Most adventure modules come in a familiar format, with the author having written the story well in advance. The characters either fit that story or they do not. The format is familiar and typical of M.A.G.U.S., Call of Cthulhu, Shadowrun, and D&D adventures alike. In many situations, people try to do something about its shortcomings. D&D emphasizes its heroic nature and tries to establish clear sides of good and evil; characters siding with the cultists are not typical in Call of Cthulhu either. In Shadowrun, the group concept settles many things, and switching sides is difficult.

I know: in many respects, M.A.G.U.S. seems to be the exception. A priest of Ranagol, a Gorvikian assassin, a witch who also follows Ranagol, and a thief working with their group are unlikely to start "hunting Ranagolites" in the north simply because a few local priests have to talk about how Ranagol ought to be persecuted, since his religion practices human sacrifice. A party on Ynev can take many different sides, with severe conflicts even within countries and religions. The question is whether an adventure can be written in a way that will almost certainly work despite this. And if so, what should such an adventure contain?

Regarding the latter, it certainly needs a map of wherever the adventure takes place and some backstory about who opposes whom. But here it is worth developing every side a little, not just the originally intended target. We can flesh out the reasons for the local stalemate, individual non-player characters, and much else. Whether this is a M.A.G.U.S. adventure built around conflict between trading houses, or a Call of Cthulhu struggle between bootlegging gangs, cults, and Prohibition enforcers, this much is shared. So is the possibility of preparing a table or two.

In these, it helps to list roughly how much each side's individual resources are worth, as well as how they would try to stop the party after suffering losses of a given size. Now they merely react to the player characters. To introduce the adventure, we can use rumors, posted notices, or even a "classic employer." For each area and group, it is also worth marking how the characters can discover its location, recording plenty of threads to follow through the adventure.

Lo and behold, we no longer need to assume either that the characters take a particular side or that many events occur in a predetermined order. Much of the story will depend on the characters' actions. The basic system is therefore relatively simple. Alongside this, we can assess what the characters stand to gain in each place and how they can capitalize on their victories. Loot, territory they can acquire, reputation... These are easy to document. It is worth spicing this up a little with how each group tries to recruit, when they fight one another, and how the world will function.

We can also describe who moves where and the characters' daily routines. Describing the environment is not difficult either, providing a little "sandbox" experience. This is only a basic adventure structure, but its point is that the participants live and react to the party, and many things do not have a fixed written sequence. Let us go further. At each location, we offer several possibilities for attack and for defending one's own forces. The adventure can quickly be expanded, for example by having the different sides recruit when necessary or possible. The party can protect its allies. In that case, we must assess each side's reserves and how well their troops fight—how well they are led. 

In other words, we can give character sheets not only to characters and NPCs but to the different sides and organizations too. Shadowrun already offers a solution for corporations, organized crime, and gangs, but the same can be done for local interest groups using statistics the players can influence. The smaller scale and faster change call for a new system. For example, we can introduce the following attributes and associated checks, independently of the game system:

Manpower: The size of military reserves. A successful check lets the group replace a previously lost platoon. Replacing larger units is more difficult. If this is needed regularly, the value may decrease. 

Supplies: This covers replacing weapons, food, money, and goods. A check suffices to replenish a stock equivalent to a small shop's storeroom. It decreases with regular use.

Influence: A check can obtain help from a small squad of law officers or locals, three to five people. Asking for many such favors naturally reduces the attribute. It also depends on the group's reputation. 

Recruitment: This produces recruits rather than trained troops, but potentially two platoons' worth at a time. Supplies still provide their equipment, of course. This too decreases with regular use and naturally depends on reputation.

If attributes are rated from 1 to 6, they drop by one point after three uses. If you prefer a 3–18 scale, they drop after every use. An attribute that does not need to be used "regenerates" by the equivalent of one use per month, starting forty-five days after its last use. 

This is a simplified basic system with which you can run many stories in many systems and worlds.