I enjoy your reports on the economic side of things. I also had a few questions if you want to waste some time answering them.
Thanks! And sure, I'm happy to answer.
- You cite the AIWars expansions as not being as profitable or as lucky as AIWars. Yet one could argue that the expansions are continual advertisements for AIWars and that the success of AIWars is entirely due to the expansions. How do you decide how much credit is due to AIWars and how much to the expansions?
Sorry, I should have been more clear. I absolutely DO think that the expansions are crucial, absolutely critical, to the continued success of AI War. And they have been enormously financially successful, too. The distinction I was trying to make is that they were not
lucky. In other words, they do not get a lot of press or anything like that, so they don't get a bump in that fashion. But the AI War word of mouth continues on largely thanks to the expansions, I believe, and AI War remains something that -- as a "franchise" so to speak -- the news DOES talk about, even though they don't talk about any single expansion in particular most of the time.
You can be successful without being lucky, or unsuccessful without being unlucky. In this particular case I was saying that I feel like the AI War expansions were not particularly lucky, and that their success is basically due to riding on the coattails of AI War, which was itself enormously lucky. If that makes sense?
- I've noticed Arcen products have been in a massive number of bundles lately. Indie developers are split in their ideas of bundles versus a fixed retail priced. Can you describe your strategies here and how you believe they will effect long term income generation? As an illustration, I've noticed that your weak selling titles are also the ones that are in bundles selling 5+ games for $1. Getting $0.05 for a copy of Shattered Haven isn't going to be helping its sales numbers very much.
Our strategy for this sort of thing really depends on the game, how well it has sold in the past, what our financial needs are at the time, how the title has been performing, etc. Once you pass a point where a game is no longer really selling anything notable, then bundles make a ton of sense. The amount of volume doesn't really matter. Hey, it's money where there was none before -- and having more people play our games and hopefully enjoy them is only a good thing. "Free" advertising, so to speak. There's always the chance those people will then buy our later stuff at full price, and they now know who we are.
The bundles also are something that really saved our butts a few times last year in terms of providing us the income we needed to stay solvent while finishing certain titles. Indie Royale and the Humble Weekly Sale, most notably. The two of those account for something like 90% of our non-Steam revenue, at LEAST.
In past years I was really against bundles, and felt like they were a devaluing factor, etc. And at this point I still do not do bundles on titles that are selling well unless we get some sort of specific benefit out of it. Aka a minimum fixed price per unit, or some other thing that makes sense. But suffice it to say, discount promotions are our bread and butter in terms of income (probably 85% of our income is made when titles are on some level of discount, I would guess, although I don't have numbers on that in front of me), and bundles are just one more form of discount.
Extreme bundles really should be reserved for titles that are otherwise not moving much, though. And for some of our titles that just aren't moving much, we always hope to see a bit of a snowball effect with more people playing and talking about those titles. And we do see that some, actually, but it doesn't translate into referral sales that I can see.
- You've talked a lot about the gross game sales, but how do they fare in number of units sold?
I honestly can't tell you, as I don't have those numbers ready to hand. I don't actually keep track of that, as it isn't something that is terribly actionable on my part. I can easily look on Steam and see what is on there, but in terms of units sold through bundles and whatnot, I would have to dig back through tons of emails. My best estimate is that there are somewhere around 750k to 900k copies of our games that have been sold, when you include each expansion as an individual unit.
I can tell you that on Steam, as of today, we've grossed $2,322,576 and sold 371,814 units total of everything with them. So that's an average of $6.24 per unit, but that's across all of our games, including our $5 expansions and $20 games and things on discount and at full price, etc.
A couple more from Steam:
Last Federation: $449,599, 26,622 units. So $16.88 on average. Bear in mind that regional prices also play into this some, since sale prices are way lower in Russia and a few places like that. TLF sold most of its units at 10% off at launch, then a solid bit at 0% off after that, and then a bump of them at 25% off and 40% off during the summer sale.
AI War Bundle: $491,028, 74,604 units. So the average sale price there was actually $6.58, rather than the usual $16.99, which sounds about right. We move tons of this at 75% off. Why don't we just lower the price? Well, experience shows us that if we lowered the price people would not buy more of it, we'd actually lose money. If we cut the price in half and then did a 50% off sale, we'd be selling it for the same price as a 75% off sale now, but we'd earn less money both during that sale and during the non-sale periods. Sad but true.
Discount sale in general have never been a bad thing for us, they really are our lifeblood. But doing a permanent price lowering (which we did with Tidalis, AI War, and AI War's expansions) is something I'm always really wary of. I think it actually did work for Tidalis and AI War, but it's not a decision you can take back, so I always am super cautious there.
- One of the themes in your post is the effect of luck in game popularity and sales. This is a theme that is central to a number of Malcolm Gladwell's writings (e.g., Outliers), that success is highly due to luck. However, some critics and researchers believe that Gladwell is really writing things that read well, but lack true scientific rigor, instead appealing to our human foibles. In this example, luck is a contributing factor, but there are many other causes as well. In attempts at ego preservation, people want to believe luck is a large factor in success, so that we can blame our failures and the successes of others on luck (and then its not our fault.
- Wow, that sounds pretty harsh.
I've read his books, but I don't really have anything novel to say about them. He makes some interesting points about a variety of things, but a lot of it is speculative, as you basically say. I think his point about the tech magnates both being at the right time and place and age as WELL as having the skills is dead on, though. Bill Gates and Steve Jobs and so on. That's not an excuse for anyone else, because plenty of other people had big successes before and after those two, if not on that level. Besides, Gladwell was talking about
extreme outliers, not smaller ones. Basically Minecraft and Angry Birds, not even something like Braid or Tiny Wings or Flappy Bird. And anything Arcen makes is on a whole other lower level than those.
AI War and Minecraft actually launched in the same month -- May 2009, I am
pretty sure -- and I think that both of them were recipients of substantial luck there. At that time it was next to impossible to get press to pay attention to indies, and yet through a lot of hard work and wiliness and having the right games, we managed to do so. Had we released our games a year earlier, I don't think the same things would have worked, at least not to the same degree. We'll never know for sure, though. Had we released a year later, we would have been more lost in the wave of other indies that were starting to come out, and thus would have had more trouble getting that early press and early mindshare.
All that said, a vast majority of it does come down to the game, of course. All the luck in the world (usually) can't sell a bad game, although the success of Goat Simulator is something that I think astounds even the devs, who openly tout the bugs and whatnot. But they did really nail the humor on the head, and I was greatly amused watching an interview video with them talking to an IGN guy while he was playing. So I mean, even there they managed to really put together a package that attracts people for likely a variety of reasons.
I come from a background in business to business software, and I can tell you that is very different. We had a few dozen clients, and they were enormous. With new clients we could personally fly there and persuade them that our software was what they needed, and that we could do a job. We could show them references of other clients who agreed that we were awesome. It was all very personal in terms of the level of interaction, and we could always pretty much understand why someone did or did not become a client of ours.
But when selling mass market to the consumers, we are by contrast pretty blind. The reason you do something is entirely counter to why person X might do the same thing. Goat Simulator isn't something that sells well because of ONE reason, I guarantee you. Different people are buying it for different reasons, and part of the reason it is so popular is likely because it has a perfect storm of reasons that keep enough people buying it.
Let's look at military history. I think that we can all agree that military strategy is enormously skill-based, and yet "no plan survives contact with the enemy." Sometimes there were armies that were just frankly unlucky, because they happened to get surprised at a time when they had no reason at all to expect it, or there was some other outside force (weather, etc) that crept up in an unexpectedly meaningful manner and changed the situation. Some of these things were legitimate mistakes, some were unforeseeable but at the same time were a "now we know not to do that again" learning experience, and other things were just simply a matter of "that was a freak thing that happened that one time, and probably never will again, for good or ill."
I'm not a military history expert, but I think that general sentiment is pretty much accepted. The best generals were the ones who reacted well to changes in the fortunes of war, and also who were equipped to do so. But even for some of them, eventually they fell because things just eventually didn't go their way despite their skill. And there are some frustrating examples of a really sucky commander getting really lucky and overwhelming a really awesome one, I'm pretty sure.
Anyway, I look at the role of luck in the mass market sales of games as being something along those lines. It doesn't mean that if you lose a battle you can just go "ah, that happens, let's move on and not learn anything!" It doesn't mean "oh, nothing I did had anything to do with that!" It doesn't mean "nothing I do matters, so let's just hang my shingle out there and see what happens."
But NOT recognizing the role of luck is frankly poisonous. Both to those who succeed and fail. There are a number of examples in indie games (which I will not name out of politeness, but I'm sure you can think of them) who had an awesome first game's performance, and they then got a big head and assumed anything they later did would be gold. So they put in a ton of time on something new, but weren't quite as "hungry" as they were when they were smaller. They were overconfident, instead, because they thought the past success was entirely based on their innate awesomeness. Then reality hit with the second game tanking, partly due to it being inferior based on their own actions, and partly due to the fact that THAT much luck that they had the first time doesn't always strike twice.
There are a bunch of other indies who have one smash hit, and then DON'T get a big head, and then continue releasing quality titles. Those titles then do quite well, but they aren't smash hits and aren't talked about on the level of the first one. Should they feel bad? Did they lose their edge? No. The first was great and lucky, and the others were just great or good.
My point, at core, is that people like to boil things down too simply. This is the problem with Malcolm Gladwell in a nutshell, really, but at least he brings up new points of view from the norm. The reality is that nothing is simple, everything has many many causes, and I'm not sure that anything is actually deterministic (at least on a level that we can appreciate). In other words, while you're insinuating that I attribute too much to luck, I actually think most people attribute too much to skill. I attribute an enormous amount to skill, but recognize the power of luck and its effect on things. Luck is basically an amplifier or a dampener, that's all it is, though.
- AIWars and The Last Federation are your two best sellers and the luckiest, but seem to have made their money in different ways.
I wasn't super clear, but I actually view Skyward Collapse as luckier than AI War. I'd say it is TLF that is the luckiest, then Skyward, then AI War. But the nature of their luck is entirely different, yes. The markets they released into were also VASTLY different, so the type of luck required was vastly different. Strategies that worked for generating success in 2009 are no longer valid. Because of the flood of new games, strategies that worked in 2013 are actually quickly becoming out of date, too.
- You cite AIWars as being lucky with advertisement, but it is arguably your best product to date. It is best seller, but it is also the best product in game engine, gameplay, and support. Its income has come from a long history of reliable consistent sales.
- In contrast we have other great seller, The Last Federation. I don't have access to sales data, but I have looked at forum data. Immediately after release, activity was reasonable. However, a few days after release, TotalBiscuit released a strongly favorable WTF is The Last Federation and forum activity increased by 300%. I'm guessing sales follow the same pattern. And it self perpetuated, with strong sales leading to further exposure and continued strong sales.
- Front page Steam advertising is amazing, as demonstrated by the impact of the community Steam Sale.
- In contrast to AIWars, tLF looks very pretty. The interface looks highly polished and the game makes great screenshots.
TLF was already going gangbusters before TB did his video. The forums went more active than sales did following his video. Sales for TLF were highest in its first 24 hours, and then dipped a bit, then went back up a bit when his video came out, then gradually fell down for the next couple of days. It was a crazy weekend.
And yes, graphics matter. The name of games also matters. I think that part of the problem with Bionic Dues -- and I have been first told this by a member of the press who I trust -- is the name. When you see something on Steam, all you see is the name and a very small icon. The Last Federation is a very strong name, while Bionic Dues is a weak one. The screenshots are much stronger on TLF than on Bionic, too, and certainly more than AI War.
Anyway, yes, success begets success. AI War's success was in large part due to my unrelenting campaign with the press and forums all over the internet to help educate anyone and everyone who was talking about the game. The big piece of luck for AI War was getting just the right press that let us get onto Steam at a time when almost nobody else did. TLF's luck was in its original weekend launch and when exactly that happened and how long it stayed on the front page. Beyond those things it was pretty much down to the games themselves and our marketing efforts (largely).
- The other Arcen titles are... not very polished looking. The art is functional, but it isn't outstanding. If you show the screenshots of the games to an naive viewer they will rate most of the other titles as being 5-15 years older than they actually are.
- I'd argue that the luck of tLF is really just good positive advertising with a very polished initial appearance. And with enough gameplay that players don't feel it necessary to bombard forums with any negative experiences they have.
- So is luck in advertisement the key to success for a small developer?
- Of course this brings up the conclusion, if initial exposure is king, then are games that focus on good long-term gameplay viable for small developers? For a statistical illustration, let's say 10% of games hit the "lucky" spot. If a dev spends a year making 1 game, they have a 10% chance of a single "lucky" success. If they spend a year making 12 games, they have a 72% chance of at least a single success.
- This is the pattern that mobile gaming is taking. And as a long-term gamer, this makes displeases me.
I think that there is no one simple model, and anyone who tries to condense it to such is going to wind up going down the one path. As noted above, the best generals adapt to circumstances that change. The market is changing every year, and the method for marketing a strategy game is vastly different from marketing a platformer or a casual game. Trying to over-generalize is one of the things that we as humans try to do too much, I think, and that's my own personal chief complaint against Gladwell. I enjoy his books because he has a contrarian point of view, but I don't believe everything he says is innately true. I think that Truth is an infinitely-faceted complex thing when it comes to understanding even something like why some games sell when others don't. I think you hit on some things, I think I've hit on some others, and so on. But I have other thoughts that I haven't typed up, and there are other thoughts that are things that I wonder but am not sure about.
If a dev spends a year making 1 game, then there is higher financial risk -- this much is indisputable. If they are smart about how they make it and how they market it, however, they can minimize the impact that negative luck can have. You can then hit a point where you have kind of a baseline level of success that you can probably count on. Whether or not it goes sky high or not is still somewhat a matter of luck, though.
The Edge of Tomorrow is doing pretty well, but not as well as expected from what I understand. That movie is amazing (I actually saw it twice, which I practically never do in theaters), and the critics seem to be overwhelmingly positive, too. So why is Godzilla beating it, despite ostensibly not being as good? I think some of it is probably marketing, sure -- Godzilla was way more visible. The names probably have something to do with it, too. Tom Cruise may be a factor. There are probably other things, too. But then there's also just plain luck in there somewhere.
- One complaint I get from others when I make them play Skyward Collapse is that the graphics are too small. What do you think about the viability of a graphic modded version with units and buildings being larger and cartoonier (and with animations) per titles like Final Fantasy Tactics?
- I get the same complaint when I try to get others to play Shattered Havens, the graphics are too small. What do you think about the viability of a graphic modded version with things being larger and cartoonier (and with animations) per titles like Secret of Mana? (You also might need a level minimap, but that's just details)
- How do you feel about opening up Arcen architecture to community modding, the costs and benefits? (Don't feel you have to patronize us here, many of the ideas proposed by formers are not remotely economically viable)
It wouldn't be economically viable for us on any of that, really. And opening it up to modding would either be just giving away the games for free, or selling them at a super low price. The graphics in Skyward are quite large, actually -- there is no problem for anyone in seeing them if they play on a screen resolution that is appropriate for their monitor size. The problem is that some folks have high-density screens and they then insist on playing our titles on those with the resolution at native. That makes everything TINY as heck. You're seeing things at like half the size of what they are supposed to be.
Shattered Haven is borderline too small for sure, though. But if anything, Skyward is on the very very large side.
- A common problem with indie developers is groupthink.
- Groupthink is a psychological phenomenon that occurs within a group of people, in which the desire for harmony or conformity in the group results in an irrational or dysfunctional decision-making outcome. Group members try to minimize conflict and reach a consensus decision without critical evaluation of alternative viewpoints, by actively suppressing dissenting viewpoints, and by isolating themselves from outside influences. - Wikipedia
- I would provide examples, but that would be rather political. Regardless, I'm certain that everyone has their own example of this.
- Of course, there is also the other extreme, the counter-flaws such as: indecision, too much compromise, and slow development.
- What strategies have you used to avoid the dangers of these two classes of issues (groupthink and counter-flaws)?
- How well do you believe they have worked?
Generally speaking, I'm a nonconformist by nature. I also come at this from a business software background, while most of the others come from AAA game development backgrounds. So my approaches and values are inherently different. A lot of times groupthink comes from the idea of "safety in numbers," or the fear of trying something new. I think Arcen, for better and for worse, is demonstrably not that.
I've used the financial buffer from AI War to do a lot of experimenting and learn a lot of things. A lot of those things did not pay off, some were major losses, but all were valuable learning experiences. Which sounds like a cliche thing to say, but what I mean is that I literally have information now that others do not have. But as of TLF, the learning period is basically over for me, in terms of wild experimenting on a high wire. We're still going to be experimental and unique, but in terms of where the worst areas of risk are, we've learned to recognize those pretty well. Not perfectly, but after each project we look back at it and decide what we did well and what we did poorly, both internally and externally and so forth. Bionic Dues in our evaluation was the closest thing to being perfect internally (and still far from it, but closer than any of the others), while AI War and Valley 1 and TLF were the best ones on the external side.
- With current product history, it seems pretty obvious that Arcen will continue to be successful and continue to grow (although, just not as quickly as initially expected). As it grows, do you think you'll move upwards in management scale and hire more programmers or would new people will be brought int to deal with management while you remain closer to project development?
At present, I don't plan to make any more hires. A lot of companies pursue kind of unlimited growth, but I don't feel that is needed or healthy. Having a highly skilled core team that can do things in a reasonably inexpensive way while at the same time compensating everyone in a way that a real business would (as opposed to how a lot of game companies do, where they treat employees like crap), is more my goal.
I'm sure that Arcen will likely grow again at some point, but there is no innate desire in me to see that happen right now, and literally no incentive to do so. The one thing that is an exception is that I would like to bring Cath back as a fulltime artist, as opposed to just being a contractor on a much more part-time basis like she is now. But Blue is just absolutely killing it on the Spectral Empire art at the moment, so...
In terms of my own role, yes, that has shifted and that's part of the success of TLF. Having more time for me to work on straight design and to then playtest and evaluate is important. Looking at the forest rather than the trees. Late in the TLF process, that broke down because of time pressure, and I think that the weaker parts of that game are largely due to those two factors -- the undue time pressure and my shift into a more active programming role instead of being able to focus on design. But there were a good 3 months there during the first part of TLF's development where I did literally no programming for Arcen. It was extremely odd.
For Spectral Empire, aside from some graphics programming and a few other bits here and there, I will not be doing much if any programming on that title. However, all the programming on TLF has been handed over to me at this point, so I'll be doing the programming and design on that while I then do design and general producer-role stuff on Spectral Empire.[/list]
Topic: Followup to last year's AI War postmortem (now discussing Bionic, TLF, etc). (Read 12178 times)