So it’s all done, you’ve fixed the immediate problems with your adventures, your tweaked a few things, added some extra details and completed that player handout you promised yourself you would be doing. Finally your adventure is done.
Really, it is.
You’ve done all you can for the adventure.
Seriously, you’re done.
Then why do you have a nagging feeling something is missing, that you missed a flagrant flaw, that the adventure is missing some added combat capability, that a new monster you just read/thought about would suit your adventure, or that you could fit that cool NPC you just made up watching TV?
That feeling is the revisionist in you trying to get out. It’s not a bad one to have, but one you must keep in check. The problem with over-revising is that you may invalidate your play-test! Yes. The last-minute addition you are thinking about may completely change your adventure, its flow, its own "thing". So you have to be careful.
So the question becomes… What do I revise?
The Encounter Flow
What you want to avoid is anything that could significantly alter the adventure or one of the existing encounters. Sometimes the alteration is in-game while other times it is a meta-gaming. As an author, you must seek to keep that flow.
Example One changing an encounter from a bear to a small dragon significantly alters the encounters and leads to a number of collateral damage. Dragons, even small ones, have hoards and thus players want to get their hands on them. Dragons often have servants. But the most important thing is… small dragons usually have a mother. So an encounter where the PCs walk through the forest, fight a bear and keep going turns into a risky and scary affair where they look over their shoulder all the time for Mama-dragon to show up or where they begin searching for a treasure hoard instead of being on their way through the forest... That does alter the flow and the adventure itself.
Example Two we replace the bear with an owlbear. Similar monster and its base purpose (a goon) is unchanged. No one thinks much of an owlbear in a forest. They don’t really have major treasure (except maybe in their stomachs). So we can change the bear to an owlbear without serious alteration.
Sometimes you may alter the flow so that the simplification completely alters the encounter. Reverse example 1. You have a small dragon, making the crossing of the forest fraught with peril and worry about the mother showing up. Instead, with the bear’s appearance the forest trek because rather mundane and not so worrisome. That can also alter the flow.
The story flow
This one is harder to spot and remedy. Here, through the addition or modification of a non-combat encounter – usually a non-combat encounter – the story changes significantly. Many of those are linked to in or out of character knowledge about an NPC or NPC groups.
Example One in the original document, the PCs met with a generic merchant who told them about the king’s evil plot. But you decide that this information should come from the local crime lord whose minions the PCs have been fighting for the last while. His inclusion into the adventure may lead to all kinds of unique tie-ins. Why is he suddenly helping us? Is the information reliable? Should we trust him?
Example Two instead of some existing NPC, replacing the generic merchant with someone you thought of and plan on using later on. Many players will catch on that "NPC joe" has a name and is not portrayed like everyone else. While the players may notice the man, they may simply note him down and continue. This does not alter the flow too much.
What to do?
Well, if your revision caused something you believe might alter the adventure in a significant way, the best thing is to run another play-test. I cannot emphasize this enough.
But if you do not have the luxury of doing that – and let’s face it running two play-tests of an adventure is a luxury, then you must sit back and play the adventure through your head, using the "QA Method". Be VERY thorough and exhaustive when doing that. If you can find a potential flaw, the best thing is to remove your fix. If you think fixing something will make it worse, then it’s not worth doing.
Why drop it? Your text is written taking into account what you had originally, not what you changed and creeps and other mistakes may appear because of the change. Many adventures in LG were branded as poorly edited or written when they were the result of last-minute edits or revisions.