A while back I was asked to give a deposition in an industry case that was going to trial. The defendant's attorney wanted to send over a crew to video tape my opinion on "average royalties" for independent game developers. I asked him if the developers themselves funded the project, or if they'd received publisher advances. He answered that the devs themselves "covered the expense." I countered that he didn't answer my question; I already assumed the developer covered the expense to develop the game, as that's how the vast majority of contracts are written. He caught on quickly, "Ah, I see what you mean." He then properly answered my question, revealing that the developers did not get an advance from the publisher, but that they paid for the development out of their own pocket. "In this case," I said, "it's not unreasonable for them to get a 30- to 50-percent ballpark royalty percentage. Of wholesale, of course." His reply was typical of what I've heard from other devs in the industry: "Wow, that high!"
I quickly added, "Or even higher."
This little story leads into my four-point topic on royalties, and how dev studios have bought into a royalty game that has publishers giggling with glee during this and every jolly holiday shopping season.
 Independent developers, especially new studios, generally do not know what level of royalties they should be shooting for. Back in the early 90's, when Apogee (pre-3D Realms) began working with publishers, we, too, didn't have a clue what sort of royalty was fair. We were lucky at the time to work with a smaller publisher who didn't have much clout and desperately needed product, and through hard ball negotiations we secured a nice-for-back-then royalty of 25% of wholesale minus COGs (see note below about royalty calculations). And this was on top of getting an advance big enough to cover the funding of the game. But things were easier back then, because there were many more publishers, all ravenous for games to fill their needs. Bidding wars between publishers was common practice. For example, the bidding for Duke Nukem 3D in 1994 included FormGen (the eventual winner), GT Interactive, and Activision. Even though we were fairly unknown back then, the CEO's of all three companies (Jim Perkins of FormGen, Ron Chaimowitz of GT, and Bobby Kotick of Activision) flew to Dallas to woo us. It's fun being in demand. (Sierra, too, sent some big-wigs over with a full company buy-out offer, but their piteous $2 million offer found them quickly ushered toward the exit door.)
[Sidebar: In late 1996, GT Interactive bought FormGen for $17 million in stock in order to get the publishing rights to Duke Nukem 3D after it had been released and proven a hit. Only two years earlier FormGen paid $250,000 in advances to original get the game, beating GT Interactive in the game's bidding war by $30,000. Being cheap can sometimes backfire.]
Nowadays, publishers run the show for the most part, with developers verging on commodity status. This means bidding wars are rare, dev studios have no leverage, and except for the most proven independent studios, they're just happy to find a deal...any deal.
Maybe only one or two dozen top-tier studios can get a 50% royalty or higher. To get this high they must either own their own proven IP, and/or be able to self-fund the development cost of their project. The highest royalty 3D Realms has executed is 70%, for a project in which we owned the license, and self-funded. That's the highest I've heard of in the industry. (If anyone knows of one that's higher, I'd love to hear the details.)
The way the industry works now, unless a studio can self-fund, it's near impossible to maintain ownership of an original IP. Publishers practically demand IP ownership in case it turns into a hit. This was not the rule in the 90's, except for when things started to change in the very late part of that decade. Remedy and 3D Realms recently demonstrated the value of owning an IP when we sold the Max Payne IP to Take-Two for over $40 million in early 2002, this after making just the first game.
The Holy Grail for any independent studio is to create and maintain ownership of an original brand, or IP. This gives a studio both leverage and financial strength to remain independent, and pursue future original projects of their liking. If you think about many of the strongest independent studios, they're often ones that own their own IP: Id (Wolf, DOOM, Quake), Epic (Unreal), Valve (Half-Life), Lionhead (Black & White, and Fable, BC and The Movies in production). Bioware is now working their own IP, as are a few other studios who are trying to establish their own IP to gain secure financial independence (not having to rely on publisher funding).
Then we have studios that do not have leverage, who often settle for percentages as low as 10- to 15-percent. This lower percentage not only means that these studios get less of the revenue pie, but it hurts them in another important way, because...
 Developers ALWAYS pay the development cost of their projects, regardless of who owns the IP!!!
Let this sink in for a moment.
You see, whether the game’s funding comes out of the dev studio’s own pocket, or the publisher advances the funding to the studio, either way the studio pays for the game’s development cost. A publisher's advance is a loan, and the loan gets paid back from the dev studio’s royalty stream.
Publishers argue that they’re still taking on loads of risk in case the game doesn’t sell well enough for them to recoup their advance. But, the truth is that publishers can break even on a game well before the dev studio has repaid the advance. The payback of any particular game’s advance by the developer is not a reliable measure of a game’s profitability.
And this plows head first into my third point...
 The way publishers recoup their advances is a long perpetuated industry scam. That’s because advances should not be repaid from the studio's royalty stream. This way of repaying advances severely punishes studios who are working for a lower royalty. Imagine a studio getting just a 1% royalty (unrealistic, but I'm making a point). Let’s say this studio got a $2.5 million advance to make a game. This game would need to sell a GTA-like 8.3 million copies before the studio saw a dime of royalties. (This assumes the publisher makes $30 per game, after COGs.) By the time this studio saw it’s first royalty check, the publisher has gotten gross revenues of $250 million, meaning they’re at least $220 million in the green!
Although this is an extreme case with an unrealistically low royalty percentage, there's an unfair lopsidedness that applies to royalties that are common in the industry, with studios not seeing a royalty check even though the publisher has made millions in pure profits.
The point I’m making is that there’s no connection between a publisher’s profit on a game and the current standard method currently used by publishers to recoup their advances. Publishers have gotten away with installing a recoupment system that on the surface makes sense, but doesn't make sense when you look closer.
This is why most independent studios live from game to game, and milestone paycheck to milestone paycheck. Rarely can a studio accumulate enough in the bank to break free from this deep rut.
If this system is a farce, is there one that makes sense?
A system that makes sense is one that combines the game’s development cost and the game’s marketing, inventory and distribution cost. Marketing, inventory, and distribution are costs handled by publishers, as these are a publisher’s primary functions. When these costs are all combined, then you have the true cost of developing and publishing a game. Whatever this cost is, it should be deducted from the top of all game earnings, and only then does anyone – publisher and developer alike – make any profit.
The bottom-line for those who understand the lingo: All reasonable costs should go into the COGs bucket. Once COGs is paid off, both sides start making money. This is fair because, although the publisher is risking their money and also paying for marketing, inventory and distribution, the developer is paying for the project's development cost and they're usually getting only a small piece of the royalty pie.
There are plenty of others in the industry that share this opinion. For another respected opinion I direct you to attorney Jim Charne, formally President of the Academy of Interactive Arts and Sciences, who recently wrote about this topic in his monthly IGDA column, Famous Last Words.
And there, folks, you have The Great Royalties Game. Play it at your own risk.
[Sidebar: Royalties -- How they're calculated in the PC industry: Generally, developers sign deals whereby royalties are based upon a game's wholesale price, which is the amount that the publisher gets from selling the game to retailers, renters and distributors. (Retailers mark the price up another 10 to 20 percent, which is the price the public pays.) But before publishers pay a royalty to developers, they first deduct "COGs" (Cost of Goods) which is the typically the cost of the game's retail packaging, such as the CD, the manual and the box itself, along with any goodies that might be included within, like the mouse pad that was included in the PC version of the original Max Payne. COGs usually ranges from $2.00 to $3.50. So, a typical deal might have the developer getting a 20% royalty, the game wholesales for $30, and COGs is $3 even. Under this scenario, the developer will get $5.40 per unit, BUT only after the publisher has recovered the entire cost of development from the studio's royalty stream, which will likely take several hundred thousand units. Note: For the console industry, COGs will also include the console manufacturer's (Sony, Microsoft, Nintendo) license fee, which can be as high as $9 to $10.]