Crunch Is Not Weather. Someone Sets the Date.

The games industry talks about crunch as if it were weather, something that rolls in before a release and that nobody could have planned for.

I think that is wrong. Crunch, the stretch of extended overtime studios demand as a launch date closes in, is the predictable result of how projects are scheduled, staffed and sold. It will go on until the people who sign off on those decisions have something to lose when they make them.

I should say what I can and cannot prove. No survey I have read counts how many developers work 80-hour weeks, and none can tell you what overwork does to a marriage. What the surveys do show is a pattern, and the pattern is enough.

What the numbers say

The International Game Developers Association, a professional body for the industry, ran its Developer Satisfaction Survey with Western University in 2021. It collected 803 responses between February 15 and April 12 of that year. One-third of respondents said their jobs included crunch. Another 22% reported long or extended hours that they did not call crunch. ^[2]

That second figure deserves a second look. Some people work punishing hours and do not use the word, because the word has been made to sound like an event and not a condition.

Then comes repetition. Among respondents who had been through crunch, 58% of employees, 64% of freelancers and 63% of the self-employed said they had crunched more than twice in the previous two years. ^[2] The IGDA said the share reporting repeated crunch had risen from 35% in its 2019 survey. ^[2]

That is not a crisis that arrives at the end of a project and leaves. For many of the people who go through it, it comes back.

The concession

The best case for the industry is that things are improving, and the numbers give it some help. In 2019 the survey found 41% of respondents had crunch in their jobs. In 2021 it was one-third. ^[2] Not every studio crunches, and a survey of 803 people is a snapshot from one spring, not a census of a global workforce.

I grant all of that. I spent twenty years filing to deadlines, so I have no quarrel with deadlines. A date is a fair thing to ask of a team.

But look at what the improvement still leaves standing: a third of developers in crunch, and, among those who crunch, most doing it again and again. An industry that tolerates that has not solved a problem. It has made a problem slightly smaller.

The four dials

Here is the argument, and it is mine, not something the surveys measured. Every game project has four dials: the release date, the scope of what is promised, the number of people on the team, and the hours those people work.

The first three are set early, usually by people who talk to publishers, trailers and marketing calendars. The fourth is the only one that can still turn when the other three turn out to be wrong. When a date is announced to the public, a feature list is promised and a team is sized to a budget, hours are all that is left to give.

This is why I call crunch a design decision. Nobody sits down and decides to wreck anyone's health. They decide the date, the scope and the headcount, and the hours absorb the error. It is a system with a shock absorber built into it, and the shock absorber is a person.

The word the industry reaches for to make this palatable is passion. People who love games will give a lot for them, and executives know it. Passion is a reason for the hours to be offered. It is a poor reason for them to be requested.

Fear does the rest

Why do people say yes? Because of what the 2026 State of the Game Industry survey, run by the organisers of the Game Developers Conference (GDC), found about job security. It drew on more than 2,300 industry professionals, including developers, marketers, executives and investors.

Some 28% said they had been laid off in the past two years. Among US respondents the figure was 33%. Half said their current or most recent employer had conducted layoffs in the past 12 months. ^[1]

That survey did not measure overtime. I am not claiming it did. But put its findings next to the 2021 ones and ask a plain question. If more than one in four of your peers has lost a job in two years, how readily do you refuse the extra weekend?

An industry in which dismissal is common is an industry in which overtime is cheap. That is analysis, not a finding, but I would be surprised if any developer thought it was a stretch.

A game people love

The example I keep returning to is Disco Elysium, the role-playing game from the studio ZA/UM. The Verge's May 18, 2026 review of its spiritual successor, Zero Parades, describes the earlier game's development as driven by overtime, and recounts the studio's prolonged dispute between founding members and the company. ^[3]

I take that as the review's account, not as an established finding about any individual's hours or health. The review does not offer them. The dispute involves allegations I will not repeat as fact.

What I will say is that a game many people treat as a work of art, one with a voice and an obvious conscience, is described by a respected publication as having been made through overtime. Players love what they love. They are rarely shown what it cost, and the cost is seldom in the credits.

What would change it

The IGDA's executive director, Renee Gittens, said when the 2021 survey came out:

"No one should ever avoid or leave our industry because they feel unsupported or unwelcome."

• Renee Gittens, executive director, IGDA

That is a decent sentence and a modest one. It names the cost of the problem, which is people leaving, and it asks for support. What it does not name is a mechanism that would make support arrive.

Workers themselves seem to see one. In the GDC survey, 82% of US-based respondents supported unionisation of game-industry workers, and 62% of professionals said they were interested in joining a union. ^[1] The survey does not say what they expect a union to do, so I will say what I expect.

A union turns hours from a private sacrifice into a negotiated number. It moves the fourth dial out of the hands of the person least able to refuse. It does not abolish deadlines. It makes the date, the scope and the headcount someone's problem other than the developer's.

What to think

The lesson I draw is that an industry should not need its workers to prove personal devastation before it takes repeated crunch seriously. The surveys already show that a third of developers had it in 2021, that most who crunched did so more than twice in two years, and that layoffs have touched more than a quarter of the workforce since. That is the evidence. Anyone who waits for a sadder story is stalling.

So here is what I ask of you. The next time a trailer ends on a release date, remember that somebody signed for that date, and somebody else is going to pay for it.

The executives will call it content. The people who made it will know better.