One thing I’d add is that, for shorter games, it’s vital that the players leave with a feeling of accomplishment. They don’t have to solve the mystery or fight the big bad, but they should feel like they end the session in a better position than they were when they started it.

Many of my previous weeknight (short session) games fizzled because the players felt like they were spinning wheels all session.

As a GM, you may want to metagame a bit and help the players get to that feeling of accomplishment. Even players that normally want to work everything out on their own appreciate a bone when they know they don’t have a lot of time.

We handle pretty much all of our in between adventure activities via email. Most people start showing up an 30-60 minutes before game time. Everyone usually brings their own dinner or some will order a pizza (we game at our FLGS).

In a typical session there will be at least one encounter that is non-combat with 2-3 combat encounters. I keep a flow chart clipped to my DM screen to keep things focused. Although I don’t use a timer during combat, I do expect my players to be ready on their initiative. With six players, they all have plenty of time to plan their action. If they hem and haw, then I put them on delay.

I do this every week, and some of the earlier posts here have been super-useful.

1) No wandering monsters. Things should be there for a reason.

2) You DO need that socializing half-hour.

3) Index cards for gear really helps. In fact, reducing paperwork in every way is a bonus.

4) If information is vital, write it down, and hand it to the characters. You can still role-play it out.

5) Know when to fast-forward. It takes Indiana Jones two days to fly from America to Egypt. Can you imagine if the movie made you watch that?

6) Therefore, start with something exciting, whenever you can. It’s also a clear end to socializing/eating that way.

7) END WELL. Try to have each session have a good ending, so that it’s memorable. Don’t have time for one more encounter? That’s cool. Have it show up, roar, and then say “see you next week”.

Still you have the definite time limit. I think one of the biggest things that can help a weeknight game go quickly and smoothly is the gaming space. If you have a nice neutral space, set up to accommodate players, no TVs or video game systems going, then you should be able to get in under way and keep focused.

You need 30 minutes of socializing. Every time in my life I have tried to cut this time has resulted in pain. You just can’t cut this. One trick is if you include dinner in game night, then that 30 minutes can happen during dinner. But then you need to be prepared to have a bit of socializing happen after dinner because someone will want to arrive just in time for the game, skipping dinner (perhaps because of work hours, perhaps because they desire other dinner arrangements).

It should be noted that Swordreader’s game doesn’t require this socialization time, because the players have the opportunity to socialize outside the structure of the game. That was probably also true of Mike Mearls’s lunchtime D&D.

If you seriously want to aim closer to a two hour session than a four hour session you will need to take care in chosing the game. Two hour bits of D&D with all it’s miniatures glory are going to be tough.

Frank

1. We played twice a week.

Everyone involved (except one of the players) lived and worked together. Scheduling wasn’t difficult so we went ahead and played twice a week. This allowed us to…

2. Focus on the game.

We weren’t socializing, we weren’t catching up, we were playing. Much in the same way you’d sit down with your family and play a game of monopoly or something.

3. We played for short amounts of time.

The longest amount of time we would play (aside from the grand finale) was three hours. This gave me enough time to pack in story, combat, and juicy tidbits for the players and still left them wanting more. This kept the play very relaxed. Though, this really only worked because of point number one.

4. I created “sections” to each session.

Any given evening I would have anywhere from two to four “sections” of the campaign that I expected the players to get to. Each section was designed to be a good stopping/starting point. If my players were having fun with a particular section, I would simply extend it out until our time was up and wait on the next ones until the next time we played. If we were nearing the end of our time when they finished a section, we’d quit early. Maybe not the most efficient way to do it, but it worked well and we all had fun.

5. We didn’t care.

No, seriously, it’s true. We didn’t worry about the plot moving forward. We didn’t worry about getting “quality” play time in. We just new we had two or three hours to play and enjoyed it as much as we could. I personally think that was the key, right there.

I don’t push my players to make decisions, although sometimes I do prompt them to declare that they’re actually doing something when they’ve decided to do it. My group is surprisingly reticent about actually saying “I climb through the hole” even though they’ve clearly made up their minds to do it.

I do the opposite of the heavy-action game. Probably a good 85% of my game is straight RP. One combat per session, tops. But this may be because combat is rather awkward in a chat room.

One of the really superior things about online chat is that you can “background prompt” people by opening an IM screen. Thus if someone is having trouble with something, you can shoot them a message with what they need to know. It saves a lot of time and confused out-of-character speech.

I also worry less about getting stuff done than keeping people engaged. As long as no one is dropping out due to boredom, I don’t care if my idealize scenario takes 1 session or 5 to unfold. If they start taking a long time, I may throw in a few extra steps or a combat to liven things up.

We play bi-weekly evenings, usually 3 to 3.5 hours. I find it very important to get a lot of “story meat” covered in each session, or the players completely lose track.
Before the session, I create a brief outline with my assumptions of what will happen and how long it would take. During the session I keep a close eye on the schedule.
In order to keep the game dynamic I put a time limit on every decision. When players start discussing options, I usually tell them “2 minutes to decide one way or the other” at some point.

During most combats the decision making time is reduced to 30 sec or you lose your turn, and no player to player discussion during combat.
In Climax encounter we usually take more time to resolve the entire combat.

We hardly ever use random encounters. If something happens, even if it appears random, the players know it will be important to the plot. The downside is that it is not “realistic”, the upside is that almost every minute of the session contributes to the story.