It was announced yesterday that Runescape’s developer Jagex has cancelled its big scifi next project Mechscape. Without knowing any details about the project itself or the extent to which it’s been tested, it’s difficult to comment, but it’s a convenient time to share a few thoughts about when calling it quits is a good decision, and when it’s not.

The first element to take into account is that an online game has a lot of ongoing running costs. Launch is just the very beginning, and online games is not really a field where you can make a quick buck and run away by releasing something which can never become great (as happens quite frequently in the offline games industry). Sunk costs are by definition already lost, and the damage you might inflict to your brand by launching a lemon may not be worth the limited money you can gain by selling boxes or lifetime subscriptions upfront. If you’re in it for the long haul, as most online games developers are, cancelling might be the best option.

On the other hand, if a project that’s been in development for a long time is in such a bad state that the best course is to can it, that’s probably an indication that something within the development process is not working right. Perhaps the troubled game been developed “in a bubble”, with features and costs piling on without any input from players. When this happens, it’s not uncommon to discover shortly before release that the current state of the game isn’t satisfying, and often the development process is too rigid for appropriate responses to be taken as a result of player feedback and metrics.

Unless you have artificial constraints (such as a box release) that put a lot of pressure on the launch, the game doesn’t have to be ‘perfect’ as long as 1) it’s basically a good experience for players, and 2) you have a framework in place to allow it to evolve and thrive. Releasing a good quality minimum service should be the focus, while keeping the processes agile enough to grow a community. The best approach is to listen, learn and react quickly to the feedback players give you. Lowering the pre-launch investment also makes it less risky to experiment, which is another key support for learning. There’s a curious dichotomy here: if start-ups are getting extremely good at learning (primarily because they have no other choice) and their games grow successful as a result, it’s a bit puzzling to see bigger companies knocking themselves out to “get things right” at launch, and shying away from the very attitude that may have helped them succeed in the first place.

The Jagex article states that the previous game (I guess talking about Runescape) “wasn’t a game the team wanted to play”. Which is actually fine, unless the development team is your target market! Delaying a game’s exposure to players is like doing improv comedy when no-one’s watching, or dungeon-mastering an RPG using only NPCs: if you don’t give people the chance to experience what you’re doing, you can’t know if they like it or not, and in the case of online games, not knowing can be very costly. An online game is a dynamic, almost living thing; there’s no such thing as ‘finished’ until you pull the plug, and a successful online game is really a team effort between the developers and the game’s community.

The past week in Korea has been very busy for us. We attended the KGC event for the first time this year, and it has been a very nice experience.

Thomas spoke at KGC, presenting not one lecture, but two! Both went very well, with a very attentive audience that was thirsty for knowledge on the European market and the European players.

The presentations are on SlideShare now:.

This presentation is similar to others we’ve done during the year. Some of the numbers have been updated, and as always we have provided additional information from previous iterations. The section on how to enter the European market, and related advice, has been fleshed out more thoroughly.

This is quite a new angle for us. We prepared this lecture at the request of the KGC organizers, and we can honestly say it was a lot of fun to prepare. We’d like to develop it even further, but we hope that in its current form it has enough material to be useful to its intended audience (non-European developers).

As ever, feedback on both presentations is very welcome!

[You may have noticed that those presentations feature a different logo than our previous materials. We'll talk more about that soon, but for now let's just say that we felt the need to move on to an image that's more representative of ICO's spirit.]