Venture Africa Post Mortem - Part 2

Part 2 of the Wildlife Tycoon: Venture Africa Post mortem focuses on business and management issues. Again, this was originally intended as an internal document, but it also might be of help to some people out there. I hope someone can glean something from this...

If you missed Part 1 (which focused on Technical, Art, and Design issues), read it here.

Revenue
Wildlife Tycoon has been a modest success for Pocketwatch Games. We met the major goal for the product, which was to establish Pocketwatch Games as a serious contender within the downloadable games space from a quality and artistic perspective. Early sales predictions were from 1,000 to 50,000 units over the life of the product. (The remainder of this section redacted)

Cost
Contracting fees were relatively minor, comprising under 1,000 dollars in costs. The biggest costs have been rent, and costs related to game development conferences. In all, about 9,000 dollars has been spent on the company since January 2005.

Management
The biggest success in terms of management of the product was the asset list. A very well-thought out GDD led to a constrained and detailed asset list, probably only 25 percent of which changed by the end of the project. The GDD and the asset list were the key parts to why the game got done on time and on budget.

The asset list also included micro-royalty percentages that were tied to the completion of each asset. In addition, 9 percent of the royalties were reserved as contractor bonuses, to be doled out after the project to those that put in more work than was expected for a particular asset. Percentages for each task were based on the predicted percentage of man-hours spent on that task (as compared with the total man hours expected for the project) and on the difficulty of finding a contractor that could complete the task. Negotiation with contractors was virtually unnecessary due to the somewhat generous royalty values and the copious data given to them about the royalty scheme and the plan for the project.

In addition, the asset list also kept our design constrained; we had to work within our limits when we wanted to expand or change the game design. Since the clean and well thought out asset list was central to the success of so many elements of the business, we must be sure that future projects are not so much larger in scope that we cannot effectively plan them from day 1.

In terms of personal team management, we got lucky in that everyone involved was professional and motivated. Only one person dropped off the project in the middle. Almost all of our communication was done through instant messenger, email, and a yahoo group. Having all our contractors be remote is only now beginning to take it's toll upon the company psyche (re: my psyche). We also got lucky that no team members tried to turn in low quality work in order to satisfy the royalty requirements. For the next project, we should slowly start to build a local team, but still utilize our remote contractors so that we can grow intelligently.

The biggest failure in terms of team management was our lack of an art direction. Environments, animal models, UI screens, and music each had a different tonal feel. One of our first needs, then, is a central person who can direct the look and feel of the game, without sacrificing the personal styles of the artists.

In a few cases, we were too lenient with delivery schedules on assets. Some assets ended up in the game that were clearly sub par because we ran out of time for the artists to work on them. I don't have a good solution for improving asset delivery schedules with part time contractors, though perhaps we should expect to do a few post-release asset improvements.

Finally, our asset file management and code revision control was terrible. Everything was done in an ad hoc manner. At one point, a hard drive failure made us lose a month of programming work. True remote source control need to be implemented to avoid major data loss and organize the delivery of assets.

Business/Distribution
Probably due to the conceptual game design, press obtained through GarageGames, and the pedigree of Pocketwatch employees, we were the subject of plenty of publisher/distributor attention. An early deal with Trymedia allowed us to set up a time-limited private beta test. Active interest with a number of publishers allowed us to get a good deal with MumboJumbo for the worldwide retail publishing rights. This attention was received due to occasional blog posts on the GarageGames website. We should continue to keep development blogs on GarageGames.com as well as to an external blog site so that we can control advertising and traffic.

We have learned quite a bit about which distribution avenues are successful for this type of game. Our greatest amount of revenue came from our own website. After than, the Trymedia portal (TryGames), download.com, and the PCGamer pack-in CD were excellent sources of revenue. Revenue numbers from GarageGames direct or affiliate sales are not yet available. Our lowest sales revenue came from European print magazine pack-in CDs (we were on about 5), and hard-core gamer download sites like GameSpot. It is still wise to have as wide a distribution network as possible, and we will strive to continue to broaden our distribution with the next game, possibly through Plimus.

However, we currently have no way to identify where a customer came if they download the game and buy it from our website. We should implement a cookie on the website that tells the game trial where the customer arrived at the website from, so we can accurately divine our best sources of advertising.

The "soft-launch" of the game was a problem. After the beta, the game launched from our website, and we expected a very small customer base to give us some feedback through forum posts and our early conversion rate. Unfortunately, a few large portals picked up the release and pushed the press engine into motion, against my will. Our initial buggy soft-launch inadvertently turned into our wide release. In the future, we should use the beta period to test and gauge reaction to the game, and only launch the game for sale when it is truly ready for the public.

PR/Marketing
Our best advertising in terms of customer conversion probably came in the form of the IGF nomination. A surprising amount of revenue is coming from direct sales on our website; most of the traffic is being driven there from the IGF website. Similarly, GameTunnel's June mid-year roundup of the best games so far and the games people were looking forward to kick-started all of the publisher/distributor interest in the game. Both of these events were driven by "tertiary" advertising sources: the IGF nomination from my articles in Game Developer Magazine, and the GameTunnel article from my blog posts on the GarageGames website. "Primary" advertising (direct to customers) is generally too expensive, but word of mouth and viral marketing depend on these "tertiary" advertising sources.

We also failed to create much "seconday" advertising, meaning reviews of the game, possibly due to fears of having bad reviews. One good review, however, is worth 3 bad ones. In the future, we should have a review schedule lined up before the release of the game, and we should attempt to increase and further control our exposure through "tertiary" advertising, like blogs, articles, and endorsements elsewhere in the media.

Our primary advertising has been miserable. We don't have very good data on it's effectiveness, but my experiments with Google AdWords had poor results. Not only did the funding for the ads have to come out of the pockets of Pocketwatch Games (after the contractors and distributors were paid), but they didn't even manage to pay for themselves. In the future, we should structure contractor agreements such that their percentage is calculated after advertising costs are accounted for, and we should only attempt to use Google AdWords if our conversion rate is over 3 percent for advertisement-driven customers.

Our targeting was also poor for both our secondary and primary advertisements. I wasn't able to reach the coveted casual audience because the general gaming press drive traffic of more hard-core gamers. In the future, we need to find press opportunities that will drive the traffic of non-gamers.

Our website and forums have been a modest success. The forums are slowly growing in membership, and they should allow us to do some direct email advertising for our next title. The website should continue to grow in content and features, and I should implement an RSS feed for the news page of the site so that others can keep up to date more easily and so that I can syndicate that news feed on other sites under the Pocketwatch umbrella.