Remember that an MMO is multiplayer. There are multiple lessons in that one phrase, but I’ll just split this one into two (the next lesson will cover the other half). One of the most important considerations when you’re making a decision about a quest, event, population, or feature is to consider the multiplayer implications of that decision. All too often I see content that isn’t multiplayer friendly; it’s developed with a single player in mind, and as soon as more than a few players are around, problems occur.
For example, content bottlenecks are pretty common with named bosses, particularly with the first major wave of players in any MMO. You get a quest to defeat High Lord Brekhalu and you ultimately find him where the quest said he’d be. Unfortunately, someone’s already fighting him or he’s just been killed. Find a way to fix it. Can you share credit to all players who helped defeat him? Can you trigger a near-instant respawn of the boss if someone who needs him is present? Can you make it so players summon him in some way (e.g. lighting a pyre or hitting a gong)?
Every decision should be made with multiplayer in mind. From the ground up (literally). The environments need to accommodate enough players, the mob population needs to be dense enough and have some form of respawn throttling (low mob population respawn acceleration), events and objectives need to account for multiple people, quests objective numbers need to be tuned against objective availability, intricate scripted sequences might best be left to instances, features should be multiplayer inclusive, etc.
The fact that MMOs are multiplayer is the genre’s greatest strength and greatest development challenge. Leverage it.