize though that the whole practice is fraught with problems. Or you might discover that it’s fantastic without any hitch. Anyway, here’s a small list of the problems I found and remedies to mitigate them.
Not enough Content
Two parties go through content twice as fast. And you’re a really busy person. In fact, they don’t. What happens is that, in a sandbox, they will either drift in different directions and stop interact (and then will go through content quickly) or will instead mooch about the trail of devastation and/or consequences left by the other parties. But this trail is, unless you are a consistent shit-hot designer and writer, much better content than most of the stuff you write, so the two groups will faff about in each other’s trail of devastation. Why is it better, though? Mostly because its fully coherent: it’s bourne out of a real story that unfolded itself, leaving bashed doors, mutilated corpses, destroyed fortresses and broken hearts following the criteria that a group of murderhobos would follow. Second, its corpses, ruins, and survivors are interesting because they come with their own drama already. And you dont have to prep it, because the previous parties will have provided all the shenanigans your poor npcs will ever need to have a terrible existence. Third, early groups will follow the most interesting bits of prep and improv you throw at them, and there’s a big change following groups will follow them too.
In short, having a band of adventurers play with your setting is the best prep you can muster, especially if many groups trampled through your campaign already. So, scratch that fear away, multiple parties create a lot of content by themselves.
Timelines & Paradoxes
Different parties do go through in-game time at different speeds. This will cause you headaches and possibly blow your mind open with paradoxes. But it will only if you care about the Game Police.
Let me explain better: it’s not mandatory to completely have the parties act in 100% coherent worlds, and for many reasons.
The first reason is that there is no Game Police telling you that you’re doing it wrong. So try to chill out and enjoy.
Second, most importantly, only the relevant stuff is worth to synchronize. Players are known not to care or notice or misinterpret small details anyway. It usually does not matter if one of the two timelines has an additional three months long trip unless someone cares or its somehow important (and, in that case, you can nimbly reshuffle events between the two campaigns unless there is a post-hoc event).
Third, in case it gets excessively problematic, just bluntly tell the players that, for the purpose of a single adventure or limited time or action frame, you’re taking the liberty of keeping the sessions not synchronized. Just tell them. This is so meta that some of your players might be turned off by this breaking of immersion but, if you want, don’t even tell them. Ultimately concurrent campaigns are a crutch to help you, the DM, run more interesting games. You dont have to justify how you run different groups unless you have a very very complicated social contract with your players. For what they know, the other party is not even real, of its actions are used by you as a suggestion.
The two parties will probably never meet: for logistic reasons, while the PCs live in the same world and maybe in the same city and maybe even work for the same master, they will never meet unless the players happen to be playing at the same time. This can be fixed by careful meta-discourse, the GM running PCs ar NPCs and understanding that the campaign has some brittle spots that are better not be prodded. On the other hand this does not stop players interacting indirectly by, for example, setting each others house on fire. Apparently wanton destruction on absent PCs’ properties is a sport with a tradition of more than 40 years, and who I am to deal a blow to such an important traditional sport?
How to start
Start a steady campaign and, on the side, play one shots in the same whereabouts with different groups, or in nearby places with the same group. The fire and forget nature of one shots make players risk prone and extremely propense to create the trail of consequences described above. Pit them against the mob. Have them disappear up in the mountains. Let them set the woods on fire. Rob a caravan. Murder the Major. Collapse bridges. Destroy dams. Kidnap princes. Make volcanoes explode. Make sure that they super-piss-off NPCs. Let them seed new adventures, then reap the results with your core group.
Once youre fine with random acts of wanton campaign vandalism, simply go bananas with many groups at the same time. The most I had were three groups at the same time: while two adventuring locales were kept effectively separated, the rest of the setting was fully synchronized. This happened even when I was running Western League for two of groups weekly and for the third once every three months. Just dont feel constrained by the relative incongruences: even if two of the groups meet they’ll be busy discussing shared lore, interesting details and experiences and what a douchebag Lord Dude Mc Duderson is rather than nitpick at your shoddy treatment of the calendar or shopkeeper inventories.