Figure out who'll be there in reality
Here's a secret that I'm honestly not proud of: I am terrible at keeping up with guild forums. I've even worse at keeping up with actually marking certain events as things I will or won't attend. It's some sort of bizarre apprehension about collapsing the waveform or something, and I wind up just laving myself in the "maybe" pile until the event has already started.
Despite all this, I'm generally pretty good about actually attending if I know in advance -- not perfect, but I do make the effort. And while it may be wishful thinking, I suspect there are a lot of other players who do the same thing. For that matter, I know there are a lot of players who will promise attendance at every event but only show up once in a blue moon.
When you're getting your event together, you hopefully know enough about the people involved that you know who's going to actually show. This is important information. Sure, there are people who will throw you curveballs, but you don't have to walk in completely blind. And that ties in very closely to the next point.
Give each player something interesting to do
I'll freely admit that sometimes you don't have this option. Someone might show up at the last minute, and you certainly don't want to have a part of the event that only Mike's character can handle when Mike might not be there. But there's a world of difference between putting in a big barrier to everyone other than Mike's character and ensuring that Mike's character will have something neat to do in terms of advancing the plot.
The size of the event can also pose problems. A plot-based event meant to run for two hours can certainly provide something of interest for six characters. You're going to have a lot more problems with five times that number. But it's still a guideline to aim for, and it should be obvious -- you want every player to log in and jump into action with a feeling of being a participant rather than an observer.
And if you can tailor part of the event specifically to be someone's moment of cool or a chance to advance a long running character subplot? Even better. Remember, whatever the main thrust of the event, there's always room to weave in something else. In the best events, you can fit time in for players to advance a subplot of their own volition, and it'll still look like it was part of the plan.
Don't make likely actions auto-failure
One of my guildmates in Star Wars: The Old Republic plays a Twi'lek who was previously enslaved. Said Twi'lek hates slavers. Absolutely hates them. There is no negotiating with a slaver, no discussing, no reaction except shooting.
As a result, if I ran an event with a guest star who happened to be a slaver, I'd have no one to blame but myself when she shot him because that's what she has to do. This is the most likely course of action for her to take. Building an event based entirely around keeping him alive is only going to end in tears.
You see this from time to time -- an event based around subterfuge with one or two characters who bull-rush through everything, a long chatty scene with players who tend toward the large and silent archetype, or one huge brawl with a bunch of pacifists -- all scenes that really require a group of characters that aren't in attendance.
I'm not saying that these events can't work, but you need to bake-in the assumption that if John shoots first and asks questions later, no matter how many in-character warnings you give, he will walk in shooting. Build up a way to work around that restriction rather than having everything fall apart once John does exactly what you know he'll do.
Have a fair method of resolving conflicts
This is, inevitably, an enormous can of worms.
The fact is that MMOs are set up to handle a few kinds of conflicts well and gloss over most other conflicts. In a tabletop game, this is where those skill points you put into skills other than "hit stuff" would be relevant. But an MMO doesn't have that, and so instead you need some fair form of arbitration. Make success automatic and there's no tension; make failure automatic and everyone is playing an old-style adventure game in which you have to essentially guess what the people in charge wanted in order to advance.
Like a lot of problems, this is one without a clear solution, and it's probably something that I'll write a full article on at a later date. For now, the important takeaway is to have a system, announce it ahead of time, and run with it. People might not like the system you've chosen, but at least they can't complain that you weren't up-front about it.
Feedback can be addressed in the comments below, or you can mail it along to email@example.com. You know, just like every other week. In fact, let's talk about ruts like that for next week because it's worth talking over -- how do you break out of a character rut?
Every Friday, Eliot Lefebvre fills a column up with excellent advice on investing money, writing award-winning novels, and being elected to public office. Then he removes all of that, and you're left with Storyboard, which focuses on roleplaying in MMOs. It won't help you get elected, but it will help you pretend you did.