Budgeting the Development and Launch

" MMP budgets almost have to be huge nowadays, but when looking at your budget, make sure that your budget is mostly content (art or world-building). It takes a massive amount of content to create a full-feeling world. If your budget is huge and it isn't mostly content, then your designers/programmers are probably trying to make something that is too complicated, too hard to Beta, and overall huge in risk."

” Damion Schubert , CEO of Ninjaneering, who formerly worked on Meridian 59 (M59) and UO2

Back in the ancient days of 1988 “1993, the production values of online games were, to put it charitably, "primitive," and it didn't matter one darn bit. Until around 1996, with the release of the last of the true "garage" efforts, M59 , online games were between three and five years behind the art and sound production values found in commercial games. M59 didn't bridge the gap, but it did close the gap a bit.

Similar to what happened with computer games in general, it is no longer possible for two guys in a garage to build and launch an online game. The playing public's art demands alone, which insist on the best possible art and lots of it, make it necessary to employ either a full art team of between 10 and 20 people or a couple of guys with seriously long art delivery timelines .

This was bound to happen once publishers got into the act. Since the release of UO in 1997, which featured a CD-ROM packed with art, publishers have been intentionally raising the bar on production values. Part of this was consumer demand for "bleeding-edge everything," and part of it was to make it impossible for a garage team to steal a portion of the market at a critical growth stage in the industry. Considering how bad the art and sound were in online games pre-1996, how loyal players tended to be to one game for years at a time, and how much money the games were making at the time, this was not an unreasonable strategy.

In six short years, budgets have ballooned from less than $400,000 to produce a cutting-edge PW to over $1,000,000 for a relatively easy-to-produce hybrid, and anywhere from $3 “$15 million to develop a PW. The costs are heavily on personnel; in the more expensive projects, baseline salary burn rates of $250,000 per month are not unheard of, rising to $300,000 “$350,000 during peak art loads. With a three- to four-year development timeline, that adds up.

Development Budget

So how do you budget for development? "MMP games are incredibly expensive to construct," says Damion Schubert, a designer on M59 and the former lead designer on UO2 . "The important thing to do is to be sure that you build your team at an appropriate speed and at an appropriate level. For example, don't hire world-builders until you have a tool for them to build with. Don't hire a lot of artists until your art pipeline is well laid out. Given that both of these factors take much longer in PWs than in standalone games, this can save you significantly in the long run."

Richard Garriott agrees that a major mistake some teams make is "hiring people too soon. A great deal of ground work can be done before you need the majority of the staff." It has been our experience that whatever baseline figure the team comes up with in the design treatment, it is probably short by a factor of two or three for a PW, or by around 40% for a hybrid. The first phase is to make the team finish the design and create a master task list in Microsoft Project or some other tracking software.

Your biggest time and money-saver in this process, then, will be completed design documents and project plans. It is impossible to get something approaching a realistic cost figure until that happens. If you're using the proper tracking software and your leads have estimated their development time and cost per task correctly, this should give you a pretty good idea of the overall development costs. An executive producer or senior management should then go over the project plan in detail with the team and not hesitate to ask for more details; the more complete these documents are, the smoother things will run. Clearly noting which tasks have to be completed before the art team is fleshed out is critical at this stage.

Launch Budget

It is becoming common wisdom among publishers that it takes about $5 million to launch a PW, not counting marketing costs. It can easily cost that much, especially when staffing costs for the full live team, hardware and bandwidth costs, and so forth for a full year are figured into the equation. Even a game that needs to support 30,000 simultaneous players at launch (as did the three major product launches of 2001) can cost close to $1.5 million for a minimal setup and the first three months of operations. If you're smart, you'll also ramp up your launch staffing over a four- to six-month period during the Beta tests, scaled to match the popularity of your game during that process.

There is no way to skimp on these costs without being penny-wise and pound -foolish. The best advertising any online game has is a technically smooth launch and a trained, helpful customer relations staff present on the first day. You can always scale down the team if subscribers don't match projections; it is impossible to scale up at the last minute and expect anything except a Type-H cluster bomb of a mess.

In general, the industry hasn't been very accurate at estimating the costs for either a game's development or launch. "After misestimating the completion costs/date, most budgets underestimate QA testing, launch costs, and CS issues," noted Gordon Walton. "It's more a question of what we are good at budgeting for this type of game and the answer is 'not much.'" As a lesson to be learned, this is an important one for the early stages of the process; it will be hard for you to go wrong by overestimating by 25 “50%, even after the detailed project task plan is completed.