A dev's guide to sustainably supporting live games without crunching

The game industry has a burnout problem.

Developers have complained for decades about the toll this business takes on their personal lives. People lose sleep, miss important moments, and sacrifice relationships in order to work in games, in part because it can feel so satisfying to stay late with the team and hit that deadline together -- especially if it’s a project you really care about.

Now that so many games are designed to be played and updated indefinitely, those late nights seem as likely to happen after you ship your game as before. Big, regular content updates keep players engaged and talking about your game, which can put pressure on a studio to overwork people in order to ship a game -- then do it again down the line to deliver updates for said game. 

We saw a version of that story play out earlier this year, when it came to light that some Epic Games staffers were complaining about having to work in near-constant crunch mode in order to keep Fortnite updates coming at a rapid clip. In the wake of that story making the rounds, the industry as a whole seemed to get a bit more proactive about taking a stance on crunch; Apex Legends dev Respawn Entertainment even made a show of saying it wouldn’t make weekly updates to its popular F2P battle royale game in order to maintain the dev team’s quality of life.

“That's the closest we've come to seeing anybody even make a statement like that in that space,” says veteran game producer Keith Fuller (read our full Q&A with him here). “Now it seems as though they're making good on those words, that they're executing on that, they aren't putting out daily deliveries or whatever Fortnite's doing. But then also, the next thing that people do is, they put up the monthly active users, daily active users charts, for those games. And okay well, great, it's wonderful you're doing that for your team on Apex. Let's compare your active users vs. those of Fortnite. Oh, look. Yours are horrendously cyclical and trending downwards, while Fortnite is skyrocketing off the planet. From a business standpoint, which one should we be doing?”

Fortnite's latest season update, its tenth in two years, introduces pilotable mechs, prestige missions, and more 

Choosing to do what's best for the business instead of what's best for your employees is the core of the burnout problem, and it's one deeply rooted in the way games are made. Fuller cites Activision Blizzard's decision to lay off ~800 people earlier this year, while almost simultaneously reporting historic earnings, as a key example of how the business of games prioritizes profits over people.

"The problem with which we're faced is that Bobby Kotick is doing a good job. He's doing what he's supposed to do. He's pleasing the shareholders, he's keeping the stock price up," said Fuller. "The difficulty therefore is in the system he's serving. By virtue of the fact that capitalism is what it is, you can absolutely be put in a position where the best job you can do is exploit the people beneath you."

So what can devs do to push back against such exploitation, especially when it comes to supporting live games?

To find out, Gamasutra chatted with a few experienced producers from around the industry about how teams can best avoid overworking on live games, and found two pieces of advice recurred again and again: establish (or find) a company culture that abhors crunch, and master flexible scheduling that handles the realities of live game development.

Unhealthy culture and poor scheduling lead to overworked devs

“The key is a studio culture that views crunch as a production failure at all levels, and a development process and pipeline built to avoid crunch at all costs. You can’t have one without the other and expect to actually avoid crunch,” says Zynga lead producer Tami Sigmund (read our full Q&A with her here).

“There’s a common belief that ‘live ops = unpredictability’, and while that’s true for some aspects of development, I believe that running a live game is a ton more predictable than a game in development,” she continues. “The trick is setting a cadence schedule that you can reasonably commit to, that has enough wiggle room within to adapt to change as necessary. For the games I work on, we commit to releasing a new feature  to our players every two weeks. This is an extremely predictable and regular cadence, so the team that is working on those features has a rigorous and tight pipeline that’s well-staffed to support that ongoing content without requiring extra hours from anyone on the team.“

Sigmund also shared a list of practical tips for creating an effective, crunch-free live ops production schedule (you can read all of them in her full Q&A) that include rotating your on-call team, establishing robust reporting systems (so you don’t make the same mistake twice), and budgeting extra time into your schedule.

“Build in at least 20 percent extra time into your schedule to allow for pivots, but don’t eat up that time if you don’t have to,” she cautioned. “Instead of stretching work to fill that schedule, build up a bank of extra lead time so that you can expend it on things like passion projects that the dev team wants to work on, going above and beyond on particular features, or learning and development time for the team.”

Veteran producer Dante Falcone made a similar suggestion (read our full Q&A with him here), going so far as to say it’s a good idea to budget in an additional 30 percent extra time and have all deliverables ready at least a week before they’re due.

“Whenever I schedule anything I make sure there is 30 percent slack time for unexpected stuff - because there will always be things you can't predict - ranging from bugs and late deliveries to all-hands-needed live issues,” said Falcone. “Along with this, I stress a regular timed release cycle, such as delivering content to live every few weeks - when you do this over time you will get better at knowing what you can and cannot do in that cycle, as well as get better at executing.”

Falcone has experience working in MMORPGs at studios like Turbine and ZeniMax Online, and he says the best live-service games he’s worked on, in terms of quality of life for the team, had expansion content ready to go months before it was due. 

That gave the team time to test and refine it, as well as suss out any deployment issues or bugs ahead of time. While you may not be able to deliver your next milestone a month in advance, Falcone recommends that you “please please please have live content development complete at least a week (or more) before releasing it live.”

He adds that having staffers dedicated to your live platform is key to the success of a large-scale game; no amount of stress and scale testing can prepare you for what might happen during deployment, and assigning devs who might otherwise be working on game features to handle server and database issues is just asking for trouble.

Share the burden by using multiple teams and shifts

Veteran producer Grant Shonkwiler gave similar advice (read our full Q&A with him here) but dug in deeper on the scheduling aspect, noting that if you have people working in shifts to ensure players have support 24/7, it’s important to make sure staffers are working the shifts they want, getting ample time off, and being guided by people who can reliably decide how important a problem is (“does this need fixing right now, or sometime this week?”) and how to escalate it efficiently.

When you move beyond supporting the game to release some new content, Shonkwiler says it’s critical you make sure players have accurate expectations (your community and marketing teams can do the heavy lifting here) and build sustainable schedules that allow for things to slip.

“[For] example: We have 3 releases coming up. We have 12 pieces of content we want released across those 3 releases. We have 2 teams working on these 12 pieces of content. So you set it up this way. Release 1, Content 1-3, Team 1. Release 2, Content 4-8, Team 2. Release 3, Content 9-12, Team 1,” said Shonkwiler. “This allows for teams to have time to work on things without the pressure of having to be part of every release. Also if something can't make it into the next release, it's not a huge problem for it to shift to the next release. “

The key, according to Shonkwiler, is to make sure you’re very tight about what you put in your sprints and what gets added in after you’ve started. Also, add in extra time for iterating and bug-fixing; even if you overshoot, it’s better to have too much time than not enough.

Scheduling out your post-release support so that multiple teams can share the workload without crunching is a sensible tactic, one Ubisoft put into practice while working on the Division games. In a recent conversation with Gamasutra, Ubisoft’s Yannick Banchereau explained that “nobody is working non-stop” on The Division 2 because each game update is (usually) handled by a different team.

Ubisoft's global network of studios affords it the luxury of having multiple studios pitch in on supporting a game like The Division 2, so that every update may be spearheaded by a different studio

“What we try to do is, each update is led by a different studio, so they can take the bulk of the work,” said Banchereau, who served as live content manager on The Division. As the lead studio on the game, Ubisoft Massive helps with the details and main direction of a given update, and then other Ubisoft studios pitch in to share the workload.

“A studio won't be responsible for a whole update,” he continued. “For example for episode 1, the missions of episode 1, the main content of it, are done by a specific studio. And episode 2, the main content of that episode will be done by another studio. Then there's the rest of the work, all the game improvements, the feature reworks, all these kind of things that are usually shared work.”

“A lot of the team on The Division 2 was here for The Division 1, with post-launch for The Division 1, so we have a lot of experienced people who know what running a live game actually means,” Banchereau concluded. “So everyone going into The Division 2 knew what that actually means. Knowing we're going to ship, then we're going to have updates, then we have a rotation of people working on the specific updates, and then they can slow down a bit while another team takes over on the following update, and we try to allocate a structure like that.”

But even if you set up a bulletproof live ops schedule that gives everyone on your team enough time to get everything done and then some, you can still run into trouble if your company culture encourages or abides crunch and overwork.

It’s worth taking a moment to distinguish the two terms: both refer to time spent working more than a standard work week, which in the U.S. is commonly pegged at 40 hours/week. “Crunch” is regularly used as a catch-all term for time spent working on a project above and beyond the standard work week, but according to veteran producer Keith Fuller, it lacks precision because it means different things to different people.

“For some people that means working half an hour past 5 PM, one day a year. Some people view that as crunch,” said Fuller. “Some people don't think much of 80 hours of work a week. But overwork is when you are pushing someone past their reasonable parameters of mental and physical wellbeing. Or demanding that they do so, either explicitly or through the veiled aspects of the workplace culture.” 

This is an important takeaway: you can overwork without actually “crunching” and you can crunch without overworking yourself.

If you’re working so much that your health is being affected, that’s overworking, even if nobody told the team “we’re crunching now!” And if you’re being asked for extra hours, Fuller believes it’s possible to do so without overworking too much -- as long as the timeline is reasonable and the company has a clear policy of prioritizing your wellbeing and fairly compensating extra work.

“One thing you could do up front is say 'okay folks, we're embarking on this endeavor, hopefully it's gonna do X. Maybe it'll be ten TIMES X, maybe it'll even be a MILLION times X, but regardless, we are NOT going to demand more than 10% additional hours from anybody, and not for more than two weeks at a time,” he said. 

“And even if we do, we are going to be setting up comp time; you will be given 2-3 free days off at the end of such a period, because we value the mental and physical health of our people above all else, even above having an enormously happy, rabid fanbase around the world.' That’s an example of the sorts of values-based decision that a company should be making in advance. Just being super clear that yes, we will serve our fans, we will make them happy to whatever reasonable extent we can, but what is unreasonable is that we then exploit our people beyond the limits of their wellbeing.”

Breathing life into a game via regular updates can be exhilarating  -- but it doesn't need to be exhausting

So what can you do to establish a company culture that doesn’t support or encourage overworking? While everyone at the studio has some impact on the culture, Fuller acknowledges that it’s hard to do much on this front if you’re not in a senior position.

The importance of good leadership

“If you don't have a studio leader or CEO on board with this concept, there is nothing you can do. Because when push comes to shove, everybody reports to that person.” he said. “And if that person says 'we're gonna make the customers happy, no matter what. We're going to deliver that build to the publisher, no matter what. We're going to get the demo ready for E3, no matter what.' Then that's what's gonna happen.”

But if you do have the power to change the culture of your company, Fuller recommends making it clear that employee health and wellbeing are paramount, and then set up systems which will hold your company (especially the leaders) accountable to those priorities.

“There needs to be explicit expectations for anyone in a leadership role. And if it's a list in three bullet points, fine; that's a start. But that's three bullets more than what most companies do,” Fuller said. “One of those bullets is, it is your job to promote and protect the health and wellbeing of your people. Full stop. That's it, that is what we expect of you.

“Then, we provide leadership training, which again, very few companies (especially in games) ever do, and we talk about this as an expectation that's on you. Here are the steps we take to make sure you're prepared to keep your people healthy. Maybe we send all our leaders to mental health first aid training. Maybe we make sure that you are trained to use specific language to ensure you're empathetic and supportive of your people. We ensure that we have regular and frequent one-on-ones; every person that you're responsible for on your team, you will have a one-on-one with them,” he continued.

“And part of that is going to involve 'hey, how are you? How are you doing? I care about your physical health, your mental health. Do you feel the workload is too much? Do you have enough autonomy? Do you feel your status is threatened in any way?' These sorts of things. So there's a level of values discussion that needs to take place, it needs to be propagated downwards by wise leaders, and there needs to be a set of explicit expectations for all levels of leadership, and training to back that up.”

Fuller suggested Kitfox Games chief Tanya X. Short as a great example of how a studio leader can use similar techniques to foster a healthy culture tied to clear values.

“She’s adamant about supporting the values they have, to the extent that, I don't remember if it's weekly or daily, but they'll have the whole team stand up and, every day, they'll go over one of their values,” said Fuller. “I don't know whether it's player design, or whatever the value of the day is, but they openly talk about them every week, at least. And you've got the head of the company right there, engaged in the conversation.”

Kitfox cofounders Jongwoo Kim, Tanya X. Short, and Xin Ran Liu

Reached via email, Short confirmed that the Kitfox team does have daily values discussions (typically at the end of a morning stand-up meeting) that have a meaningful impact on how the company operates.

“We recently had our first clear violation of our policies and values, and it was from a key team member, so it made us re-examine our values and re-prioritize them,” she explained. “We realized our people do matter to us, long-term, more than the products we make -- as much as I love the games we make, I value the fantastic people I'm working with and their growth even more, because I think if we keep investing in them, our next game will always be even better.”

Short went on to say that it’s hard to give definitive advice for keeping your studio happy, healthy, and productive. Every employee has their own unique needs and capabilities; it’s management’s job to learn them, and ensure that nobody is being overworked. 

“It's not just about working a certain number of hours -- for me, it's necessary to create personal, 1-on-1 opportunities for colleagues and employees to be observed for signs of overwork,” said Short. “Even then, the solution is often not only to work less, but also differently. I certainly don't envy large corporations that need to figure out systemic ways to care for their people. I'm not sure you can create an assembly line machine of good managers.”

For Short, those opportunities can often pop up during lunch breaks or office socializing, but she also takes pains to keep regular, scheduled 1-on-1 meetings with every employee on the books in order to give them room to talk openly about issues they may not want to discuss in front of anyone else.

“The three questions I ALWAYS ask are: ‘How's life?’, ‘How's work?’, and ‘Can I do anything to help you be more effective/efficient?’, and usually a bonus fourth, ‘Do you have any questions or suggestions?’”, she explained. “But depending on the person and the situation, the discussion varies widely. The important thing, I think, is to deliberately create opportunities for ‘little things’ that aren't a big deal (on both my side and the employee's side) to bubble up BEFORE they become a big deal.”

While Short acknowledged that these sorts of formal grievance-airing meetings sometimes felt stilted and awkward (“there’s not always really anything to talk about”), especially when Kitfox was only four people strong, she believes the potential benefits make them worthwhile.

“Studies say rapport with your supervisor is the #1 influence on whether or not you enjoy your job, so...it’s a bullet we bite,” she concluded.

In fact, your relationship with your supervisor was a focal point of every developer consulted for this article. Each offered unique advice and perspective, but all said that the responsibility of eliminating crunch and overwork from the lexicon of live game ops (and game development at large) rests, at least in part, with management. 

Leaders define the culture of a game company. They help establish your goals and schedules, decide what to prioritize, and how your time should be spent achieving them. If you’re not in a leadership role, then your power to avoid overworking may be limited to finding another job or organizing to resist it, as Sir-Tech devs did in 1998 when they threatened to walk out over crappy working conditions on Jagged Alliance.

But if you are in a leadership role, then you have the power and the responsibility to do something about the health and wellbeing of your colleagues. If you don’t, their suffering is on you. 

“Was one person fired because they wouldn't work a weekend? Was one person driven to the point of tears because of mental and physical exhaustion? Then you've failed. You've failed as a leader, you've failed as a company. Because you've set up an ecosystem that allows that to happen to one human being,” added Fuller.

"That's judging a company's actions on the human impact, and I'd say even if that's only 5 percent or a 1 percent outlier, what was described in the Fortnite article, that is still a horrible mistake that needs to be openly addressed. And needs to be changed.”

The producers quoted here shared so much useful advice that we've gone ahead and published the full transcripts of our conversations with them, in case you'd like further insight on how to sustainably support live ops games: