Video Game Budgets Are Broken
With games like Starfield touting $400M budgets, it's clear that AAA gaming is flirting with disaster.
Most people are just looking to have fun with a good game. A good game doesn’t have to be complicated, huge in scope, or graphically face melting to be a massive success critically and financially. You wouldn’t think that, though, given the current state of the video game market at large where apparently three A’s weren’t enough and so now we’ve got a fourth A being added to try and express just how much money a company is willing to throw at a single title, so here we are. Why are so many games taking so much time and money to make these days? Why do some games seem to accomplish more with less resources? Are time and money the only two resources that play a part in creating great games and if so, is the current formula sustainable? Are there any other ways to make these games happen? I’ve got a few ideas that might answer these questions and, like revisiting an expensive hamburger and fries, it begins with a roller coaster.
Listen, I swear I made a pact with myself that I’d try to avoid being the millionth person to use this example but like that one friend who raises their finger when Aragorn kicks the helmet, I’m doing it anyway. RollerCoaster Tycoon, which is considered by many to be one of the greatest games ever made, was programmed by one absolute mad lad in Assembly language across two years back in 1999 when frosted tips was just a boy band driven hair style and not a euphemism on Urban Dictionary. If you don’t have a point of reference for what that means (Assembly, not the spicy google search), it’s just this side of programming a game with ones and zeros. This is what allowed RCT to run like a dream on potato powered computers everywhere which was useful because this banger was getting handed out for free in @$#%$ cereal boxes before RCT2 took the reins and the collection sold 7 million units by 2004. You’d perhaps think that as the industry barreled forward with technology becoming far better, teams far larger, budgets far greater, and development times far longer, games would have gotten far better and played by far more people as a standard, right? Surely, even with rising player expectations and game complexity, games are getting better and game development more efficient across the board. Well…other than what happened to RollerCoaster Tycoon not long after Chris Sawyer handed it over being an example of why that’s not necessarily the case, there are more recent and pointed examples.
Not to beat a dead horse that could inexplicably walk vertically up a rockface, but compare 1999’s RCT2 to something like 2023’s Starfield which was developed by some 500 people over a full eight years on a reported $200M budget, or $400M including marketing. It doesn’t run well compared to its peers, it’s buggy albeit less so than other Bethesda games at launch and is likened to a stifled yawn by many who have played it. Including Game Pass players, Starfield’s most recently released player count sits around 13 million but they have yet to give out any sales numbers, which probably means the vast majority of players have checked it out via Game Pass. As much as Game Pass is a crazy value for consumers, player buy-in to try an otherwise $70 game is not the same as actually buying the game outright and so it’s safe to say that the reach of Starfield is at best about the same as RollerCoaster Tycoon 2 – unless we’re ready to compare RCT cereal box reach to Game Pass. Spending an increasingly long period of time and an increasingly large amount of money on games, even for seasoned developers, is clearly not a panacea for success. A video game does not need to be a $500M exercise in unfettered capitalism to be a success – even if it can theoretically provide a better shot.
Maybe that’s not a satisfying enough comparison. Perhaps comparing the scope and complexity of RollerCoaster Tycoon 2 to a game like Starfield or other modern blockbusters is a bridge too far. Well, how about World of Warcraft – that was a pretty complex game to make. WoW was developed in four years and change by 40 people before growing to about double that leading up to the release in 2004 on a budget of $64M which, if you dragged that dollar value through the two financial crises leading to the current year of 2024, would be roughly $110M. That’s about $40M less than the total budget for 2022’s The Callisto Protocol as a point of reference. The WoW development team over time is estimated to have hit numbers as high as 300 – for those curious. How about a more direct comparison in 2012’s Mass Effect 3? It cost $40M (about $55M today) and, perhaps criminally given how much content was cut to make a specific launch window, was completed in two years. I don’t have the developer numbers for ME3, but the original Mass Effect had 130 developers in 2007 so it’s safe to say that ME3 would have had no more than double that on the high end. For those doing some mental math at home, yes that does in fact mean that Bioware released Mass Effect 1, Mass Effect 2, Mass Effect 3, and all DLC in less time than Bethesda took to get Starfield out the door.
Jordan Middler
I think that we’ve established there’s something going on here when it comes to explaining current development times and costs that’s more than just game complexity or that making games is super hard because obviously, they are, but that’s not a new hurdle. I’m not going to pretend what I’m about to provide as answers encompass everything that’s going on because game development at the high end has a million moving parts. That being said, some of these parts have far more influence than others and I’m going to focus on a couple I think are most important.
If we look at the developers in the examples we used earlier and compare them to what development houses look like today it will highlight the first problem that I believe has contributed to longer development time, costs, and lower game quality – that being institutional knowledge. The Blizzard of 2024 is not the Blizzard of 2004, or even 2014. The folks that gave us the original Mass Effect trilogy are not who are currently fighting for their lives to make the next installment across nearly a decade. Big business has done a hell of a job eviscerating teams in such a cyclical nature that it makes it nearly impossible for a development house to build up invaluable institutional knowledge – all that knowledge and wisdom born of experience working together for long periods of time with consistent systems. It can’t be overstated how incredibly valuable this is. Developers staying together for many years spanning many projects are largely the reasons the decade roughly spanning 2002 to 2012 make up a true golden era in gaming. Perhaps the best example of leveraging institutional knowledge in gaming today is Nintendo who has benefited from it in ways that I’m sure any developer would dream of, but in the modern era this has become the exception and not the rule.
When you lose that kind of knowledge, and you are trying to hit a constantly moving target as you are when developing a game, you’re already behind the 8 ball and development time is going to naturally balloon – assuming you care enough to delay a buggy game, of course, and that leads us to the next reason game development is so expensive and time consuming: poor project management.
If you give two teams similar projects to complete with similar team size and skill, they should be able to complete their tasks in about the same amount of time with a similar budget. If one team completes their project faster, at a higher quality, and for less money then it usually points to poor project management in the other team. If Bioware can make one of, if not the greatest science fiction trilogy in gaming history with fewer people, for less money, and in less time than it took Bethesda to make Starfield I would be willing to bet that there was a project management problem. Even if you were to give Bethesda the two worst years of the pandemic for free, the math still ain’t mathin’. This is a similar story in a load of modern games. It used to be incredibly rare to hear about a game taking four years to complete – now it’s sort of the average. More and more games are taking six plus years to complete and in some cases are taking an entire console generation to get across the finish line and of course time is money, so costs go through the roof.
Of course, downward pressure from even further up the chain makes project management harder than it has to be. Suits in the C suite desperately chasing shareholder value without an original idea in their heads results in all manner of unreasonable expectations being placed on the developers throughout the development process. One of the first horrible trends we saw was multiplayer modes being shoehorned into every game imaginable whether it made sense or not because multiplayer games were making more money. Bioshock 2 famously was subject to this trend, though the main development team was blessed to have the multiplayer offloaded to a separate team. That evolved into all manner of things we get to endure today like microtransactions, games as a service, day one DLC, and more – again, whether it makes sense for the game or not. These sorts of pressures come in different forms as well, and one of the better examples is EA deciding that every developer under them would be forced to use their Frostbite engine – a decision that they would eventually walk back after disasters like Mass Effect Andromeda.
With unreasonable downward pressure from the suits, poor project management, and abandonment of institutional knowledge, games taking so much longer and costing so much more money despite more easily available and powerful tools at the developer’s disposal begins to make sense – if not in an unsatisfying manner. So, the question that remains is can this current model be sustained indefinitely? I think this is also a complicated question to answer but I have a couple of thoughts about it.
Aiden Crawford
The modern gaming landscape is a vastly different place than it was 20 years ago and the majority of the shifts have happened in the last 10 years. Gaming is bigger than ever, and has long dwarfed all other media combined both in dollars made and the size of the market and it’s no longer mostly guys between the ages of 18 and 35. In fact, 45 percent of gamers in the US are female and about a quarter of all gamers in the US are under the age of 18. With such a huge pool of people to sell your games to, you’d think it would be easier to sell your big AAAA title to enough people to become profitable but that’s not necessarily the case.
One of the biggest problems facing game developers, especially those spending so much time and money, is that there are now a handful of games representing a black hole in the market. This black hole sucks up a huge percentage of play time across the market. What makes up this black hole? Games like Fortnite, Minecraft, Roblox, League of Legends, and GTAV accounted for some 27% of playtime across all games. That leaves significantly less play time for all other games to battle over, and more games are being released now than ever before. In fact, there were over 14,000 games released on Steam in 2023 alone. Despite many of those games being shovelware, that’s a lot of competition even for big budget games.
This means that despite a much larger market than developers and publishers enjoyed decades ago, it can’t be assumed that even a great game is going to sell through 10 million copies. That hasn’t stopped publishers from ignoring that reality and placing unrealistic expectations on game sales, calling clear successes failures. Square Enix was guilty of this with the original Tomb Raider reboot back in 2013 – a game that sold a pretty incredible 8.5 million copies. Still, Square Enix claimed it underperformed. More recently, the very well received Dead Space remake sold 2 million copies – double that of the original release. Nevertheless, this didn’t meet EA’s expectations and subsequently resulted in planned follow-up projects getting put on hold.
That kind of expectation becomes particularly dangerous when coupled with how much time and money is being spent on major projects. A single flop is costly enough to tank a developer these days, and this has led to a homogenous blob of rehashed IPs and genres to minimize risk while gambling hundreds of millions of dollars on known properties and currently successful game types. If previously proven IP is a baseline requirement to get greenlit, but the expectations and bloat necessitate more money than many developers can afford, what’s the solution? Well, the big one we’ve all been watching is the consolidation of the industry. Whether or not this is a particularly good idea or good for consumers (though it’s probably not) is another discussion entirely, but it has been the way in which major entertainment companies have avoided the risk of developing new IP. Microsoft moving to buy Minecraft might now make even more sense if it didn’t already.
None of this sounds particularly sustainable, does it? A bigger market getting dominated by a small group of monolithic games, forcing everyone else to fight over a much smaller share of the pie while expectations and demands exacerbate existing issues pushing development time and costs through the roof doesn’t sound like something that can go on forever – and I don’t think it can. That’s not to say I think we’re going to experience a collapse in gaming like was seen in the 70s and early 80s, but I do think that something has to give. What that is exactly is hard to say, but I want to say that big budget titles can maybe look to smaller budget games for some ideas. It’s likely that four $50M games are going to have a better shot at profitability and sustainability than one $200M game, as a start.
It’s not all doom and gloom, though! There’s a load of games being released for a reason. Developing games is more accessible than it’s ever been, and that means we’ve been treated to some surprising hits that I think signal a new trend in what you might call a burgeoning single and double A market – not to mention a flourishing indie market. Games like Palworld and Helldivers 2 are prime examples of just that, and have more or less come out of nowhere doing huge numbers while big budget gambles on even well-known IPs and studios like Rocksteady’s Suicide Squad: Kill the Justice League scramble to make their money back. Of course, there’s also great success stories from developers breaking the mold and finding their own way to success like the incredible redemption arc of No Man’s Sky, and the astounding success of Baldur’s Gate 3.
Part of the problem of course is that once in a while these gambles pay off and everyone experiences that amazing high of a big budget game delivering in spades. Unfortunately, more often than not we’re experiencing the lows – but hey, look at it this way, if everything was always amazing, nothing would be amazing. The highs feel great because the lows preceded them and just like any golden era you’ve got to endure what comes before. Things might be a little rough at the moment, but give it a little bit more time and I think this whole roller coaster thing might pan out.