Watch Where You're Aiming Those Incentives!

The prologue
Squall and his instructor, Quistis, engage a funguar in a training exercise.
Squall: A Funguar? Isn't this guy a bit weak?
Quistis: We're not here to fight him. We're here to learn the Draw system! Focus your mind on drawing magic from your enemy!
Squall: Neat, he has fire and cure magic. Should I toast him with fire?
Quistis: Even better, you can stock up to 100 of any spell you can cast.
Funguar: *hits Squall for 1 damage*
Squall: Uh, I only got like 5 charges of fire for that.
Quistis: Lucky! You get 3-5 per draw attempt.
Squall: So... you're telling me that I have to prolong every encounter I face so I can spend 20-34 turns stocking 100 of each spell the enemy can cast? What happens when I get 100? Do I burn him to the ground with like 100 fire spells at once?
Funguar: *hits Squall for 1 damage*
Quistis: NO, silly, spells are for junctioning to your stats! You link your stored fire spells to your attributes and they get a bonus, but that bonus drops if you spend any of your spells, so you should never actually CAST anything!
Squall: You're telling me that I spent all this time in mercenary training, and it turns out that my only two combat moves are autoattack and drawing curative magic from the enemy for immediate use when I'm injured?
Funguar: Uh... is anyone going to fight me?
Quistis: Yes, Squall, now you've got it!
Squall: Screw this, I hear Seifer wants to duel.

The problem
The problem with misguided incentives is that players will do what the game rewards them for, rather than what the designers had in mind. I don't know if Square envisioned the conversation I wrote above for Final Fantasy 8's "draw" system, but that's the way I played the game, because that's the way the incentives worked out. Why weaken my characters by casting spells, when I could kill the enemy with auto-attacks? The downside is that I will never play FF8 again, because wasting that much time stocking magic spells I could never actually use was not very much fun.

That relatively obscure example aside, there are any number of MMORPG incentive design issues that make games less fun. A few examples:

- Wintergrasp daily quests that don't award credit to players who died fighting and released so they could be revived and rejoin the battle
- Players who go AFK in battlegrounds/scenarios hoping to reap rewards
- PVP siege situations in which enemy armies avoid each other to get a better reward/time ratio by having the battle end more quickly - There's a lot of these, like the infamous Alterac Valley "race" match, the Hellfire Peninsula world PVP areas, or keep trading in LOTRO's Ettenmoors
- Reputations that penalize players for doing the non-repeatable quest content before hitting a certain rep level (Blizzard seems to have mostly fixed this one since TBC - it wasn't fun to reach a new town and think "better ignore all the quests until I get my rep up in dungeons")
- Pressure on newer/less experienced players to perform at a high skill level so that the rest of the group can farm content more quickly (see recent issues with WoW Heroic PUG's, I also ran head first into this problem at a pretty low level in FFXI)

Perhaps someone is going to come into the comments with a lengthy defense of one or more of the above, but the specifics aren't the point. My point is that these players aren't screwing around with the game design, and, sometimes, making harming their own experiences or those of others playing the game, just because they want to mess with the designers' heads. Players make their choices based on the incentives that they're provided with.

Know your core mechanic
Scott Jennings is arguing that companies literally cannot afford the dev time to "finish" their games before releasing them. Perhaps he's right, but these incentive design curves are one area that can't wait, especially in the your game's core mechanic - the main feature that's going to hold your subscriber base.

The core game mechanic in World of Warcraft is actually solo leveling. That's the difference between the hundred thousand subscribers that would have been a hit game in 2003 and the hundred times that that WoW draws today.

If you switch over to LOTRO, you might get the impression that it is also shooting for a solo leveling core. That's what I expected when I tried the game out, and I was deeply disappointed that the leveling content I needed was not ready. Turns out that I was wrong. The core of LOTRO, the thing that they do that no one else can, is creating a virtual world based on the works of Tolkein for their players to enter and explore.

Now we turn to Warhammer, where the intended core mechanic was keep seiges. Some of the issues the game has had in this department were technical, but some of them are tied to incentives. Snafzg is now reporting that players prefer Warhammer's PVE dungeons to its much-touted RVR due to the frustrating random loot distribution system. Meanwhile, Massive Gamer's Warhammer column says that the game has a keep flipping problem, where it's better to let the enemy capture your keeps so that you can get influence for recapturing them.

Of course, neither WoW nor LOTRO launched in perfect, final condition either. However, in their little niches of solo PVE content or Tolkein world building, they were the best on the market. Take away the incentives to participate in Warhammer's non-instanced RVR, which weren't ready for the game's launch and - according to some - still aren't quite right even now, and you're left with instanced dungeons and scenarios, where it is much less clear whether how Warhammer stacks up. (I'd argue that Warhammer scenarios still outclass WoW battlegrounds, but it's a moot point since the lack of other content means that I don't really have a reason to care if I do get to max level via scenarios.)

If you have time for nothing else before your game launches, you have to get the core game working properly. Incentives can be part of the solution, or, if you're doing them wrong, part of the problem.

Bribery comes back to bite
If there's one bit of advice I could offer, recognizing the time constraints on the development process, it's this: Be very careful what you try to bribe players to do.

You CAN, in fact, bribe players to do things that they don't want to do. They just won't like it, and will take whatever steps necessary to get out of it, whether that means players AFK'ing, PUG's requiring raid gear to do non-raid content more quickly, etc. Players don't NEED to try and exploit shortcuts around incentive grinds if you're providing alternative ways to get the rewards by doing stuff they WANT to be doing in the first place.

Players are going to run out of stuff to do and leave your game. You can accept this and let them go, or you can offer up a bribe to try and get your players to try and convince themselves that they actually want to do X portion of your game that they weren't doing because they don't LIKE that part of the game. In the long term, you'd be much better off letting them leave with the fond memories at the front of their mind, so they'll come back that much sooner after your next patch.