I’m sure you know the drill by now: A big publisher does something destructive that we don’t like. Maybe they lay off some people. Maybe they force their teams to crunch to hit a deadline. Maybe they close an entire studio.
In response, we get a bunch of people saying, “No! Don’t do this Destructive Thing. The loss should come out of the CEO’s salary. The CEO is paid millions of dollars. Just take a few million bucks from them instead of punishing the peasants! The CEO won’t even miss it!”
All three of these destructive things are bad, but to limit the scope of this video let’s just focus on the problem of crunch. That is, the practice of having your development staff work painfully long hours in an effort to meet the desired ship date for a game.
The most recent version of this problem was when we found out that the team at CD Projekt RED was working crunch hours to meet the ship date on Cyberpunk 2077. A lot of fans and critics were outraged by this. People were saying, “CD Projekt RED promised not to do any crunch, and here they are breaking their promise! Why can’t the company just let the game ship late and take the losses out of the CEO’s ridiculous salary?”
And yeah. This answer is simple. And it seems just and fair. We can save all of these overworked developers from hardship, and all we have to do is take some money from a CEO who won’t even miss it. Why can’t we do this obvious thing? The thinking is that the executives make SO much money, that if we could just get them to give a little of it up we could solve all these other problems.
So… I’m going to try to convince you that this argument doesn’t make any financial or business sense, won’t solve the problem we’re worried about, and in fact will make the problem worse and create several new problems in the process.
I know this isn’t what people want to hear. In fact, tons of people have already stopped the video and hit the thumbs down, followed by the back button. It’s okay. I was young once too. I know the allure of simple answers. But if you’re willing to give me twenty minutes I’ll explain why things aren’t this simple, and offer a better argument to use against the executives.
To pull back the curtain a bit: When I sense I’m fighting an uphill battle against an audience that isn’t likely to be receptive to my message, I’ve found it helpful to try and make peace with them before I start taking swings at their sacred cow.
Here I was trying to do two things:
- Remind the audience I’m a real person and not some faceless narrator. In the video, a picture of me from 1994 pops up when I talk about being young.
- Allow that their position is understandable, even if I disagree with it. The idea is to disagree, while making it clear that I don’t think the reader / viewer is stupid. Saying “I used to have the same opinion, but then I changed my mind” is a great way to signal to the audience that you’re not holding them in contempt.
You could interpret my line as saying, “Hitting the thumbs down and leaving is an immature move”, which is fair enough. But you could also read it as “we disagree because I’m old and wise and you’re young and clueless”. So instead of an olive branch it might feel like I’m leading off with a slap in the face, which is the exact opposite of what I wanted to do. I didn’t really notice this problem until near the end of production. At that point, it would have been impractical to record an alternate line, get it to fit the flow of the original, and modify the visuals to fit the timing change.
So, I decided to live with it. We’ll see if it ends up biting me in the ass. Anyway, back to the article…
Now, I’m not going to argue that current executive salaries make sense. And I’m not going to try and convince you that crunch is a good thing. I’m just going to try to make the case that you can’t use the first problem to fix the second.
I’m Just a Peasant
Before you accuse me of being a greedy CEO type myself, I should point out that I’ve never been an executive of any kind and I’m definitely not rich. I was a programmer for most of my career, but these days I write for my Patrons. You can look at my Patreon and see exactly how much I make if you’re worried I’m some wealthy out-of-touch blueblood.
In my younger days I worked at a tech company and I personally knew several executive types. I was friends with a couple, and I met quite a few more during the dot-com bubble at the turn of the millennium. I got to see some weird shit in those days, including millions of dollars evaporating as the dot-com bust wrecked the industry.
The point is that I have first-hand exposure to things like project management, technology companies, and corporate finance. I don’t pretend to be an expert, but I have enough experience to explain why things work the way they do without resorting to simplistic narratives about CEOs being scheming cartoon villains.
So let’s start off by talking about…
To figure this out, we need to know how much it costs per month to work on a AAA game. I don’t mean the total budget. We have pretty reliable numbers for that. But the total budget includes a lot of one-time overhead costs like licensing tools and middleware, legal fees, and that kind of thing. What we want to know is: What would it cost us to let a game miss the planned ship date? If a game takes an extra three months of development, how much is that going to cost the company in additional payroll?
Unlike the total budget, we don’t have really solid numbers for this. I looked at several job-hunting sites to find some numbers to use for average salary, and added in the typical employment expenses you’d expect from technology jobs. The first thing to understand here is that salary is not the same as the cost to employ you. According to the CEO types I’ve met in my adventures, the cost of employing someone is about one and a half times their salary. If you pay someone $80k in cash, then you probably pay an additional $40k a year for liability coverage, unemployment insurance, health insurance, office space, supplies, and per-seat licensing costs for various tools.
Anyway, I ran the numbers to see how much it costs to employ someone per month in this industry, and it comes pretty close to $10k. Then, while fact-checking my results, I ran into an article from Jason Schreier, who asked numerous game studios this same question. Those studios offered the same $10k estimate. So I think we can have a lot of confidence in this number. On average, every member of a development team is going to cost the company $10k dollars every month.
Now we just need to figure out…
Teams are incredibly big these days. Rockstar reportedly has teams with hundreds of people. At the low end, it’s been years since I’ve heard anyone mention a AAA team with less than 50. I Googled around, and like salaries, the numbers are all over the place. Still, 75 seems to be a very reasonable estimate for “average AAA team size.”
For a 75 person team at roughly $10k per person per month, you spend $750,000 on labor every month to work on a game. There are a lot of other ongoing expenses to operating a large company in a major metropolitan area, and we haven’t figured in costs for management or contractors. We’re ignoring those costs for now because I don’t have good estimates to work with. Just keep in mind that we’re being very conservative and the real monthly cost is probably going to be significantly higher than that $750k.
So, can we solve the problem of missing a ship date by taking money from those evil predatory executives? To answer that question we need to know something about…
In 2019, joint CEO of CD Projekt RED Marcin Iwiǹski made zł5.7 million (Zloty)In the video, I anglicized the word to “zloty”. But since then I found out it’s pronounced more like “zwotty”.. That works out to 1.5 million USD. This discussion began by asserting that a CEO wouldn’t even miss a pay cut of a few million dollars, but even if we took EVERYTHING from Iwiǹski and made him work entirely for free for the entire year, his salary would be gone after just two additional months of development time for Cyberpunk 2077.
And that’s ignoring the fact that the game has already been delayed multiple times. It was originally slated for release in Holiday 2019, then slipped to March 2020, then to September of 2020, and now to November 2020. If Marcin is going to personally pay for those delays then he owes the company everything he’s made in the last 6 years.
People hear big numbers like “billion” and “hundred million” and the numbers seem so unfathomably large that it feels like the companies have limitless cash. But companies can and do go out of business. Even really big companies. THQ ran out of money and imploded. Sega ran out of money and had to cut their operations down to the bone, knocking them out of the console market and basically turning this publisher back into a game studio that survives by leveraging the properties they developed in their heyday.
If a publisher routinely lets their games run over budget and ship late, then eventually those losses will take their toll and make the company unprofitable. Yes, these companies seem invincible when you hear they make a billion dollars a year, but a big publisher can burn through a billion dollars pretty fast when they have dozens of games in simultaneous production.
Now, maybe you’ll argue I’m cherry-picking executives here. Sure, CEO Marcin Iwiǹski can’t personally keep the company afloat by giving up his salary, but… what about the big dogs?
The Big Dogs
So I’m sure you’ve heard the number thrown around, claiming that EA CEO Andrew Wilson gets $21 million in compensation. That makes for a good headline, but the truth is a little more complicated and a lot more boring. Wilson gets $16 million in stock and $5 million in cash. The thing is, stock is very different from cash and stock given to a CEO is even more different. You can’t just take that stock away from Andrew Wilson and give it to your employees as pay. To do that someone would have to dump the stock, and routinely dumping stock is not good for the company or the owners.
This is a big topic and I can’t do it justice here. If people are curious I can cover this in another video, but for now just humor me and accept that we can’t use Andrew Wilson’s stock to magically produce an extra $16 million dollars to spend on game development.
Now, what about that $5 million he gets in cash? That’s real money. Couldn’t we take some of that and use it to help out a game?
And yes, you could do that. The problem is that EA publishes a LOT of games. They published 10 games in 2019 alone. And that’s just their AAA releases, not counting various partnerships and mid-budget games. Gamedev is an uncertain business, and there’s no way all of those titles arrive on time. Which means Wilson never gets paid anything because there are always one or two delayed games, every year. Sometimes games get pushed back by months or even years. We would need tens of millions to cover that mess, and his salary is peanuts compared to that.
Now I have to confess something. We are still leaving out some major costs. Things are actually much worse than I led you to believe.
Time Keeps Slipping into the Future
If you’re 12 months from release and you realize that you’re going to need an extra month or two, then sure, just tack on a couple more months of production costs. An extra 1.5 million bucks isn’t a huge deal at this scale. As long as you can afford the extra time, you’re fine. But things get dangerous when you start talking about pushing back a game when you’re less than 6 months from release.
Let’s assume that – like so many titles – you’re aiming for a release in October or November so your game lands in time for the holidays. But now it’s late July and you realize you’re really going to need another two months. Maybe some of your big-name voice-actors are having scheduling conflicts and you can’t get them into the booth in time. Maybe your programmers are having trouble getting a handle on the new console generation and they need some more time to make sure the framerate is stable. Maybe the team suffered from feature creep because people kept sneaking in extra features and quests and worldbuilding details and now you’re a few weeks behind. Maybe one of your middleware suppliers is causing you headaches and you need to dump them and switch to something else. Maybe you realized at the last minute that you could make your game massively more appealing and unique by changing the art style into something different and eye-catching, and you need a little more time to make it happen.
Whatever. These things happen all the time. Engineering is hard, and doubly so when combined with artistic concerns.
So you need another month or two. So just push the game back to December, right?
When you finish a game and your team burns that gold master, that’s when things get really crazy. You need to send that master copy to an industrial facility so they can burn and label the millions of copies you need. You need to print and package all of those boxes. You need to arrange to have those boxed goods shipped to retailers all over the world. You already have a deal with those retailers that you’ll have your game on their shelves by such-and-such a date, and they’re not going to be happy that they reserved precious shelf space for your game and now that space will go to waste. And let’s not even get into the lengthy certification process required for console games.
You can’t just call these various suppliers up and move your job back a month, because they’re already contracted to print and ship other games for other companies on those dates.
This is why AAA games are never delayed by just two weeks. When a game slips at the last minute, it misses its original ship date by a few months because it takes time and money to reschedule an entire delivery chain like this.
But that’s not the biggest cost of missing your ship date. No, the biggest cost is…
Publishers spend between one and four times a game’s budget on marketing. So if you spend $60 million dollars to make a game, then you’ll spend an additional $60m to $240m on marketing. This marketing spending might sound pointless to those of us who follow industry news and upcoming releases, because we already know about the game.
The thing to remember is that we’re generally not the targets of these marketing campaigns. The vast majority of consumers are not like us. They don’t watch lengthy YouTube videos by random gaming idiots. They don’t follow the news, they’re not super keyed-in to the personalities of the various publishers, and they don’t know anything about a game until they see an advertisement for it. If you want to sell games to the masses, then you need to spend tens or even hundreds of millions of dollars to tell those people about the game and remind them to buy it.
I’m sure you’ve noticed that marketing campaigns usually begin slowly and ramp up gradually over time, before abruptly intensifying to a crescendo at release. They’re designed to get people curious, then get them excited, and then overwhelm them with reminders to buy the game and make them feel as if they need to get this game if they don’t want to miss out on an important cultural event.
The problem is that a lot of old-media advertising – and television advertising in particular – is incredibly inflexible. Prospective advertisers need to bid on spots months in advance. You work out the demographics of whoever you want your campaign to reach, then you get the data on which shows and time-slots will reach those people, and then you pay money to secure advertising in those slots. If your game slips, you can’t just call up the network and ask them to move your advertising time ahead a couple of months. You paid for that original time-slot, and if you want to advertise in some other slot then you need to go and bid on them just like everyone else.
So now your advertising campaign is designed to hit maximum saturation in November, but the game isn’t going to appear on retail shelves until (say) the following March. By the time March rolls around, all of those casually disconnected customers will have moved on, forgotten, or will have been captured by a more recent advertising campaign and spent their gaming dollars on something else. What do you do? Do you pay for ANOTHER massive $100 million marketing campaign, essentially doubling those gargantuan marketing costs? Or do you let the game launch with no additional marketing, and allow it to fade into obscurity in a crowded market? Either way, you’ll lose tens of millions of dollars.
Like Andrew Wilson, Activision-Blizzard CEO Bobby Kotick gets about $5 million in cash and then the rest of his comical $30 million compensation comes in the form of stock and stock options. But for the sake of argument, let’s pretend Kotick gets his entire pay as actual real folding cash. Even Bobby Kotick – the most outrageously overpaid executive in gaming – could not cover the losses of a game that slips in the last few months of development, even if he surrendered his entire salary.
Oh, and don’t forget about the problem where delaying one game by a few months will have it directly competing with ANOTHER one of your games from the same genre. I don’t have any idea how you’d do the math to figure out how much THAT will cost you, but it’ll be a lot more than you can get from the CEO.
Look, I’m not trying to convince you that executive salaries make sense. I could do an entire video on how these guys aren’t worth half the money they’re being paid. I’m just saying that the suggestion that executives should pay for a late game instead of having the team crunch is a nonsensical idea. It’s like realizing you’re ten thousand dollars in debt, so you cancel your $15 Netflix subscription. That’s a gesture, not a solution.
Maybe you’re still in favor of the idea. Sure, Kotick can’t pay for a late game himself, but shouldn’t he at least be penalized a bit? How about we take a few million bucks from him just to punish his failings as a leader? If his employees are going to work 80 hours a week, shouldn’t we make him a little uncomfortable to encourage him to do better next time?
I appreciate the sentiment, and I totally agree that leaders need to have better incentives to run their company properly. But I think this particular punishment wouldn’t work as intended.
Paying The Man to Make Things Worse
Think about the incentive structure this would create. Kotick is going to suddenly be very worried about late games. This incentive won’t make him care about his people more or encourage him to make better working conditions for his teams. It doesn’t encourage him to make better products. Quite the opposite: It will encourage him to avoid delays. In turn, this would push him to pursue extremely safe designs that aren’t likely to create problems or challenges. Well, okay… Activision pretty much already does that, but do we want to create even MORE incentives for him to behave this way? Do we want every gaming executive to be financially encouraged to act like Kotick all the time? Because in a world where the CEO is personally penalized when a game gets behind schedule and needs to crunch, complex, ambitious, unique games like Cyberpunk 2077 will be the first things that get cut in favor of more cookie-cutter titles.
Moreover, think about what happens when one of Kotick’s projects hit a snag. Kotick will have three choices:
1) Make his people crunch and he’ll lose millions of dollars of his personal income as a punishment, or…
2) Allow the game to ship late, and lose tens or hundreds of millions of dollars of the company’s money, or…
3) Ship the unfinished game on-time and blame the failure on the developers, possibly leading to the closure of their studio.
You’ll notice that we’ve just created an incentive for Kotick to always choose the most destructive option. That third option maximizes damage to consumers, the franchise, and the developers, while not punishing Kotick in any way. We’re essentially paying him to ruin games.
Sure, cut Kotick’s pay. You can even fire him. Into the Sun. I don’t care. But cutting executive salaries isn’t going to stop games from being late, and it’s less than one-tenth the money you need for covering problems like layoffs, late games, and studio closures. As big as executive salaries are, they’re a rounding error in the grand scheme of things. If you took Andrew Wilson’s entire salary and distributed it among EA’s 9,700 employees, it would amount to giving everyone a 0.7% raise. When you’ve got dozens of games in production and you’re spending billions on development and marketing, that five million you’re paying the boss just does not matter from an accounting standpoint, even if it personally offends you.
I’m an engineer, and I believe that the first step in solving a problem is correctly identifying the source of the problem. Attempting to take away money from executives is an emotional reaction. It might feel good, but it doesn’t help the employees or correct for management’s failures. Calling for a cut to executive salaries just reinforces the notion that consumers are economically illiterate and unreasonable, and that we should be ignored.
If you really want to hit these guys where it’ll hurt, ignore their five million dollar salaries and criticize the hundreds of millions in losses they inflict due to mismanagement, opportunity cost, lost sales, brain drain, bad press, and missed opportunities. The problem isn’t how much they make, it’s that they shouldn’t have the job in the first place.
 In the video, I anglicized the word to “zloty”. But since then I found out it’s pronounced more like “zwotty”.
Push the Button!
Scenes from Half-Life 2:Episode 2, showing Gordon Freeman being a jerk.
Resident Evil 4
Who is this imbecile and why is he wandering around Europe unsupervised?
Batman: Arkham City
A look back at one of my favorite games. The gameplay was stellar, but the underlying story was clumsy and oddly constructed.
I Was Wrong About Borderlands 3
I really thought one thing, but then something else. There's a bunch more to it, but you'll have to read the article.
PC Gaming Golden Age
It's not a legend. It was real. There was a time before DLC. Before DRM. Before crappy ports. It was glorious.