To split or Not to split

This is not a very much discussed topic in the world of Online gaming where game design, technology or business models make for sexier discussions. Maybe there is a consensus and everything below is so obvious that it isn’t worth time spent on it.

In our experience though, it is an aspect that is just not considered, whether for economical reasons (the studio will licence the game anyway and need to separate the different services) or for a lack of interest (operators just do it the way their main competitors do it). When everyone is pushing for the “It’s a service, not a product” motto, it is a very important aspect to be considered.

But first things first, what do we mean by split services?

In a world of global economy, of Internet connectivity, there are a few different ways to structure your service. You can make one global service offer for all the players, regardless where they come from. They all play the same game, pay for the same things, have access to the same server(s). This is the service offering of EVE Online, Runescape and Guild Wars (even though technically Guild Wars at launch was offering Separate services and EVE has a separate service for China).

Players don’t necessarily pay the same price, they essentially are serviced through one big common offer and can play all together and in the same way.

But, you can also offer separate Services. For the purpose of the discussion, we will exclude consideration on Asia and focus on the separate services offered for North America and Europe. The players may play the same game (but not necessarily, you can introduce service specific rule sets for instance), they don’t have access to the same service offers (it can be different pricing, different business models, different server sets, all of the above) and they are managed separately to a certain extent. Good examples of this are World of Warcraft and Habbo.

As mentioned before, this might be considered a non-issue, but it actually has a lot of long-term consequences. For a starter, you cannot that easily switch from one to another. Actually moving from a Split model to a Global model is easier than the other way round and probably one of the first reason we tend to prefer the Split Services approach. Ok, let’s dig deeper in some Pros for Split Services.

Why would you want to have separate services?

Licensing - Briefly mentioned in the beginning, if you want to licence your game to another territory, you will have to maintain separate services. No local operator will accept a global model and risk their business by sharing its service offer with the other licensees (including if the studio developing the game is the only other operator for the game).

Business Model - Different territories, even ones culturally close like Europe and North America, may benefit from different business models. The current economy situation alone and the drastic change in the different currencies exchange rates is an example of this. With one global offering, you need to be vigilant on the different pricing for your service and the corresponding value for money. Of course with separate Services you also need to be careful on this but because they are different services, you can actually manage this a lot more efficiently. If you run a game that is based on virtual items sales, it is quite key to manage the item shop with care and separate services make it that much easier to optimize the prices for your audiences and maximise revenue. If you have a box distribution business (or if you integrate some pre-paid card scheme), it is an efficient way to protect yourself against grey imports.

Marketing - Your marketing team will always want to create something specific for a partner, have some sort of exclusive with a distributor (digital or physical). Separating the services will offer a lot more margin of manoeuvre for those operations and any exclusive will be on a given service, making those assets re-usable for other deals on the other services. Ideally, if you have the ressources, you want to develop something unique for each service as well. In the same spirit, you can tailor special events (marketing or in-game) to local celebrations. Europeans don’t care that much about the 4th of July and Oktoberfest isn’t celebrated across that many countries.

Community - This one is tricky.  Most of the time split Services are good for your communities. Player retention is always better when you force players to play with other people from their region (and I am not talking about players that don’t speak the same languages): it creates more tightly-knitted communities. Like in the game design, you don’t want to give the player the capacity to make a choice that will make his experience worse, when defining the service, you want to take away a choice that could ruin the player experience. With a global offer, you might have players choosing to start to play within a foreign community “for fun”… That cannot provide them with the best first experience with the game.
Quite a few times, we have had that discussion and an argument that would come up surprisingly often: “but the players don’t want to be segregated, they want to all play together!”. That’s a very strange argument to make. Players also want the xp to be faster, the challenge to be easier, the subscription price to be lower… At the end of the day, players who will really play outside of their territory form a very small minority. It is also a player base you won’t lose if you don’t offer a global service. They might disappointed but they won’t leave. While you may lose players who can’t find players to play with because they play in a different timezone from the one they live in.
Sulka Haro, in one his lecture, also described the very difficult situation Habbo Hotel ended up with when they launched their Japanese region. Their existing players swamped the area, initially eager to meet with Japanese players to the point the later felt overwhelmed and decided to stay locked in their rooms.

Legal - Different countries have different laws. As a global service, you have to comply with all of them… As a separate service, you can tailor your terms to make them work locally. It is not something to consider lightly. It is already a big headache to be able to create a service offering that is compliant to European countries, adding the US laws in the mix makes it quite difficult to manage… And usually to the disadvantage of the US service offer. As with everything legal, it is not a problem… until someone decide it is and then you wish it wasn’t a problem. It gives you a better control on the legal definition of the service offering and you don’t want a loose definition there.

Tailored service to local standard - A good example are the means of payment that varies greatly from country to country. Having separate services allows you to include only the relevant solutions for specific users, as well as tailor the pricing and the offering. Another example would be Customer Support standards: a lot of US customers are used to have access to a call centre, a solution which is not very practical in Europe, notably because of the cost of a multi-lingual call centre and the related EU regulations – it is very uncommon for any European game service to offer support via a call centre. Separate services allow to deeply customize all aspects of the service to best fit with those standards.

So why would you want to have a global service?

We believe there is actually some excellent reasons to still consider a global service:

Critical Mass - If your game is quite niche, and you don’t expect to be able to reach the critical mass to make it fun to the players because it would feel empty, going global is an option, as it allows you to aggregate all the niche audiences interested in your game. Besides, since it is niche, players can be more enthusiastic and willing to go out of their comfort zone to play with foreigners, use English to communicate, etc. “Mass-market” consumers expect to play in their own language with fellow nationals anyway. You can consider at some point, when you have healthy population on different territories, to split into separate services but it is a tricky exercise to go through.
Especially for browser based games that lack the ability to create a launch-momentum with boxes appearing on shelves, it is a tricky position to be in. You have to go through the cycle of being small and growing every single time you launch a separate service, but it might be necessary anyway the moment you start servicing your game in a different languages. This isn’t true for all games though, it really depends on the communication  channels you give to the users and how it influences the experience of the user. Online collectible card game Urban Rivals has been pretty successful in managing different languages in the ir communities through one international service offering.

Scalability - Similar to critical mass, when the game is struggling in one territory it won’t be as obivous if you have a global service. Additionally, you don’t have to take the hard decision to close a service or merge it together with another one to remedy critical mass issues if your player base dwindles.

Game Design (it is game design in the sense it will influence the game experience, it could also be called Service Design) – One of the best example would be EVE Online. And this is an excellent reason to promote a global service approach: the mere fact that there is only one service is core to the user experience. There is one global game economy, one storyline based on player actions and one “truth” the whole player base can refer to. It also comes with a price: in the case of EVE, a good part from the steep learning curve comes from the sheer size of information the global service creates. It is more difficult to find people speaking your language, to understand the political situation, to understand the market situation. The very things that makes it so popular to its existing user base, makes it that much more difficult for a new user. Not to mention the drama, that EVE had to deal with on many occasions. Any scandal is felt personally by the entire player base. This being said, EVE wouldn’t be EVE on split services and it wouldn’t have reached its current size. A Global service service doesn’t necessarily mean a more hardcore experience, but it certainly can contribute to that effect if not closely watched.
In the same spirit, games like Trackmania, Urban Rivals or Power Challenge, and their international competitions, are benefiting from their global offering.

How far should one go?

Now, how much should separate the service. We mentioned the North America / Europe split, but the arguments made could encourage for even further separations: France/UK/Germany ? West Coast/East Coast? And essentially, it is a matter of common sense and making it manageable. Critical mass can help dictate the best way to go as well. But, theoretically, there is no limit to the level of separation you can have for the services. It has been hugely successful for Habbo that went to the country level.

Like in all things, this is in no way absolute. There are solutions and work-arounds to all the problems the global service model is bringing. The most important point we wanted to make is that it should be carefully considered, all the options weighed and be made sure the decision taken is the best for this specific game.

Games using the concept of shards will find it more easy to actually follow a split services model, as for games with a younger audience. In the same idea, the organisation operating the game is important to factor in, as a purely American company will have a harder time to manage a global service compared to a more widespread or even just multi-cultural one.

Games having mechanisms that benefit from a rich and vast economy, games with a competitive dimension coming before the community dimension (Urban Rivals would fall into that category as would Power Challenge for instance), games with a system that allow within the world for national repartition (whether it is geographical or via instancing), they will probably find more advantages into a global service model. In any cases, there is also some room for some hybrid solutions: split services that are sharing multiple features or a global service with “invisible walls”. It also makes sense if you offer split services to have the flexibility to allow player to be able to move from one to another. In all cases, it is some important considerations to take into account while developing a game and to evaluate regularly even after the initial launch.

Tags: Eve Online, Global Service, Habbo, Split Services

blog comments powered by Disqus