
Writing brings out the surreal…
Source: http://surrealistisch.blogspot.co.uk/2009/11/roberto-kusterle.html
The hard part of writing a scenario is writing it.
Having noted facts and names from HP Lovecraft’s Arkham and written down bullet points for key people, locations and events, I now start writing in the blanks.
I start at the top, writing a ‘background’ section for the Keeper. This basically explains the recent events that lead up to the involvement of the player investigators, who the main characters are, what they have done and anything else important (such as the potential consequences if the player investigators do nothing).
As the scenario exists only in my head at this point, as I write many ‘new’ ideas rise to the surface – if it feels good and fits, then I write it in (going back over the preceding text and back-filling or altering as necessary to ensure the new idea fits). If the idea is good but doesn’t fit, then I make note somewhere else and may one day come back to it for a different scenario.
I usually focus on the facts first, people and places. Each one is written up, why are they involved and what might they do – this latter bit breaks in to two parts: a) what the NPC will do during the scenario (i.e. he will do X that will cause Y), and b) what the NPC will do if the player investigators meet him or her, or what he/she will do if the PCs are liable to mess up the NPC’s big plan etc.
At some point, I’ll spend a day or two stating-up the various NPCs (I find doing all these in one go easier and give my brain a break from being ‘creative’).
Once I have written out all the NPCs and locations, I then write up the key events – potential major events (Y happens if X happens) and so on. Once this is done, I fill anything required, which might include extended Keeper notes, an introduction, a conclusion (including player rewards, possible outcomes – ways to solve the scenario), and player handouts.
So it’s written. I’ll then read it through, correcting and aligning contents – deciding if events should be placed after locations or whether they should be combined – this amounts to presenting the information in a logical play order – making sure the information is accessible to the Keeper in the middle of the game.
It’s at this point that you might realise that your clue quota is insufficient, i.e. you have inadvertently made a ‘clue bottleneck’, where an essential clue can only be found in one situation or place. If discovered, then I’ll look for ways to place the clue (or variations on it) in multiple locations. Some clues, however, are ok to be left as is; if the PCs don’f find the clue it will not derail the scenario.
Next step – playtesting.

Testing, testing…
Source: http://www.doctormacro.com
I’ll gather my group together and run them through the scenario, having first prepared a set of pre-generated investigators.
This is the chance to make sure the scenario works. Do the players do what you expected them to do? Did they visit all of the locations you prepared? Did they meet all of the NPCs?
More importantly, where did they go and who did they seek out that you hadn’t prepared? I found that when I ran Missed Dues the first thing the players all decided to do was speak to someone I hadn’t even considered writing in to the scenario. I couldn’t believe I had missed so obvious a person!
Whilst playing, I make notes in the scenario – new names, places, events, that will act to remind me when I come to revise the scenario after the playtest.
Often the most difficult bit is determining what to add in and what to leave out after the playtest. You have to take the middle road, adding in stuff that you feel the majority of groups will need, leaving out the less likely bits (otherwise your scenario will begin to look like a novel). Wherever possible I will try to include as much as possible without over doing the write-up, perhaps using bullet points to list some of the possibilities, allowing the Keeper to embellish these as they need or ignore them. For example, in the scenario the PCs will visit a certain building at the climax, however they really only need to go into one room (and will be directed to do so). So I’ll write up a full description of that room and what happens when the PCs go in there, whilst only listing a few ideas for the other rooms in the building in case the PCs decide to be more cautious in their approach. The Keeper then has enough ideas to rift from without me giving them an additional 5 pages of text, which for most groups will be ignored anyway. This is particularly relevant because the scenario is intended for convention play, i.e. within a four hour time slot.
Ideally, I’ll try and playtest the scenario will at least two different groups of people to ensure I’ve got it right. If possible, I might have another Keeper read and run the scenario too – this is very useful when you intend to publish the scenario.
Thus, having playtested and revised the scenario accordingly, it’s pretty much done.