For a game that depends on a stream of income from subscribers or RMT shoppers, the first hour of play must be the top development priority. This is where you hook players. After that, the endgame is important because that is where your players will be spending time indefinitely and where your game’s chatter will come from in the long run. Next is the early game, when you build momentum. The mid-game has already fallen this far down the list, as you have certainly seen in a lot of MMOs, and frankly few care much how good the late-game is because they are already fully committed and racing for the end-game.
I stand by my repeated claim that optimizing the new player experience is of paramount importance. You must grab my attention within five minutes, and you must deliver a satisfying hour or two for my first play session. Without that, any free trial is worthless, and you may even lose some people who have thrown down $50 for a box. This is the part of the game that every single player will see on every single character, and if you cannot do a good job here, I have no hope for the rest of the game. Yes, it is hard to make things interesting while giving the player only a few buttons to play with. Suck it up, we all have hard parts in our jobs. That’s why they pay us.
In retrospect, the original Asheron’s Call tutorial dungeon was truly horrible, only tolerable because these MMO things were just so new and exciting. The Lord of the Rings Online™ does a great job with its introductory instances, basic gameplay while introducing the setting and giving you some big name characters, and you don’t realize it is foreshadowing when an NPC saves the day while you watch. City of Heroes has a weak tutorial (including a “run in a straight line for 20+ seconds” segment), which City of Villains does better. World of Warcraft makes the less common choice of opening in its main game world, with no tutorial instance, but it manages to be dull for every race. Warhammer Online makes the same choice brilliantly by immediately tossing you into a warzone (best: Dwarf versus Greenskin newbie zones). The Champions Online tutorial just feels laboriously long. The Chronicles of Spellborn has a LotRO-style opening that ends well in a big fight with chthonic horror, but the gameplay along the way manages to be tedious even while very short. I remember starting A Tale in the Desert back before there was a tutorial, just drop you in the world and go; easily the most hardcore PvP game (with permadeath!) ever made. Wizard101′s tutorial explains things very well but is painfully slow and impossible to skip or hurry on a second character.
That hurdle overcome, the next question is where the most time is going to be spent in-game. Correct me if your game’s metrics suggest otherwise, but for most MMOs, it seems to be at the level cap. If nothing else, that is where your loud community is: the hardcore, the devoted, the guild community leaders who style themselves opinion-leaders or -makers (and may be). Any sane amount of content will occupy casual players, so giving people something to do at the “end” is how you keep and mollify the hardcore. How you do this well is widely disputed and the main topic of hundreds of blogs, so I will table (American sense) that issue.
World of Warcraft does this part famously well. Even if you do not like the WoW end-game, or the current end-game at any given moment, they have done a great job of recruiting and retaining players by putting an emphasis on late-game dungeons and raids. (Personally, I heard “the game begins at 80″ so many times that it was part of the reason I quit.) City of Heroes does this part famously sparsely, launching without the last ten levels and encouraging altoholism rather than building lots of level 50 content. Years into the title, there are exactly two raids and not a whole lot of level 50 task/strike forces (we try not to count the Shadow Shard content, out of politeness). Warhammer Online seemed to collapse (still does?) horribly at the level cap. Dark Age of Camelot had excellent realm-versus-realm combat but had horrible backlash when it added alternate advancement PvE content at the cap, creating a higher effective (and therefore required for PvP) cap. Back when I played, Asheron’s Call had a soft cap that amounted to an endless late-game. EVE Online has its PvP empire wars, to which Darkfall aspires. The Lord of the Rings Online™ has the ersatz version of the WoW end-game, taking the same approach but with very little content and alternate advancement grinds. It does, however, recycle its old end-games into new late-games better than WoW as the level cap rises.
Next up is the early game. If everyone is going to see that tutorial and new player experience, this is next, where you hope they all continue. It would be #2 on the list were it not for the amount of time your players can spend at the level cap between expansions. It remains very important, especially if it will consume most or all of the average player’s first month. A good start gave you a chance, but this is where you seal the deal and get the player to subscribe past the trial week or free month.
Age of Conan excels here, with near-universal acclaim for the Tortage experience. World of Warcraft varies between races/zones; playing on the Alliance side, I found I did not much like any early zones except for humans, although I recall a fondness for some early undead content. Dark Age of Camelot was good for its time but grindy and punitive in retrospect. City of Heroes/Villains does well except for a few painfully placed missions; maybe some of those are intentional, to make the travel powers that much sweeter. Warhammer Online is exquisite in tier 1, and if you have never played, you can go player tier 1 for free right now as much as you like. This is probably the worst time for EVE Online as players reach the “now what do I do” point. I have not tried the re-done LotRO low-level experience, but I always loved the Shire.
(Cynically, we also note that this is as far as most get in beta. There will be few to judge you on anything past this at release. This makes it a high priority while downgrading the importance of anything that will therefore have a smaller effect on your box sales.)
At this point, importance tapers off until you reach that end-game. Unless there is some modal point where most players end their second and third months, you focus on building the game out linearly. That early hook gives you some momentum through the mid-game. As long as the late-game is not so horrible that it is not worth getting through, players will get through those last few levels to see the glorious level-capped wonders they have heard to much about.
(Cynically, we note that promises to work on this area will carry you a long way. Wherever the population is centered at the end of the first month, just before subscription renewal time, announce you are going to fix that point and the range just beyond it. Repeat at month two. Warhammer Online did this brilliantly with developer letters just before renewal time in the early months. It helps if you can predict this point and really have improvements coming down the line, but developers are notoriously poor at predicting how quickly players consume content.)
You can see a great many games that have already embraced this approach. Part of it is just a natural consequence of sequential development. You worked really hard on the newbie zones in early beta, you worked on the glorious end-game wonders so you could show them off for the press, and then you fill in the middle as you get a chance, ideally trying to keep just ahead of the bulk of testers and/or players.
Some games really do fall down in the mid/late-game, hard enough to start seriously losing players. I love the 30s and 40s in City of Heroes/Villains, when all your powers and slots are finally coming together, but many people find it grindy without the quick progression from the early levels. Warhammer Online was appalling in the mid-tiers at launch, with poor PvE (“and such small portions!”) while the PvP balance problems were becoming apparent as all the powers and talents finally came together.
I’m stopping that thought so we can reflect. The mid to late levels are where you character finishes getting all of its abilities, with that “complete” point varying wildly across games and classes. If your game has horrible balance problems, they may be hidden under new shininess and quick growth, but they will become apparent in the mature levels. This is where the steam runs out for we the gameplay-Explorers. It is also where Achievers can jump ship as advancement slows down. This must disappoint the Killers: the sheep leave just as the wolves get the really fun ways to kill them all.
Zubon, it is sounding a lot like you’re saying that every part is important. And yes, I would love to say that, but experience suggests a few reasons why these later (but not end-game) levels are less important for retaining subscribers.
First, I am suggesting an extreme case of the game imploding. I do not know how many people ever experienced the Age of Conan end-game because the MMO blogosphere sounded like wailing from the fiery pits of Hell as people left Tortage. It is clearly possible to do far too little in that range, but many games get to “decent” at least.
Second, many of the extreme collapses are also end-game failures. They are balance problems or flaws in the fundamental systems that are to sustain players through the rest of the game, and there is no good news to reach after suffering through a near-empty, just-after-release late-game. These problems are not apparent in the early levels or not important enough to care about, while they first become visible in the mid or late levels. The Warhammer Online problems with city sieges compounded issues with the late-game open world RvR (plus a bit more), while the game had the same balance structure as most editions of D&D: just fine early on, when the numbers are small and luck can trump design oddities, but exploding into catastrophe as you multiply those oddities over many levels.
Third, “good enough” works. I would love to say that MMO players have discriminating tastes and high standards, but that is obviously false. We will put up with a lot of crap and flame anyone who suggests that quality and professionalism are below acceptable standards. One thing that Star Wars: The Old Republic has going for it is that at least some of their developers understand that the mass market will not put up with the crap we will, so selling to all those non-MMO addicts will involve improved accessibility and functionality (whether that idea has survived the EA merger is beyond me).
Kvetching aside, think of MMO players in two categories. For newbies, it is all new and exciting. Think back to your cherished memories of early struggles in your first MMO, and realize that you would never put up with EQ-at-launch today. Many of the problems in MMOs are not so bad once, just that we keep hitting the same bugs/grinds/AARGH for years. You will deal with it to see the new shiny when everything is new and shiny. New players are also more likely to play at a sane pace, perhaps try to experience everything on a first character (they don’t realize it is “first” not “my” character yet) before moving on to the next zone, thus giving more development time for that mid-game.
For veterans, we are obviously insane enough to put up with it, and we are already thinking long-run. Hardcore players are going to blast through come Hell or high water, and if the late-game content is weak, that is just more reason to push through to the promised land of Level Cap. You know common workarounds from previous games, you are tapped into the community to get tips on what is bugged and how to circumvent it, and you are already inured from years of suffering in previous MMOs. You have a community to help see you through, a guild of people to talk to, and you are not going to abandon your guild because (a) you like them or (b) you tell yourself you are playing to spend time with these people rather than get the next Ding! pellet.
So for all those reasons, I believe that if you sink the hooks in deeply, your players will probably view their first 40 or 80 hours as an investment rather than a sunk cost, and they will keep pushing on unless the game is truly painful with little promise of improvement. Or they are the much maligned, possibly mythical “tourists” who were never going to stay anyway, so again it does not matter.
Get the first day right: bait. Get the end-game right: long-term storage in the fish tank. Get the early game right: sink the hook. They may wriggle, but you will keep quite a few on the line with even a decent mid- to late-game. Or without the horrible fish metaphor: your early word of mouth gives the game life, and the long-run word of mouth sustains it.