It Always Starts Small
I think a lot of people start off interested in making games the same way I did:
“That doesn’t look hard to make. I bet I can do better.”
If that’s accurate then it’s hardly surprising that the two causes of developer failure, underestimation and hubris, pop up so often – they’re the impetuous for starting in the first place.
Good and bad, right? Many people fail without starting in the first place, but that misplaced motivation can come back to bite you if you’re not careful.
The first blog post was about how I failed in the development of Ignition Impulse. This second blog post is about how I failed in the development of Steam Marines.
Play It, Sam
It started as a small game called Quad. Conceived as a small, turn-based, squad-based RPG, I started posting about it in my forums in early 2012. Feedback was minimal, but there was some interest there so I ran with it.
Aside: You can already see one big problem – “Quad” is a terrible name for a game. Just try Googling it. It took me about six months to rectify that!
I was doing a lot of other things behind the scenes; “real” development didn’t kick in until around April 2012. In-between I tried prototyping a bunch of other games, did more research on the games market, and other assorted failure prevention type stuff.
The core of Steam Marines is having a four marine squad moving through a spaceship and fighting off robots and aliens. Pretty straightforward! It was turn-based, used an action point system, and made some probably questionable design decisions like locking movement and firing to cardinal directions only.
The hooks for the game, the bits that set it apart from my point of view, were that it was brutally hard and straddled a middle field between shorter, more casual games and longer grindfests. During development the “Steam Marines is hard” angle overshadowed everything else.
But It’s So Bouncy
There was not one failure here, but a series of failures of communication between myself and potential players. Steam Marines looked bouncy, toony, and colorful. But it was, at least from the average player’s perspective, overwhelmingly difficult. What makes this worse is that I knew this – in fact, I deliberately balanced the game on “Hard” difficulty since I knew I had a personal preference for difficult games.
In other words, I was catering to a niche within a niche and I was not explaining this well.
Many early players and Let’s Players got it immediately, primarily fans of traditionally harder/permadeath games. Everyone else sent me hate mail.
Aside: It is a very weird feeling to have made something that people literally hate you for making. It’s not all bad. Sometimes you get an email reading somewhat like, “I hate this game. I hate you. I’m starting a new game.”
If I had paid more attention to the mobile games sphere I probably would have gotten this right away. People simply associate bright colors and bouncy animations with easy/casual. But I didn’t and the resulting backlash from players was both confusing and damaging to morale.
I like to think I have thick skin but the truth is after a few dozen people tell you your game is crap and you are crap (and for the same reasons!) you start questioning every decision you ever made. Riding low on the results of Ignition Impulse did not help any.
In the end it was a combination of stubbornness and people giving me actually constructive feedback that got me back on track. I really want to stress this: you need to be both stubborn and open-minded to do this (well.) You need to be open-minded otherwise you will miss good feedback and opportunities, but you need to be stubborn enough to reject the crap feedback. And you will get a lot of crap feedback.
One might argue that the pertinent trait here is the ability to accurately assess a given situation, but I think the problem is deeper than that. Frequently the issue is not one you can parse out objectively. Issues of preference, balance which is contingent upon the individual, et cetera abound during game development. At the end of the day you’re going to make some bad calls no matter how sharp you are. Practically speaking your goals should be to minimize the frequency of those and mitigate the really bad mistakes you inevitably make. “Always make the right decision” is about as useful as “never make spelling errors.”
Art was coming into the project and I was feeling good. The game was starting to look like a real game, I had some really cool theme music, and I was on Reddit and Twitter talking with other developers and promoting my alpha.
In August 2012 Steam Greenlight was launched, ostensibly to help indie game developers get onto Steam in a more streamlined and orderly fashion. I can’t be overly critical of Greenlight. Mistakes were made and the process was far from perfect, but Valve made a large effort to do something that they did not have to do that did end up helping many indie developers.
Steam Marines landed on Greenlight the opening day… and took about 400 days to get Greenlit. A note on Greenlight to any developer currently in that pipeline: you need to push hard to get eyeballs and yes votes. Yes, there was an initial surge of eyeballs when Greenlight was new and shiny. It still took me and my tiny community over a year of bush kicking to wrangle up the required votes to make it past the finish line. I’m sure developers who have run successful crowdfunding campaigns agree.
Aside: Even if your game is so awesome it can make it without external effort on your part, I can’t recommend doing that. Push it, promote it, don’t stop talking until you make it. Out of all the walls you will face during development Greenlight might be one of the longer, more torturous ones, but it isn’t the highest.
Don’t think I did everything right in this case. I missed the opening hours of the Greenlight launch because my artist at the time assured me he would have the necessary assets (logo and so forth) ready on time. He did not.
I did get lucky in some ways. Negative comments on the Greenlight page were few and far between. Some cries of “Clone!” but they were mostly shot down by other commenters. It was a good trial by fire for learning to deal with people who had no prior investment in my game, and how to answer soft but firm.
I did reach top 100 relatively quickly, although there were some very dispiriting months where Steam would Greenlight a few titles… and Steam Marines would actually drop a few places in rank. I don’t have good advice for that kind of demoralization.
Sometimes there are no clever tricks; you just have to roll up your sleeves, get out, and push.
Abracadabra (Or, “Problems Go Away If I Ignore Them, Right?”)
This post is about failures with Steam Marines. It would not be complete without mentioning management of contractors. By and large I believe I lucked out. I ended up with mostly great people to work with.
My intention is not to point fingers at anyone. The following is a non-exhaustive list of problems that can crop up with any contractor, employee, et cetera.
- Have a contract.
- Have an exit clause. Think hard on what the terms should be. Do not ever think, “Oh, well this will never be used.” If it will never be used then why is it in the #[email protected]%ing contract?
- Be very clear on what the expectations are in the working relationship.
- Be prepared to sever the working relationship if necessary.
Being saddled with someone who does no work is not the worst thing. The worst thing is being saddled with someone who is actively damaging your project. Please be aware of this.
Aside: There are a myriad of ways someone can damage your project. They can produce subpar work. They can produce no work while saying they will. They can undermine your processes. That someone can also be you. Try not to be that person for yourself or anyone else.
I’m a fairly caustic individual. I’m blunt, willing to be rude to make a point, and I found it difficult to pull the rug out even though all the red flags were there. I don’t really have an excuse but I think my only explanation was that I felt I didn’t have many good options. Given that the project survived and thrived despite those issues, that was clearly a misunderstanding of the larger situation.
I can see it with a bit more clarity now, so let me offer you this advice as explicitly as possible in the event you find yourself in a similar mind-and-decision-blurring predicament:
You always have better options.
I’m not guaranteeing you’ll see them, but they exist.
I’m skipping ahead a bit here. Steam Marines was Greenlit, it was voted for Community’s Choice, and made a bunch of sales and money on Steam, Humble, Desura, and IndieGameStand. Into profit territory, woohoo!
There were a lot of little details and failures all over the place. I don’t have too much to say on any of them individually, but I want to impress upon you the volume of problems, or potential problem situations, that can crop up:
- UI/UX. Even if everything is rebindable people will complain about your defaults. Some people will love your buttons. Other people will hate your buttons. Some will even say your buttons are uninspired.
- Most players will never look at your menu options.
- Enemy AI that announces its intentions is widely considered more intelligent despite the fact that it is an incredible act of stupidity for any party to announce their actual intentions to the opposition.
- 99% of people don’t read anything in-game; 1% read everything.*
- You can make an announcement, sticky a thread, blare it in-game, and people will still miss it.*
- You can make the screen shake, throw up character dialogue, and center the camera on an event and people will still miss it.
- You need to be very, very, very specific about what you are promising and what you are not promising. People will still crucify you but at least you have something hard to point at after the fact.
- I used words like “Hard” and “Normal” to describe the difficulty levels when creating a new game. Also on the ship decks which had an entirely different meaning.
- People really hated ‘Z’ as the default keybinding for firing a marine’s weapon. Okay, that one was one me – I could have picked a better default.
- Gamepad bindings are driver specific. There are different drivers on different operating systems – you get the idea. I wonder why so many consoles don’t allow rebinding?
- Linux players are few. They cannot buy in numbers of Windows players… but they seem very tech savvy and figure stuff out without contacting tech support (that’s me.) Linux users will also cheerfully report esoteric bugs on their esoteric rigs with a cheerful, “I know this is esoteric – don’t worry!” (If you are good at ignoring problems that aren’t really problems to your users this one is probably fine.)
- Mac users are not as tech savvy. Less than 5% of the player base but 50% of the tech support.
- People will tell you they have up-to-date drivers. Then you see their dxdiag dump files.
- People will tell you they did XYZ to resolve some issue. They did not do XYZ.
- What are system requirements?*
- Get called an X because a customer did not realize Y (my favorite was someone who called me a scam artist because he did not realize Steam Marines was a 2D game.)
(*This is why marketing is hard. I am a strong advocate of slow burn, early and often. There are valid arguments for timing release of information and such, but my position as a small developer is that marketing is an enormous wall I need a head start on.)
There’s a minefield of things that can go wrong. Even if you are kind, cheerful, and helpful people will spit in your face. That’s just going to be something you (or someone you hire) will have to deal with.
My advice: be polite and be as helpful as reasonably possible.
To be honest I don’t follow that advice 100%. I may or may not be known for being somewhat snarky on the Steam discussion forums for Steam Marines. I think the right combination can be snarky but helpful and/or comic.
If you can’t swing comedy then just be polite.
Aside: Sometimes I respond to Steam reviews. Sometimes the negative ones. Some people do not take any kind of developer comment well, apparently. Sometimes having that PR person, that layer of insulation, is a good thing. Assess your own personality profile when it comes to this, too.
So many ways you can fail to so many people, so little time.
Are We There Yet?
Probably not. To recap I failed pretty hard at game naming, conveying what Steam Marines was actually like to play, didn’t manage contractors as well as I could have, and probably made a bunch of miscellaneous bad calls during the design and implementation of the game.
I got some decent coverage from some sites and YouTubers, and they definitely helped me through Greenlight, but if they were timed better they would have been spent actually getting sales as opposed to votes.
I started selling at too low of a price. Every subsequent price increase actually increased my unit sales. Oops. There are some confounding variables here – obviously the quality of the game was increasing alongside the price – but I’m fairly confident I priced too low early on.
I have merchandise (shirts) but never pushed it hard. I still assume my player base is on the small side to make that a profit center.
I didn’t go to any major game development events like GDC. I didn’t enter into any game contests like IGF. To be frank I consider these mild failures at most. While I am biased as an introvert I don’t consider physical hobnobbing to be that much of a leg up, and contests… well I don’t put much stock in those, either. I consider time spent on social media, and perhaps some physical location promoting at local gamer meetups, to be more productive. But maybe I’ll change my mind in the future. Steam Marines is, after all, a niche game. A game targeted more broadly would probably benefit more from these larger events and platforms.
That’s the end of Part 2. I don’t know if there will be a Part 3. Oh, wait, one more thing. Let’s end where we started:
“That doesn’t look hard to make. I bet I can do better.”
Maybe your game will encourage the next generation of game developers. Your game doesn’t look that hard to make after all. They can probably do better.
Thanks for reading,