Eldritch, an impressive roguelike indie game I wrote about here on New World Notes last Fall, was developed by ex-2K Marin developer David Pittman, who left the studio after working on BioShock 2 and The Bureau: XCOM Declassified. David just published a post-mortem of the game, which has some interesting sales figures for the voxel-based Lovecraftian action adventure, and a slew of valuable tidbits for indie developers and enthusiasts alike.
I reached out to Pittman with a few questions about his post-mortem analysis, and his answers are a must-read for anyone wanting a glimpse behind the curtain of indie game development. After the jump, Pittman talks Steam success, PR failure, and developing in the shadow of a voxel giant.
The core principles of the game -- roguelike structure, stealth/combat mix, lo-fi art direction -- were planned from the start, but the details emerged over the course of development. For example, I knew I wanted the player to have magic powers, but I never had a master plan for what those powers would be. I added them as ideas occurred to me, while I had time.What I did plan and rigorously stick to was a schedule of major milestones: When the game needed to be playable by friends, when it needed to be content complete, when it needed to be bug free, etc. The way I worked was loosely inspired by Scrum. Every week or two, I would set a new high level goal toward the next milestone, but not define exactly how to solve it until I was actively working on the problem.
The press coverage that Eldritch received was diffused over a pretty long timeframe, so avoiding the huge releases possibly didn't make a significant difference. Although the initial numbers were modestly successful, I feel that awareness of Eldritch was still relatively low at launch.
I'm sure some players see the pixelly, voxelly visuals and immediately assume that if it looks like Minecraft, it must also play like Minecraft. I expected as much, and tried to highlight other aspects of the game. For example, on eldritchgame.com, the reader first sees a non-blocky illustration and the keywords "roguelike" and "immersive sim"; anything resembling Minecraft requires scrolling down the page. But a lot of press featured the game with screenshots instead of the key art, so a lot of first impressions were shaped with the visuals in mind. I hope that in time players largely come to recognize voxels as just another tool in the developer's kit and not something uniquely associated with sandbox games; but in the meantime, I'll probably shy away from using voxels in future games because of the confusion.
Return on investment and risk (of failure, feature creep, etc.) are the key factors for me. I was able to create [New Game Plus] in about two days without any significant risk, so it was an easy decision. Other suggestions are more difficult to act upon. A lot of players (including myself, frankly!) would like to see online co-op added, but that would require rewriting virtually the entire game and make QA testing an order of magnitude more complex. In between those extremes were a lot of things I could have done, until the financial need to move onto the next project became significant. I chose to develop [the Mountains of Madness DLC] because there was a demand for more content and the risk involved was quite low. The return on investment could have been better if I had sold the DLC instead of offering it for free, but my finances were sufficient that I was willing to trade that short term reward for what I hope will be long term player investment in myself and our company.
The need to be on Steam to be successful didn't impact development of Eldritch, in part because I didn't actually realize how important the exposure on Steam was until after it launched. I expected Eldritch to take longer to be greenlit, and had planned to launch only on [the Humble Store], with the Steam version coming out later.
For our future games (and for other independent developers), I think the impact will be felt in negotiating the timing and message of the Greenlight campaign. If we start too soon and have too little to show, we may get a lot of downvotes and never really recover. If we start too late, we may not be greenlit in a timely fashion, which means we couldn't announce a release date with any certainty (or we could announce a date, but might miss the critical opportunity of being on Steam on that date). Or if we announce a game with niche appeal, it may take a very long time to find its audience on Greenlight.
The timing of the launch day press release (right when people were taking off for holidays) was a big and easily avoidable mistake. My other critical PR fumble was choosing to only contact press on launch day, not when I initially announced the DLC (which I did on Steam, Facebook, and Twitter). I was trying to err on the side of not spamming people with a lot of press release emails about the game, but the result was that the story got picked up without any real information, because I hadn't provided any.
In fact, the broad lesson I learned about marketing and PR from the entire Eldritch experience was that I need to communicate more often with more people.The other thing I might have done differently is to make Mountains of Madness paid DLC, but offered for free to everyone who already owned Eldritch. The gesture of goodwill would remain, but I would generate at least a little bit of revenue for the time I spent on it.
When I wrote that, I was thinking that players were probably getting tired of sandbox games, or that if they wanted that sort of experience, they would just keep playing Minecraft. Then the Starbound team announced that they sold one million units in a month, so it's possible I don't have any idea what players want!It's given me a lot to think about for the future, in any case. I feel like the overwhelming success of Minecraft indicates a few things that the industry was not paying enough attention to. Deep or complex yet family-friendly games are a huge and mostly untapped market. Players can accept lo-fi graphics, for at least certain genres of games. There's a whole lot of people who like building things more than destroying them.
Iris Ophelia (@bleatingheart, Janine Hawkins IRL) has been featured in the New York Times, and has spoken about SL-based design at the Fashion Institute of Technology in Manhattan and with pop culture/fashion maven Johanna Blakley.