The Binpress Podcast

The Binpress Podcast


Binpress Podcast Episode 24: Brad Wardell of Stardock

December 16, 2014



This week we talk with Brad Wardell, founder and CEO of Stardock, the software and game development company behind everything from Object Desktop to Sins of a Solar Empire and Galactic Civilizations. Brad discusses the importance of delayed gratification when building your business, how to distinguish between a hobby and a business, and the importance of location when recruiting talent. He also covers how he got his start, why cognitive dissonance kills more businesses than anything else, how to make sure partnerships are worth your while, and much more.


Listen to the podcast in the player above, or click here to download it directly. Subscribe on iTunes or do so manually by using this RSS feed.


Subscribe oniTunesPocketCastsStitcher


Show notes

Transcript

Alexis: Brad, thank you for coming on the podcast.


Brad: Oh, glad to be here!


Alexis: So, as I’ve already introduced you in the intro as the head honcho over at Stardock, before we dive into Stardock let’s go back into your past and find out how you got into this whole programming mess.


Brad: Oh yeah, well, it’s funny because I started Stardock in college – this was over 20 years ago – and the idea was to make a little bit on the side to pay for school. I was going to Western Michigan University to get a degree in Electrical Engineering and at the time what I would do is I’d build PCs put OS/2 on them.


I was a big fan of Sid Meier’s Civilization, which had come out, but I wanted to make a space one. So I literally picked up Teach Yourself C in 21 Days and learned how to program and made Galactic Civilizations for OS/2 – this was back when I was still in the dorms.


Alexis: Did this take longer than 21 days? I have a suspicion it did.


Brad: Oh the game did, yes. Learning the basics of C enough took about three weeks to do. We did it pretty straightforward. Back then, especially the graphics requirements for sprites so you could draw your own graphics, put them on the screen, move them around – very easy compared to today.


From there, the game did well, but we didn’t get paid. Our publisher didn’t pay us anything, but the game made Stardock well-known in the OS/2 world and from there I leveraged the awareness to start publishing third-party games and I made my own sequels.


Alexis: What were some of the hurdles that you had to overcome in order to get your game in front of people? Because nowadays, it seems like it’s a lot easier than it was then in terms of distribution.


Brad: It was quite a bit different. One, we used to charge a lot more for games. People think of games today as expensive, but my little OS/2 game was two floppies and I charged $60 for it.


Alexis: Man!


Brad: I know. That was in 1993 dollars, and I was literally making the copies myself or asking my – at the time my fiancé, now my wife, to be making floppies. I was big in the Usenet, which is in the modern day would be Reddit or Twitter, but I used that to communicate with OS/2 users. For that, I was able to – we were able to get at retail, which was big. IBM was really pushing OS/2 at retail and so they really helped pave the way to get our stuff at CompUSA and Micro Center and all those places.


Alexis: I think there’s a jump here that people have to make, from when programming starts out as just your hobby, to something “Hey, I could make money on this!†When did that realization or transition happen for you?


Brad: Well for me, when you’re working on something, it doesn’t matter what it is. It doesn’t have to be a game. When you reach 80% of the way – you can make anything 80% and still be having a pretty good time. For me, if I’m doing something I really don’t enjoy or that involves things I don’t enjoy, then I’m going to charge money for it.


When I’m making a game, if you go online and you talk to indies, a pretty common thread is they work at a game and they get to that point where it’s no longer fun to work on it, and they usually stop, and that’s what makes an indie hobbyist game developer what they are. There’s nothing wrong with that; that’s how almost everyone is, but it’s that last 20% where you’re just doing all the cruddy work that’s involved in taking it, to finish it – that’s when I decided that I was going to charge for it.


It’s not that I didn’t have fun on the game; I had a lot of fun for 80% of it, and then the last 20% of it was just working on – you usually just know when it stops being a hobby and when it becomes a job. It’s when it becomes a job that you decide – in my case, I decided I was going to charge for it.


Alexis: For me, as a web developer in my spare time, it’s always when it gets to the polishing everything – make sure everything works on one browser and the other browser and that it looks good and that the UX is on point and everything. I can totally sympathize.


Brad: Yeah, that’s a great analogy. It’s when you’re like, “I don’t use Opera, but I’m going to have to modify my CSS to make sure it works with Opera. This is a job, and I’m doing a job.â€


Alexis: Right. Let’s see. At what point did you decide Stardock can grow or needed to grow?


Brad: When I was still back – many years ago, I would say it was in 1995 and I realized that if –. First of all, we had a tremendous opportunity because at the time, I was very passionate about IBM’s OS/2. In the same way you hear people are very passionate about Linux, I was that way for OS/2. I was 20 years old and as 20-year-olds tend to be, we get very passionate about strange things.


Alexis: [Chuckles] This is the truth.


Brad: And I was a rabid, crazy, OS/2 zealot and I thought, “Well, I had to do this for the cause,†and that is everyone needed to learn how important multi-tasking was. It’s very geeky in hindsight – in 32-bit! If I was willing to basically pay myself nothing but enough to afford an apartment, then I could use that money to hire other people who could take care of things to help us grow, and then keep reinvesting it over and over again.


Alexis: Who were the first people that you hired? Did you go for marketing and sales? Did you go for another developer?


Brad: Oh, tech support [laughter]. That’s easy! Tech support was the first thing I hired, and then the second person was someone to handle orders coming in. Those were the first two key things. I was doing the programming, and then I needed someone to handle the tech support for people who had problems. I always considered that my first responsibility is to people who have already given me their money, so that’s why support was first.


And then the second part was to handle the people who want to give us their money, and that’s where having someone answering phones or faxes and fulfilling orders came in.


Alexis: You’ve got your priorities straight. Those are some very important number one’s and number two’s.


Brad: Yeah, exactly. A lot of people make business way more complicated than it has to be. What is a key, what is a core component? What is the core of any business? Well, you have to make money if you’re a business, so the first step is to have a thing that you can exchange for goods and services. Money is exchanged for goods and services – well, I have to have a good or a service to exchange for money. I have that good, and I can take care of all three parts myself at first; you have to have that nucleus.


And then so I can concentrate on more things, well then I need to hire someone who can help the people who have already given them money and have enough money to help me help people give me money [chuckles].


Alexis: Yes, please. Give me more! [Chuckles]


Brad: Right, exactly, and then you just keep growing from there. The next person I hired was an artist, so I didn’t have so much programming art; and then marketing, someone who could help me on promoting the games so that other people would know about our good that they could give us money for so they could have the good and then be able to use our support if they needed to.


Alexis: Now continuing this thread of support, you’ve been doing support – Stardock has been doing support for years and years now. What are some of the things that you’ve learned that if you could bestow this knowledge onto fellow programmers who are selling their own software, whether it’s a desktop app or a game, that would make life easier?


Brad: Well nowadays it’s changed so much. I mean, one of the things that we make so much use of are analytics. It is worth your time to put in mechanisms into your game or your piece of software that tells you where it’s crashed for a user and automatically submit it.


There was a time where it would actually bring up a dialog saying, “Do you want to submit your crash report?†It’s like, “Duh!â€


Alexis: Copy-paste this and –.


Brad: Yeah, that’s insane! Yes, you always want it and as long as it’s anonymous – I mean, it’s not some violation of –. Someone’s going to get unhappy about this, but the idea is the more crash reports you can get – because don’t take any of their information; you just simply have a crash at this location. That’s all we get. But putting that in and finding out, “Oh, wow. We’re getting a lot of crashes at this spot; we need to find that out†whereas if you rely too much on forums or tech support reports only, you don’t get a good idea of where the real problems are, because you don’t want to be fixing the – they say the squeaky wheel’s what gets the grease. But what about all the people who just go and don’t send it? They have rights.


Again, everyone who’s my responsibility or any of the responsibilities of someone who sells something is to make sure that their customers are taken care of. They have a moral obligation to them. I don’t consider it a customer’s responsibility to go and put in any effort to help us to fix their problem; that’s our responsibility.


And it pays off. If you can put in the time to put code in to automatically report where something has crashed, that is absolutely worthwhile.


Alexis: So it sounds like you bootstrapped Stardock out of your college dorm room. Did you ever at any point take investors or venture capital?


Brad: No. We’ve always been self-funded from day one to – there’s not a cent of investment capital on Stardock.


Alexis: Was that a “Well we don’t need investment so we’re not going to take it†or “I really don’t want investment; I’m not going to take it�


Brad: That has come up over the years, and part of it is because I didn’t know any better. Nowadays, with the Internet and how aware we are of how businesses and how tech companies, tech startups, function, it’s easy to forget that in the ‘90s that didn’t exist. It never even occurred to me that I could get investment.


And then by the time that I knew about it, it wasn’t worthwhile because at that point I already had – it was too late. Every few years, a big company will come and try to acquire Stardock. There’s so many hardware OEMs and pretty big companies – huge companies – it would probably not surprise your listeners who has come and tried to acquire us over the years.


The problem is that it’s too late, realistically. If I’m effectively the sole owner of the company, at this stage, there’s nothing they can do for me that would make it worthwhile. But early on, I would’ve accepted an investment for those first five or six years. Certainly in the late ‘90s, I would’ve, when OS/2 died. We’re not using OS/2; we’re using Windows – that was a very rough time. I would’ve totally taken an investment then.


Alexis: That’s actually one of my questions that I’ve got here. It’s a bit further ahead, but let’s dive into it since you brought it up. It was a big change, needless to say, and how did you handle that? If there were any lessons that might apply to other seismic shifts that often happen. For example, maybe the most recent one for most programmers is the mobile shift, where if they didn’t get on that train they were pretty late. What are things to keep your ear to the ground on or how can people handle change?


Brad: I would say, broadly speaking, cognitive dissonance kills more businesses than anything else. I consider myself unusually susceptible to cognitive dissonance. My personality profile makes me very much blind – I would hate to say willfully but looking back, I had had plenty of warnings that the OS/2 market was dying. I just chose not to believe it.


Usually, if people can just take and accept that there is a tendency to believe what you want to believe – to believe things that aren’t true to be true because you want them to be true – that’s what I think kills of a lot of businesses.


Probably the other thing – and this is an ugly truth of businesses – if you talk to a banker or an investor about what kills most businesses, it’s their unwillingness to lay people off. To be perfectly honest, I hate to say it, but when the OS/2 market – what happened with the OS/2 market is that when Windows NT 4.0 shipped, the OS/2 market died within a month. We went from making hundreds of thousands of dollars a month in software sales to making less than $20,000 within – yeah. Except for our enterprise sales – those are hit and miss.


I had had plenty of warning; I had had the NT beta. My gut told me that this was going to be really bad for OS/2 and I just kept deluding myself – self-delusion. But I didn’t take any proactive steps and laid people off, and because also they could have gotten another job, and it could’ve been done so much easier than waiting until the last minute where I did have to lay off a bunch of people but under much harsher circumstances, because I had no other choice. By ’98, I was close to not being able to make payroll.


Alexis: Man, that’s a never a good situation to be in.


Brad: Oh, it was awful! Luckily, I borrowed everything I could from the bank on my house and everything I owned in order to make that transition to Windows. Obviously, if I had not been self-delusional, suffering from so much cognitive dissonance, I would have been able to pre-plan and move to –. I mean, I had done something. We had a game called Entrepreneur and we had made a Windows version.


Alexis: Ironically.


Brad: Yeah, exactly. Luckily, having a game available for Windows made all the difference. It was ironically – it wasn’t the game that saved the company; it was our customer loyalty that saved the company. But we could’ve done that a lot better if I had been paying closer – not closer attention, but – you know what I mean.


Alexis: Yeah. If I remember correctly – I don’t know if it was that piece of software or another – you had asked customers to do a subscription model was it to support the company?


Brad: Yeah, it was Object Desktop. Our main product on OS/2 was called Object Desktop and when the OS/2 market collapsed it would take us – Object Desktop was a nontrivial piece of software and it would take us two years to make this game.


Some people have recently started to know who I am on Twitter and they’re surprised why I’m very, very pro-customer. People say it’s cliché-y. They just say, “Customers are this†and it’s like, “You know, no.†I hate it when I see people say stuff like that and they don’t really understand it, but I lived it. I’ve seen what customer loyalty is.


Think about it. Let’s go back – it’s 1998. There is no such thing as Kickstarter. There’s no such thing as early access or anything like that. There’s no such thing as digital distribution; our customers had bought our software in a box at a store. It’s going to take us two years to make Object Desktop for Windows. I don’t suppose you guys would preorder it right now and wait two years. We had probably somewhere between 10,000-15,000 people do that.


Alexis: Wow.


Brad: Yeah. We delivered what obviously became a pretty fantastic product. I mean, Object Desktop delivered to Windows users things that people take for – if you look at Windows today, you look at what Windows was then, it’s pretty obvious the effect Object Desktop had. Zip files were treated as folders – that was completely unheard of back then. Skinning, GUI skinning, the control center, a dock on the bottom – all kinds of stuff. That all happened only because I had capital, so in a way, our customers – it was like Kickstarter in a sense.


Alexis: Very much so, yeah. So speaking of the very close relationship you have with customers – in fact, one of my notes says “The title for coolest CEO either goes to you or T-Mobile’s John Legere†[chuckling], in part because if you check out your blog or your Twitter account, you see lots and lots of “Hey, we’re working on this game; this is our progress. If you want to join in on the beta or the alpha or this kind of thing –.†You’re very, very involved; that is very cool to see.


Brad: I very much enjoy it. I started out as a creature of Usenet. I made games and software specifically so I can interact with those people. I’m very much an extrovert. Nothing has changed, not that it mattered, now it’s even better. This is the best time ever to be making software because this is going to – I’ve actually told marketing that this should get out there, but no one would want to cover this, they said.


Sorcerer King, which is a game we’re working on, has hundreds and hundreds of quests. One of our lead writers at Stardock, his night job so to speak, is writing for cracked.com. He writes these amazingly hilarious stuff for Sorcerer King, and he’s also working on Star Control.


We had the quest, but they had to be implemented so I went on Twitter, and in fact, I had gotten some trouble for this recently, but I recruit people off Twitter all the time and say, “Hey, would you like to work on here?†We do these contracts – little, tiny contracts – with people for just a few hundred dollars here, $200 there, but it’s the only realistic way I could get all this content into the game. They get credits in the game and they get paid, and we get hundreds of quests into the game, and it’s such a win-win.


Alexis: You still program on these games, right?


Brad: Yeah, on Sorcerer King I’m the Lead Developer.


Alexis: Yeah, I was surprised by that because I’ve talked to quite a few CEOs and founders over the past 22 episodes or so now, and most of them – if not all of them – have said, “Well, now I don’t code anymore because I’m too busy handling this set or the other thing,†and it is quite impressive to see you still getting your hands dirty with this kind of thing.


Brad: Well, I don’t sleep very much. Apparently it’s a condition or something, but I only sleep a few hours a night – no matter what. Even if I go to bed at – and that’s what gets me into so much trouble, too, online over the years too. I’ll be up at four in the morning, waiting, and while I’m waiting, “Well I wonder what’s going on in such and such social network of whatever form†because I have nothing better to do because there’s no on here yet.


Alexis: [Chuckles] If I may inject a personal statement here, I think getting yourself into trouble constructively and selectively is what makes life interesting.


Brad: It does. I have to say overall, with only a few exceptions, it makes it where it’s totally worthwhile. I love talking to people on forums and I’d lurked on even more forums and obviously social media, and it’s just very enjoyable. I think it makes our stuff a lot better too, just in terms of being able to hear what people think about different things – and not just about our games, I’m talking about games overall. It just makes all things very worthwhile.


Alexis: In following this close relationship to the community, your games have no DRM, right?


Brad: Nowadays, since they’re on Steam, they have whatever’s on Steam, which we don’t really consider – I know some people consider a DRM, but that argument, logging on to a forum or logging onto Twitter is DRM, right?


It’s not that I’m anti-DRM; I shouldn’t admit to this, but sure I’m going to and this will probably blow up in my face, but Seth’s not here to stop me. Back when I used to teach classes in college, Assembly language, and I would crack copyright protection as part of my demonstration of why you’d want to learn –. They were like, “Why do we have to learn Assembly language? What are going to learn Assembly language for?†I’m like, “Well, let me show you,†and I would blow it up.


There’s a game called – what was it called? Powermonger. It had the most obnoxious copy protection. I’m like “See? Look here. You see this CMP here? You just change it to – you just say jump, no matter what.†Anyway, I don’t like being made to feel like a chump when I paid money for something – that’s what I don’t like. It really annoys me.


Now if they make a DRM that can sense it’s me, then I don’t care. I don’t like piracy; piracy annoys the heck out of me, but I don’t want to be entering codes. I don’t like entering codes unless there’s a real justification – I don’t like having to remember “Did I buy this? I don’t remember. Do I have the CD key? Did I lose the CD?†If it inconveniences me, that makes me unhappy.


Alexis: So you’ve made the shift from game dev to game publisher as well. I mean, Stardock is still developing games, but you’ve also taken on the responsibility of publishing other developers’ games. Was that a natural progression? Why did you decide to make that jump?


Brad: Well we’ve been publishing third-party games since the ‘90s on OS/2. In fact, that’s how I got to be friends with some of the guys at Ensemble; we worked together on OS/2.


Realistically, part of it is that there are some amazing talent out there, but the publishing system is very formulaic now. At large companies, all policies are designed around working around the average person’s capability, and so they don’t really know what to do with really cutting edge stuff usually, because no one wants to be able to make a judgment; they want to adhere to a policy or a metric or a study.


When you do see amazing stuff out there – and all they need is capital; they just need money to get their shot at it and that seems like a no-brainer to go and help that come to market. I mean, look at what Mohawk’s making. That game is going to completely blow people away, because they’re going to go “Oh my God! There’s nothing out there –.â€


Like I said, I have to explain what off-road is. I’m like, “Well, remember M.U.L.E.?†No one remembers what M.U.L.E. is, which is a sad thing because it’s one of the best games ever. It’s like that, but times a thousand.


Alexis: For folks not familiar with Mohawk, tell us a little bit about what they’re working on.


Brad: Mohawk is founded by effectively a bunch of Firaxis veterans. Soren Johnson designed Civilization IV; Dorian Newcomb, who’s their Vice President, he was the Lead Artist on Civilization V and I think Civ IV as well and just all kinds of other stuff. They’ve put together an amazing team of people, but like I said, most of which came from Firaxis at various parts.


They’re making a game called Offworld, which is supposed to go into early access sometime before GDC, and their game is like an economic war game. You’re not moving soldiers around or anything; you’re basically crushing other people in the market. It’s got a really good – the single player is pretty compelling.


Alexis: Quick! Play high frequency trading! [Chuckles]


Brad: Yeah, but on a multiplayer, it also makes a – even in beta, the multiplayer is so compelling that we had to ban it because people were “helping†test it around the office and it got out of hand because the game is so addictive.


Alexis: Well I’m looking forward to it.


Brad: What’s amazing is that the team on there – I mean, it’s got to go on early access. The game won’t even come out until 2016 and it already, like I said, we put in the ban on multiplayer or playing at it last summer. The game is going to be so balanced and part of the reason —.


Alexis: It has a lot of depth.


Brad: Yeah, it has a lot of depth, and the reason why it’s has such a long gestation is because no one’s made a game – I can’t even point to a game, like I said, other than maybe M.U.L.E. – that it’s similar to. I don’t have a way to describe it, but it requires tons and tons of feedback to make everything feel good and right and fun. It is extremely fun, but we just want to make sure there are all kinds of different ways to play the game.


Alexis: If I understand the role of a publisher correctly, part of their billion of responsibilities is to market the game and to help spread the word. When it comes to doing that, and more broadly at Stardock, and especially the beginning and how this has changed over time, how have you spread the word about your products and your company?


Brad: That is always evolving, especially in our market. Back in my day, it was actually fairly straightforward. You worked with retailers to make sure you had market development funds so they could promote their game through their channels, and then you did advertising with PC Gamer and Computer Gaming World and Computer Games Magazine and a bunch of other places. You just wrote checks and put out a big ad and made sure you sent three review copies out to all three game magazines and that was it.


Nowadays, and I commented this in an interview somewhere where the marketing team gets back and they have all this budget stuff to do. A huge chunk of their budget next year is targeting YouTube people. I shouldn’t be using that term because there’s probably an official term for people who do this on YouTube.


The only one I’ve been watching –.


Alexis: Let’s Players.


Brad: I’ve been watching TotalBiscuit because I’m a Starcraft guy for a long time, but I didn’t realize it had become such a huge thing. I mean, I was a Shacknews or Rock, Paper, Shotgun kind of guy and that’s where I traditionally – or PC Gamer, Joystiq – I don’t want to play favorites, but I generally like them all. That’s how I looked at it, and now things are moving so much more to – it’s so much less, what is the right word for it? It’s dispersed. Excuse me, I don’t know a right way to describe it, but there’s a billion channels —


Alexis: Fragmented.


Brad: Fragmented, yes.


Alexis: Yeah, we spoke to a developer at Yacht Club Games, which developed Shovel Knight and released it recently and when they did their Kickstarter campaign, something that they said that I was surprised to hear was that 50% – roughly or maybe even more – of their Kickstarter backers and money came from YouTube – from let’s players and YouTubers who had mentioned their game or played the demo and talked about it. That baffles me as well.


Brad: Yeah, the thing is that I have three children; my daughter watches let’s plays to the point where we have certain days she’s not allowed to watch let’s play because she’s so addicted to it. I’m not sure which one she watches, but she loves watching people play Pickman on YouTube, and my sons are big into it. My son only uses YouTube now to decide when he’s going to buy games.


Alexis: So in 2008 I believe, Impulse came into the world, which was your digital distribution platform and you eventually sold to GameStop in 2011. Guide us through that journey and what made you decide, “You know, it is time to sell.â€


Brad: We had started doing digital distribution back in the late ‘90s for Object Desktop. We couldn’t make a box for Object Desktop for Windows, so we built a digital distribution platform with the fancy name of Component Manager, because that was the kind of marketing geniuses we were. And then we eventually renamed it Stardock Central, and then we eventually renamed it as Impulse. Impulse is when we opened it up for third parties to put their stuff on.


We were already using Stardock Central to distribute, so someone could go up, get Stardock Central, purchase, download and update other Stardock software, and we’ve been doing that for many years. It never occurred to us that there would be a market – and this is something people always were surprised even back in the day when we had Steam that I would say, I’m a big Steam fan. You’re not supposed to say that, right? This is one we’re supposed to be in hot competition. I’ve been in interviews and I get these – my PR people would be like, “What did you say?!?†“But I love Steam! I like what they’re doing! It’s good for the industry.†I like Impulse too, but I never had anything to criticize Steam for because I thought they were doing everything right – and I still do.


The curation thing is one of the most ingenious – anyway, I don’t want to gush about Steam, but I’m quite a fan. Impulse did really well. If you say well, it was nothing compared to Steam – and that is true. Steam had two-thirds of the market; we had 10%, so it’s much smaller. But 10% of a gigantic market is still a pretty good chunk.


But there was one downside for it, and that is in order for us to do what we had to do – this is one of those things a lot of people take for granted. Recruiting talent is not easy, especially if you’re located in Michigan. We’re not in Seattle; we’re in Michigan, and so –.


Alexis: Or San Francisco.


Brad: With the story of Impulse, it always boils down to that simple fact – we were in Michigan, and in order for Impulse to be effectively competitive, I had to take my best people off of other projects and put them on to Impulse. There was no way a company our size could do something as – Impulse, especially at the time, was just absolutely state-of-the-art, but it meant that we had to take people away from our best stuff. You really saw the consequences of that with Elemental: War of Magic where I had my best developers run Impulse – and it’s not that we had bad developers on Elemental; our biggest problem was that they weren’t experienced


Alexis: They were shorthanded.


Brad: Well, they weren’t so much shorthanded; they just weren’t as experienced. They’re a lot younger and less experienced. The other thing is that the other guys knew how to work with me. My Elemental design at the time was essentially, I want Master of Magic, multiplayer, better graphics. I’ll be back in two years.


Galactic Civilizations – so much of the excellence of Gal Civ was because my team would take my vague instructions and turn it into something amazing. The whole battle viewer in Gal Civ II was because I went and said, “I want a dialog that pops up like Warlords†– the old Warlords game, it shows you the results or shows units dying. And Jesse our lead on that took it and made something amazing with it.


But in 2010 he is on Impulse, and Cari was on Impulse. He was our Lead Developer in Gal Civ I and Gal Civ II; he was on Impulse and on and on down the line. I even had the Fences team and WindowBlinds team all working on Impulse. At Elemental, we had decided, what do we want to do? If we’re just here to make money, well then I don’t want to be part of that. That’s not fun. We either had to make games, or we’re going to have to do Impulse; we couldn’t do both. We decided we were going to – or I should say not even just make games — or we’re either going to make software. Most gamers aren’t aware of this, and it’s just like the software aren’t aware, but there was a big dead zone at Stardock at both Object Desktop and on the games because all of our top talent was on Impulse.


We decided we wanted to get out of that and we wanted to be able to – we found what we thought as a good dad, a good new dad, for it and sold it to GameStop. We used the capital from that to put it into this fund, which we’ve been using for investing in these new startups.


Alexis: Really now? Tell me a little bit about that.


Brad: Which part?


Alexis: [Chuckles] The startup fund.


Brad: Oh sure, a lot of people don’t realize this, but even at the time we sold Impulse, even during Elemental, we were profitable. The difference was that when we sold Impulse, we got a lot of money, and so I took that and I put it into a portfolio fund, which we call the strategic investment fund, because again, we’re the same marketing geniuses that came up with the name Component Manager.


The idea was that anytime we saw something that looked really promising, we’d use this fund to help them fund their company and that’s how we got Mohawk and Oxide and Mothership and the various unannounced titles – companies that I can’t yet discuss – and then allowed them to do some pretty amazing stuff.


Alexis: Okay, let’s see. Partnerships are one thing – actually, before we dive into partnerships, that split between the two sections of your audience, the gamers and pretty much everybody who buys the desktop software like WindowBlinds and all these kind of gooey things, you’re clearly conscious of this, but have you tried to make each bucket aware of the other? If you have, what have you done to unify the two?


Brad: In a way, Impulse unified them because they’re all working together. They get to find out about the different sides of the company. Next year, I would be willing to bet that most of our revenue will come from the games, which is quite a change from the past. It’s amazing how you see online people still say, “Wait, Stardock makes games?â€


Microsoft gave us the biggest gift with Windows 8 ever with Start8. We were able to make the Start menu replaceable, which is also depressing because that took a weekend to make, and that makes more than most PC games gross.


Alexis: Wow.


Brad: That thing is insane. If you think of how many Fortune 500 companies, they upgrade their machines to get Windows 8; they don’t want to do training, and it costs five bucks, and we give a discount if you buy more than 5,000 or 10,000 at the time. It was just unbelievable.


Now Windows 10 is not going to be so good for that, but there’s always something. They’re doing some dumb stuff with their Virtual Desktop, which we’ll be able to fix, but there’s always stuff to do. I would say now, the game side and the software side are still not super aware of each other. There are some intermixing; I sometimes whine to the software side to come give me a technology to put into the games. I wanted our games to be more high DPI-aware, because I’ve got a 4K monitor and some of our games look awful. You couldn’t see it – it’s so tiny! I’m like, “Alright, I need you to make our games more high DPI and you have two days. Give it to me.†And I got it.


The Sorcerer King looks amazing on a 4K monitor, and it’s not blown up looking. There’s all kinds of crazy stuff to make it look amazing.


Alexis: That’s was one of your blog posts, yeah?


Brad: Yeah. I couldn’t believe how much better they made the game too, in terms of reading the text and how sharp it is.


Alexis: You mentioned hiring and that last count, I think, Stardock has 40+ employees?


Brad: Oh, no. That was years ago. I actually don’t know. It gets a little blurry too, because –.


Alexis: Where the portfolio company is?


Brad: Exactly. I think there’s 70 in the building here, and there’s another 20 – I don’t know. The group in Baltimore, and there’s a group in Austin.


Alexis: [Chuckles] There are a lot.


Brad: Well, not a lot. I don’t want to imply that we have a thousand employees, but it’s probably over a hundred, but it’s not like hundreds and hundreds. We’re still a small company.


Alexis: Right. What are some qualities that you look for when it comes to hiring?


Brad: Initiative. Self-starting.


Alexis: Mm-hm.


Brad: If someone’s just sitting there, waiting to be told what to do, that’s generally not the way to go for us. We need people who will fill in their gaps in their direction and they have good judgment.


Alexis: Alright. Now for partnerships, that was a bit of a diversion there for a while. HP has had a partnership with you; you’ve had a partnership with them with Fences Pro on their Envy line of laptops; and Microsoft has contracted you for – I think it was the animated wallpapers for Windows Vista Ultimate; you had something like an OEM deal with AlienWare, so partnerships are a key thing for Stardock. It seems so, from the outside. What led you to get into these partnerships and what have you learned from them? What have they meant for the company?


Brad: We used to make the joke back in 2010 and there would be someone who’d write online, “Stardock’s in big trouble because of Elemental,†or “I’ll never buy anything from Stardock or something.â€


Alexis: And you were behind it.


Brad: It’s like, you already did. If you bought a PC, you gave us money. It didn’t matter – Dell, HP, Lenovo. It didn’t matter. Heck, even AMD Catalyst’s Control Center thing is it’s user. I mean, they still use our tech and that’s another discussion. Working with other people allows you to one, solve their problems and we have all kinds of stuff. We worked with Intel on their Wide-Eye stuff; we worked with AMD on Mantle and there’s Intel – oh my God, I can’t say enough good things about Intel. They are one of my favorite partners to work with. I just didn’t realize – sometimes I forget how many different things were working with them. Microsoft’s really good.


One of things why it’s really fun to work with them, with these partners, because you usually begin with the really sharpest parts of their company. I don’t know if that’s because of the nature of what we make and so we get to team up with their best and brightest, but I like having win-win situations and almost every partnership I can ever think of off the top of my head, it’s been such a win-win for everyone involved, and ultimately, that’s what determines whether a partnership’s going to work out or not.


Alexis: Say, I’m getting into a partnership with company x and let’s say they’re much larger than me or something, what are some pitfalls that I should watch out for, some things that I might screw up or they might screw up in?


Brad: Someone who’s not experienced with needs to realize that if you signed an agreement, it’s not their responsibility to make sure you benefit from it. It’s your responsibility to make sure you benefit from it. Just like it’s their responsibility to benefit.


Now, a good partner will absolutely work with you, making sure it’s a win-win out of it, but a lot of people think that if they have an agreement somehow, the other party, it’s their job to not just their part of the bargain, but to also make sure you get what you wanted to get out of it, and that’s expectation that can lead to some dark places.


Alexis: Right. A lot of devs with talked with have complained about pricing, that it’s some of one of the hardest things that they have done and continued to do. Has it been that way for you? What have you learned about it?


Brad: I’ve learned that I’m always wrong. The only thing that I know is that 29.99 is a bad price point [chuckling]. I struggle with that so much; it’s one of the things that I – like early access. I would love to have early access, be steeply discounted, but where our distribution is now, I can’t do that. Unless I want to set the main price to be that, then people don’t realize they’re getting a discount.


Steam has certain policies that if you have a game on early access, they don’t want you saying your game is 25% off for nine months. I have always wanted to have steep discounts – steep, ongoing discounts – on early access games, but on pricing in general, I don’t know. I tell you, I just suck at that.


Alexis: [Chuckles] Hire somebody else to do it for you?


Brad: No, I don’t think anyone – I think we all suck at it.


Alexis: You just learn as you go along?


Brad: Yeah. I’m very bearish when it comes to the mobile market. I don’t think the mobile market as it stands today – and by mobile, I mean iOS and Android. They’re not viable as they stand now. There’s not enough market there to justify a $2 game; you might as well be winning the lottery, and I will say that that’s one of the things –.


My biggest concern about the game industry is that there’ll be some sort of race to the bottom where we’re expected to spend a million dollars or $5million on a game and then give it away for two bucks. That’s insane. You can’t do that.


Alexis: Heck, just look at Angry Birds and Rovio how lately they’re not doing so well.


Brad: Yeah, and remember, they’re the ones who hit the lottery, right? For every Angry Birds there’s literally 10,000 people you’ve never heard of who made – Angry Birds is literally winning the lottery, because that game was available on DOS as guys wearing bananas. Remember the ape drawing bananas thing? That is a very classic game design.


Alexis: So when it comes to other kinds of monetization, you also do support, right?


Brad: In terms of?


Alexis: Enterprise support for other programs.


Brad: Oh yeah, we do enterprise support agreements.


Alexis: I’m curious here, and it’s alright if you can’t or wouldn’t like to divulge, what the breakdown is here in terms of your revenue for games and for support and for desktop software.


Brad: Oh, it varies so much from year to year. If a game is – I think I’m going off the top of my head, but I’m pretty sure the only one year that games notably made more than the software, that was when Sins of a Solar Empire shipped. That was a big year.


Generally speaking, software’s been the bulk of our revenue, but I think next year, because we’re supposed to release four titles next year and then we’re supposed to release another four the following year, and none of those being Star Control because the 25th anniversary of Star Control 2 is in 2017 so that’s our target. I think when that happens there will be a permanent switch. I don’t think that software will grow fast enough to keep up with that bump in game revenue. That’s going to be a weird thing.


Alexis: [Chuckles] So what advice would you give to folks when it comes to building a community around their software?


Brad: Good form software [chuckling]. Good form software, accessibility, you need to dedicate time to talking to them. This is one of those debates that people have all the time is you have to remove the toxic elements from your community aggressively or else they will poison it. A lot of people are afraid to lose anyone, but for every poster there’s ten lurkers; that’s one of the things we’ve always seen. I mean, our communities are almost unmoderated at this point, because when we first started, there was a code of behavior, of politeness – not politeness, mutual respect – that people were expected to have and people then adhered to it or were removed. Now it basically takes care of itself. It’s so rare for someone to come in and behave badly because people won’t tolerate it, that people won’t conform to the tone of the existing community.


I guess my point is, if you want to set up a community, decide what you want your community to be like at the onset of it, and police it aggressively. If you do that, it will take care of itself to the point where, like I said, our forums are largely unmoderated now.


Alexis: So in your 20+ years of Stardock – I’m not pulling the age card here; I’m just pulling the experience card [chuckling] – what’s one mistake you’d rather not repeat?


Brad: Oh, Elemental. Not a day goes by where I don’t think about that game. I mean, that game – I probably think about that game a couple of times a day.


Alexis: [Chuckles] It pushes you to do better?


Brad: It does. I mean, it completely changed the company. To this day, I have enjoyed working on that game more than any other game I’ve ever worked on, and that’s one of the reasons why it was so bad.


To be fair, it’s not as bad as – a lot of people know this, but Stardock, if you were to go on metacritic.com and look at companies who developed and published games so the key thing is developed and published, we have the highest metacritic average of anyone in that criteria in the industry.


At the time when we released War of Magic, we had never shipped anything that had gotten less than – a full-priced game, we had never shipped anything like that gotten less than a 75. When that thing came out, it got a 55 or whatever – that was just, oh my God, it was the end of the world!


Alexis: A shock, yeah.


Brad: Yeah, but people forget that you have all these people who have never played that game and they’ll say, “Oh yeah, I heard War of Magic was just total garbage.†It’s like, “It wasn’t total garbage; it was just –.†We were so disappointed with it that we basically did our anti-marketing on that by saying, “Yeah, that game was a total disaster.†“Why was it a total disaster?†“Well, it only got three stars.†[Chuckling]


And the reason why people are so angry about it – and why I’d been so angry about that particular title – is that it had such huge potential. It was so, so close to being something just amazing.


Alexis: I remember seeing screenshots and art from it; I’m like, “Man, this game looks cool.â€


Brad: Yeah. I mean, the whole world was transformable – this was in 2010 tech. You could design your own units; every single unit in the game, you could design. I mean, full-blown clothing and stuff. You could get married and have children and your children, actually there was a genetic algorithm in there to combine parts of the models together to create a new model – a new texture, actually, not model, to be specific – that didn’t exist otherwise. A completely new texture was generated from this, and it was just absolutely amazing.


From a modder’s point of view, you could do anything with it. I mean, you could’ve made Baldur’s Gate with that, with the engine, but at the end of the day, what we’d released was less a game and more of a technology demonstration. At first, a very buggy technology demonstration.


Alexis: That’s a good way to look at it.


Brad: Yeah.


Alexis: Now on the flip side, what’s one decision that you’re particularly proud of?


Brad: Professionally, or –?


Alexis: I was thinking professionally, but give us what you got.


Brad: Well, personally, marrying my wife. That was my best decision, but professionally – oh gosh, I don’t know. I think in terms of something that would be beneficial to other people – delayed gratification. I’d lived very poorly – not poorly as in unhappily, but in terms of economically –. I basically didn’t pay myself for the first several years of the company so that I could keep reinvesting it. I didn’t go and make my first million and then –.


Alexis: And get a Tesla, well not all the time.


Brad: Well back at the time it would’ve been a Porsche or something. I did eventually get a Porsche, but that’s the thing – people were really surprised. Have you ever seen a picture of the Stardock building online?


Alexis: Yes.


Brad: I bought that building before I had a big house. It was a regular house. And the building that we’re in, I own it. It’s not leased. My point on that is that delayed gratification – if you are willing to spend your 20s investing and you’re delaying that urge to get stuff, you can spend your 30s and 40s reaping the benefits of that investment, because it compounds over time.


Alexis: Now curiosity here, did you ever decide, “You know what? I’m going to –.†Let me rewind there. Before starting Stardock, programming your first programs and games, did you work for somebody else or did you always just jump right in and gave it a shot at starting a new company.


Brad: In high school I worked at an excavation company. My job was to clean trucks and paint – the very unpleasant stuff. In high school, I worked at the mall; I was at B. Dalton selling books. Then I drove a van for the bank also in college, and I taught classes for the university. You know those terrible lab instructors that people get? I was one of those people.


Alexis: So you kept busy, yeah?


Brad: Yeah, I was always just trying to generate enough income to pay for school. I didn’t intend for Stardock to be my career; I mean, I was only doing this until I could get a real job.


Alexis: [Chuckles] What did you intend to be your career?


Brad: Oh, I was going to design CPUs.


Alexis: Huh. Do you still dabble in, personally, working with Arduinos and hardware and that side of things?


Brad: Not as much as I used to. I mean, I still have a breadboard and stuff. Most of the time when I have free time, I like to do stuff with my kids. My kids are so amazing; my eldest son – do you watch The Simpsons?


Alexis: Oh, absolutely, yeah.


Brad: There’s an episode where Homer pulls Bart out of school to work at his grease company, do you remember that one?


Alexis: I can’t remember, no.


Brad: Well, I felt that way. Before I went online and went on Twitter and started recruiting people to help us work on the quests and stuff, and I’m looking at this XML and like, “I wonder if I can pull my kid out of school and maybe he’d get a credit hour to implement this stuff.


He, interned here in the summer fixing key frames in Maya. They’re all self-taught. I mean, they’d grown up with all this stuff around that they’ve been messing around with computers since they were kids, but that’s what I enjoy doing if I free time, doing stuff with them. We have all kinds of weird hobbies; we got chickens and bees and –.


Alexis: Do you mind if I ask how old your son is? Because you mentioning an eight-year-old messing with key frames in Maya is kind of blowing my mind.


Brad: Oh, no. The one that did that just turned 18.


Alexis: Okay.


Brad: And then I have a 15-year-old, and then I have a daughter who’s eight.


Alexis: So the last and perhaps the most important question of all – when it comes to code, what’s your text editor of choice?


Brad: I am still a big fan of Visual SlickEdit.


Alexis: Huh, this is a new one! Okay.


Brad: No, it’s been around for years!


Alexis: I should say, new to me.


Brad: Unfortunately, because it takes a long time to set up a good editor to be just right, I usually just sit down with Visual Studio and do everything in Visual Studio. They’ve done a good job, especially on more recent versions. They’re using WPF under the covers, which makes it a lot slicker.


Microsoft getting back to their bread and butter now, there are new seals and – Visual Studio is my –. I like Notepad++ for XML. There’s probably other stuff out there, but that tends to be the one I use for editing.


Alexis: Alright. So if folks would like to learn more about – I was about to say SlickEdit, but it’s not SlickEdit – [chuckling] Stardock, where should they go?


Brad: Just go to Stardock.com – S-T-A-R-D-O-C-K (dot) com. Next year, 2015, is when it’s going to get really, really crazy. I’ve tried reminding people here, because they’re saying, “Well we didn’t get coverage on such and such.†I’m like, “Guys, on the game site, all Stardock is, right now, as far as the public is concerned, is that we make these little, niched 4X games and we publish them. They don’t know – they better not know, anyway – about the titles in development right now, so hold your horses.â€


In a year from now, oh my gosh. We got four titles coming out next year, two of which are AAA – and we’ve never done a AAA game before. I mean, all of our games have been single A. Sins of a Solar Empire is probably a single A; it’s the closest thing we have to AAA, but traditionally, we’ve made those cool, niched 4X games and then next year it’s going to be a whole new world.


Alexis: And if we’d like to read up on your blog, where can we go?


Brad: I write all my stuff on littletinyfrogs.com.


Alexis: And if we’d like to follow you on Twitter?


Brad: I have the worst Twitter handle ever! It’s @draginol – D-R-A-G-I-N-O-L.


Alexis: Alright, and for us, you can follow us @Binpress and myself @alexissantos. Well, Brad, it’s been a pleasure!


Brad: Yeah, thanks for having me. Good to talk to you!


Alexis: Likewise! And hopefully we can check in next year once maybe three out of those four, or maybe we could wait until all four games are out and we can get a progress report from you.


Brad: Absolutely. Another time would be, if you happen to be at GDC, that would be a great time, because all four will be announced by then.


Alexis: Oh cool! Okay. Well that, sir, is a wrap. And we’ll catch you listeners next week.


Brad: Sounds good. Alright! Talk to you later!


The post Binpress Podcast Episode 24: Brad Wardell of Stardock appeared first on Binpress.