Kate Edwards likes to do cosplay (costume play) as characters like Thor, a valkyrie, Indiana Jones, and Brienne of Tarth from Game of Thrones. That may give her the fortitude to handle her next big task as executive director of the International Game Developers Association. Last week, during the Game Developers Conference, Edwards announced the IGDA is going further in measuring the “crunch time” practices, or mandatory uncompensated overtime required to finish projects, at major game companies.
The group has already been doing those surveys for the past couple of years. Now, it will gather more precise data from employees of game companies, and it will report on the companies with the best crunch time practices later this year. And next year, the IGDA will reward the companies that are doing the best. But if companies with bad reputations refuse to change, the IGDA might publicly report companies that continue to force employees to work uncompensated crunch time.
In an interview with GamesBeat, Edwards said that surveys for the past two years show at least 37 percent of game developers say they are not compensated for crunch time, when they work long hours in a day or week to finish a game.
“We know it is a persistent problem,” she told me. “Now, what do we do about it? We’d prefer to highlight the companies that are doing really well. If there’s an exemplary company, we will highlight them. If we found an example that is grievous, we’ll probably highlight that as well.”
Uncompensated crunch time has been a challenge in the industry for the longest time, and it was drawn into the open by the “EA Spouse” controversy, when developer Erin Hoffman wrote an anonymous screed criticizing Electronic Arts for requiring employees to work long hours in the final process of shipping games — and then moving them on to new crunch time projects as soon as they finished. The 2004 episode drew a lot of attention to crunch time, but the issue has almost been forgotten again.
Edwards said the IGDA board of directors will privately raise their concerns with leaders at companies practicing uncompensated crunch. If those companies do not change their practices, then the IGDA may take more action, including publicly speaking out about those firms, Edwards said. Here’s an edited transcript of our conversation.
GamesBeat: It looks like you’re starting gradually, but it seems like you have to get your data absolutely right now. It can’t be vague. If you associate it with named companies, it has to be very strong data.
Kate Edwards: This will be the third year of the Developer Satisfaction Survey. We’re launching today. We have two years of data, which is built on several years of older quality of life data we’ve been doing since 2004. Looking back over the stretch, it’s been obvious from the data we’ve collected — we all know crunch is an issue.
We also know compensation can be an issue around crunch. Seeing the numbers in the last couple years has given us an indication. Between 2014 and 2015 we saw 38 and 37 percent of developers, respectively, stating they don’t get compensation for crunch time. That’s ringing an alarm. That’s a huge percentage. More than a third of developers get no compensation for something that’s common in our industry.
What we’re trying to do now with this initiative, which for now we’re calling the Crunch Comp Initiative, is the next step. What do we do? We see data that tells us this is a persistent problem. We’ve known anecdotally for many years that it’s a problem. We’ve collected data that more specifically shows it’s a problem. So what do we do?
The approach we’ve decided to take is to extend beyond that. We’ve collected data from developers through our survey, but if we can get developers to start offering their own experiences through a mechanism that allows them to report in a style like Glassdoor, that kind of method, they can show us. “Here was my experience. Here’s how I would rate the company on this issue.” Hopefully, we’ll get a significant accumulation and be able to see across the board.
It’s going to take time to ramp this up and get feedback, but as we get that feedback, we’d prefer to highlight the companies we see that are doing well and say, “Look, this company is getting good feedback on this issue.” The board does a positive impact award every year. Maybe if there’s an exemplary company, we’ll say, “Let’s specifically highlight this company for what they’ve done.”
Conversely, though, if we found an example that’s particularly bad or grievous, we would probably talk about that as well. If we’re seeing, on a scale of 1 to 100 or whatever, they’re rated at five by a lot of developers, that’s obviously an issue.
GamesBeat: You do it as a survey now. Would you have to change it at all to have that Glassdoor capability, where people can approach you and say what they think?
Edwards: We’re planning to continue the survey every year. That’s still a benchmark we use. We collect a lot of data in there that would not be collected elsewhere. We’re looking into partnering with a third party like Glassdoor or one of the others out there. They have the mechanism already.
If you look on Indeed.com, for example, they have subcategories for the workplace. They have a work-life balance category for companies. If we can work with them to get a couple additional categories, specific ones about issues related to the game industry, and partner on those categories … ultimately I’d like to see, if we can get this set up, we could roll out all kinds of issues. How are they doing on diversity from an employee perspective?
GamesBeat: Your data is good enough to point people out publicly.
Edwards: Now that we’re in this era where that kind of mechanism works and it’s something that people are used to, we see it across the board with Yelp and TripAdvisor and all these methods. There are imperfections in all these methods we have to be aware of. Obviously, all of it’s going to have to be vetted. We’ll have to look and see what people are saying, so someone’s not going on there and randomly commenting. Which is why, again, we’d like to use a method from a third party. They’re already doing that.
GamesBeat: Have you found that some people already feel like they can’t be honest because they don’t trust the anonymity of it?
Edwards: Yes. That’s part of it. You look at some of the examples out there, like Glassdoor, there is anonymity there. You decide what level of anonymity you want. I’ve seen Glassdoor reviews where you can pretty obviously pick out who it might have been or narrow it down very closely. Maybe the person doesn’t care, but they have the ability to dial in and out of anonymity.
Something like TripAdvisor, you can set up an anonymous account with a fake name or something. A lot of the stuff I see there is people logging in with Facebook, though. They have no qualms about tying their name to reviews. Typically, when people do that, they’re giving good reviews.
GamesBeat: What do you already know as opposed to what’s going to be in the next survey? You already know that 37 and 38 percent are a problem. Do you know other things associated with crunch time?
Edwards: We’re not just tracking compensation, of course, but just the occurrence of crunch. When we look at the data between 2004 and 2014 — we didn’t incorporate 2015 data into this particular segment — the occurrences of crunch are declining. It’s not quite as often, and when it does occur, it’s not quite as long. But it’s still happening. I’m expecting we might see that continued trend. Looking at it on a yearly basis, maybe it’s not quite as good of a snapshot. I don’t think it’ll change that radically. But we’ll have to see.
We always ask — and I don’t expect the answer to change — about the cause of crunch. It’s pretty consistent. Poor project management. Inexperienced managers. Things like feature creep. Those are usually the big ones.
GamesBeat: I always thought that mobile was going to be different. I wonder about the categories now, whether you’d be able to identify problems in certain categories. It seems like mobile would be worse for crunch time because the updates happen all the time in free-to-play games. They operate 24 hours. They’re always connected. It’s an always-on environment.
Edwards: The thing that helps with mobile, though, when it’s always on, there’s a level of predictability. It allows you to predict staffing and content flow and work flow better than when you have something like a massive RPG that you’re trying to do over the course of years.
A lot of the mobile stuff tends to have better control over feature creep for example. It’s usually a specific type of game. There isn’t often room for feature creep, depending, although you might have cases of something like adding a whole new set of power-ups you can buy. But even there, when I’ve talked to people in the industry, they have an understanding of what work that entails. It’s not as if at the eleventh hour someone says, “We need another power-up set.”
The potential is there for any game project to feature creep or just run away with people. But in mobile, it comes down to a company culture issue. If they want people to be constantly crunching on their content that will never end, that’s a pretty dismal future for most developers. I don’t think it would be sustainable from a workplace standpoint. With something like a large triple-A title where you expect a light at the end of the tunnel, even if the light keeps creeping further away because of schedule changes or features.
The problem I’ve seen in the triple-A space with crunch is partially just because it’s part of the creative process. Whether or not they call it that, people crunch in film, in television. Writers crunch to finish that last chapter the editor is yelling for. It’s the nature of creative work, that it’s never really done. The only reason a film releases is because someone sets a date and backs up the schedule with marketing and everything else that has to happen.
It’s why we have director’s cuts. “Well, that wasn’t exactly what I wanted. Here’s this version that shows you what I would have done.” With games, we can kind of do that. We just release DLC or do a patch.
GamesBeat: Although it seems like DLC’s made it so you lost whatever downtime you used to get when you finished a game. Now, a game’s never finished.
Edwards: Exactly. That’s one thing that brings it closer to the mobile model. With mobile, games tend to be more confined in their scope and more predictable in how their content rolls out. The DLC model being applied to a triple-A console game is almost like you’re trying to apply the mobile model and expecting people to sustain that same workload — especially if a game becomes really popular. “Oh, we need more DLC.” That’s not sustainable.
GamesBeat: The Zynga team on FarmVille got more relief when they opened an India office. They didn’t have to be on 24/7 anymore. They could hand it off to a team on the other side of the world. It seems like that’s one way to relieve the time pressure.
Edwards: Absolutely. It’s just making sure you staff and plan for it. We’re seeing a lot of companies where .. as we see crunch time diminish a bit in the industry, that’s part of a maturation process. I hope it’s a maturation of management at game companies, understanding that this model had its time in our industry’s histories, but … as industries mature, like we’ve seen in the general IT sector, a lot of companies are pushing their employees to take better care of themselves. Even if it means being in something like a company-town environment. “Go use the gym at our office.” They lavish their employees with ways to take care of themselves. We’re seeing some game companies do that too. But I’d say not on the scale that we see outside the game industry.
GamesBeat: They’ll also do it just to keep employees at work all the time.
Edwards: No, certainly. Developers think about the ulterior motive. “Why do they give me this incredibly good food for dinner every night? Oh, they want me to stay here and eat it.” It can be a trap. I understand why companies do it. But that kind of model, where it becomes something we’re used to … during the DICE awards, some of the winners on stage accepting awards were talking about how they missed … their kids. It struck me as I was listening. I’d love to ask them as a parent — OK, 10 years from now, what would you rather have? The award or the time with your children? I can guarantee what the answer’s going to be.
It’s about the value of your time. Trying to get a perspective on that in our industry is something we still struggle with. We’re passionate about making games. We love making games. Developers want to make games indefinitely. When we ask them these questions, they’re very clear about that. But you don’t want to do it at the expense of everything else in your life.
GamesBeat: You’re almost trying to tell people this for their own good. Sometimes they’re not realizing that crunch has a cost for them.
Edwards: It does have a cost. It has a long-term effect. It takes a real toll on health. If you work 70 hours a week, that’ll affect your health without a doubt. We understand that there are reasons it happens, but there are also many good reasons it doesn’t have to happen or at least not to the degree it does.
Better project management is part of that. But one of the things that will feed that, especially in the game industry, is having a better handle on the creative scope of what you’re making. That’s a harder task. When you set out to create whatever massive project it might be — I understand. I’ve worked on many games. I get how the energy of the creative process is so important that it changes the scope and changes what you have in mind. I’ve been involved with projects where that last-minute inspiration hits. It’s the weekend before cert and somebody says, “Oh my God, we need to do this. It’ll change everything. Let’s stay for a 72-hour stretch and get this done.”
You have to weigh it. Is the player going to care that much about the time you’re spending? Or is it something you can do in another version or in DLC later? Is it really something you have to put yourself out there for? I don’t know if game companies tend to be as judicious as they need to be in thinking about the overall effect of that. It’s hard. They know their launch must be successful. But in my view, if you’re not already sure about that in the creative vision you started working on, maybe you should have had a different creative vision.
GamesBeat: Do you get a sense of whether good managers are the ones who restrain their employees from doing crunch? Do employees want to crunch without realizing the cost to it? Or are people being forced to crunch when they’d rather not?
Edwards: It’s a mix of both. Long-standing tradition in our industry has been one where it’s been more about forced crunch. Sadly, I’ve heard this many times from different developers. They say, “I can’t complain about it, because the standard response is just, ‘Go ahead and go. We’ve got 20 people to take your place.’” Which instantly ascribes a lack of any value to the specific employee in the company. You’re just a cog and easily replaced. That’s not a great morale-builder to begin with.
When I give talks, sometimes I have a slide that’s a frame from Bridge on the River Kwai. In that movie, the Japanese general kept telling the POWs, “Be happy in your work.” Sometimes we get that attitude coming in the game industry. Just be happy. You’re lucky to be here. We want to see that go away. People should have a choice.
A lot of it comes down to, when you’re hiring into a company, if you’re interested in working for a company, companies should be explicit about how much they typically crunch. Be upfront about it. “We look back on our work data over the last few years, and we generally crunch a third of the time.” Disclose an idea of what crunch is like, and tell people about whether or not they’ll be compensated for it. Then, developers can at least make an informed choice.
When it comes down to individual managers, it’s important for them to be monitoring both the physical and mental health of their employees. “Hey, you’ve been here three days straight. Maybe you should take a break?” At the same time, there’s a certain level of creative freedom you want to allow. If you have someone who says, “I don’t need to go home. I don’t have a family to see. I want to work another 12 hours on this because my train of thought on this issue will be complete, and I’ll get this done. Then, I can walk away feeling good about it.”
There’s a balance. To some degree, if somebody is that into a particular task — even then, managers need to step up and draw a line. “You’ve been working on that for three days, and you still don’t have an end in sight. Maybe you should clear your head and come back to it.”
GamesBeat: Is there already any established practice as far as compensated crunch at particular companies?
Edwards: I don’t know specifics. It’s hard to get that information from companies. We’re looking into it, though. If you look in the DSS data, we have people reporting that they get certain bonus structures. We ask all kinds of questions around bonuses, around whether or not the bonuses are performance based or project based and so on. The 37 and 38 percent we’re identifying are people who say, “We don’t get any of that. We just do it.”
GamesBeat: For the companies that show repeated problems, how are your board members going to communicate with them? What are the steps you’d take before going public?
Edwards: First and foremost, board members will approach the company directly. “We’ve been noticing a lot of people reporting issues around this problem in your company.” We’ll just sit down and have a conversational approach. We’ll get the company’s perspective and understand their point of view on the issue. We’ll see if there’s any acknowledgement or understanding that it’s an issue. If there is, we’ll try to discern if there’s a plan in place to fix this and change the way business is done. Or, is this based on a specific anomaly, a specific project? Maybe it’s not normal, but it spiked because of something in particular.
Basically, we’ll just do some discovery and learn more about it. The process from there is ultimately going to be up to the board to decide as far as how vocal they want to be. Our perspective is that if we see an example of a company where we’re able to determine with certainty that it’s a problem, they know about it; they’re not willing to do anything about it; they’re not taking steps to fix it; then, it’s likely at that point that the IGDA will speak up publicly. “Hey, developers, if you’re looking for a job, here’s an example of a space that you probably don’t want to consider. After our repeated engagement with them and after collecting data on the situation there, it’s probably not the best situation.”
GamesBeat: Companies won’t be happy about that. Have you felt any backward pressure yet on that topic?
Edwards: Some companies are uncomfortable with the topic because they recognize that it’s an issue. A lot of what we’re dealing with, too, is a certain degree of legacy. We’ve been on this industry train that just keeps chugging along at a very fast pace. “We got the first game done. We have to get the sequel out. We have to get the DLC done.” It keeps moving, and there’s no chance to step back for a moment and say, “What are we really doing here? What’s our workplace like?”
I don’t know to what degree there’s any deep introspection at the company level. I know some companies do that. A lot of companies say they do. I’m not saying I disbelieve that, necessarily, but I don’t know about the degree to which they take action when they hear from their employees about issues that come up.
That’s why, when it comes to the method we want to use, we want to do a quiet engagement and give the company a chance to talk with us. Along with this initiative, we’re going to take some of the data we have and stuff we’ve done in the past around best practices and how you can avoid crunch and create an environment that works. We’ll share with the companies and see if they want to employ those ideas.
I don’t think it’s going to be swift. Obviously, we want to give companies a chance to react, especially if they’re open-minded. But I’m sure there are some that won’t be. In those cases, the last resort we’d want to do is basically outing a company as a potentially problematic workplace for developers. That’s an absolute last resort.
GamesBeat: Do you have a timetable for when you’ll get to finished data?
Edwards: We probably will do some kind of listing, like what we do out of the developer survey right now. We ask developers which companies they’d most like to work for from a general perception standpoint. Valve is always at the top of the list. We’ll do something like that — which companies were the highest ranked every year in terms of how they deal with this specific issue. I’m hoping that we can crank this up and get something open by Q3 of this year. That’s when we’ll have something we can invite developers to start giving input on.