Game Production and team Management at a Chinese studio

June 9, 2021

Disclaimer: This post describes my role as GM at a game studio. I scribbled this down to reflect and decided (and after confirming with the co-founder) to?publish them here. These insights are my own and it’s my intention to recap my journey there and back. Whilst I was in the lead, the achievements – of course – are the sum of the team, the work we did – and support I’ve gotten. I’ll describe how it was to work in a game studio (as General Manager, and as a game producer introducing scrum), the life in China, and our decision to leave again.

I took a job as General Manager at a Chinese game studio in 2016. China was not unfamiliar for me and my wife; we lived there before for a couple of years but I didn’t think of coming back until I was approached for a job interview.

Let me tell you about that job interview which happened just a few weeks later after the first contact; At the time, I was living in The Netherlands and I quickly arranged a visa and took the Friday off to fly to a city called Chengdu to do the interviews in person. The interviews went well and my first impressions of the culture and office were good. The trip was quite heavy with back-to-back interviews in the office, but also outside with dinners and drinks. I gave everything and was happy to sleep on the flight back. I was excited: The studio with 50+ FTE was a serious step up for me (At my previous job I managed 11 FTE team). Back home and at my old job someone asked:?“How was your weekend?”?I wasn’t sure what to answer. (fighting the jet-lag and such an experience!). Within days, my wife and I agreed and I accepted the offer. Things went fast. I quit my job in Hoofddorp, packed, arranged paper works and flew over within next 100 days. (This was the Summer of 2016)

Back in China! The reception was good, with a welcoming atmosphere and the company booked us a luxurious hotel in the same building as the office. (It’s one of those giant Chinese malls with office, hotels and shops included). We started to look for an apartment around the city and HR helped us acquire a residence permit. (read: lots of visits to government buildings, getting medical tests, stamps… the works.) I found a small apartment right in front (15 min walk) of the office. (The yellow building on the picture) It was convenient, at work from my desk, I could see my wife and daughter waving from the top floor’s balcony. Not distracting but I could go home so easily – Just perfect. More on general life in Chengdu later.

The first month, starting as a producer

I’d join as a Game Producer shadowing the co-founder to learn the ropes, with a goal to get promoted to General Manager later in the year. My first day arrived and I introduced myself to the entire team (gulp! – in English) and then proceeded with a little team-building game to get acquainted. Building games is interesting and requires a lot of different disciplines to work together. There are game designers, 2D designers, 3D model makers, animators, programmers, light effects makers and sound engineers. It’s a diverse group of people, and together they can make better products than the sum of the parts. In my experience, programmers like clear process, and art designers like more ambiguity in terms of planning, ‘creative spirits’.

My mission was that I wanted to talk to everybody one-on-one within one month, and jot down stories for reflection. As I didn’t speak Chinese, HR helped translate – a mistake I realized afterwards, – nobody spoke their mind. Actually, I was naive at the time and unknowingly made friends and some ‘enemies’. Anyway, it’s much better to take colleagues for a coffee, or lunch to connect, rather than sit in a meeting room. All in all, talking to everybody was a good idea, I had a full notepad to prove and I many things to think about; I already knew the process of making games, and this studio raised the bar on quality and I learned a lot with the coaching of the co-founder. However, I was?alarmed, I found that some colleges did not like each other at all and I had listened to stories of extreme overtime (All-nighters in the office, skipping weekends) which caused heated emotions and a bit of a blaming culture. (Frankly, this is not unheard of in game development in any country. Games need a lot of polish to be done right with a strict deadline). What was uncommon was that some micromanagement created a ‘rebellion’ mentality – watching over shoulders during overtime if someone is working feels patronizing – for some talented team. The last was something I could address; Game production is about talented and motivated teammates who trust each other, the situation wasn’t good and needed to change. So none of that anymore, let’s focus on output. The successful studio had a numerous quality releases, but they were very dependent of the co-founder to get to that level that the fans were expecting. Watching the floor, I observed idling and constantly asking for next steps. At the end of the day, there would be a ‘inspection’ – art director walking by screens and direct feedback on the work. In other words, it could not?scale, lacked trust?and had a big dependency for the organization to grow.

Baby steps towards new ways of working

Based on previous experiences making websites and games, I presented to the co-founder a plan to introduce?Scrum?as a project methodology. I wanted to replicate the production methods that I knew from before that worked on a smaller scale. It was my mission to enable bottom-up – to create self steering teams.?Scrum?is a behavioral framework that enables a group of people (scrum team and its stakeholders) to use their collective intelligence, to build, learn and deliver solutions. Now, this was in 2016 and Scrum was barely a thing in China, I wanted to change the team?planning, the?office layout?and delivering increments in so called ‘sprints‘. I build a presentation and shared it as well as I could; However, I got the cold shoulder. In retrospect: I wanted too much at once.

A ‘Big Bang’ transformation was out of the window, so I had to proceed with baby steps. As a producer, I had the opportunity to build using scrum – together with a small group of seniors willing to try a different way of working. We started with a working prototype to play-test and demonstrate. Instead of a budget, I got approval for a prototype funding. Total budget was in line with ‘gut feeling’ of a production this size. Each release, we had to ‘prove’ to be worth the next investment. I knew we would get more certainty after each ‘sprint’ (product iteration), and that counted for the budget estimation as well. Our team behaved different – the whole company was paying attention to us and the outcome, and we had to work hard at the time, but we delivered. It wasn’t all without obstacle. Then… we met an issue when we were 60% complete: The CEO of the company did not like the art style of the game. We literally had to go back to the drawing board and start over. Artists started drawing completely new styles for the game, which would require rewiring of the code as well. That was tough. And working with many small releases, should not have happened. Instead, the CEO could have given this a few weeks earlier. We were all learning, and after clarifying this, we got two weeks extra to deliver the game, and that deadline was met! I still feel a special connection with those who made that first game together with me. We proved the success of scrum.

I officially moved to General Manager role. *Yay!*, that first game with the agile production technique led to trust throughout the organization. I started to get involved with company-wide team planning and $4MM budget of the 50+ team. By the way, here I made my second mistake – perhaps because I was too excited and it was quite a lot to handle – looking back, I had a vision and started implementing, but I wasn’t including the management team enough. Because of that, I got into a headlock with of the key leaders of the organization. Here’s what happened: I wanted to stop moving people from project to project, as it was creating a few heroes and not teams. I believed building strong fixed teams would help, as teams can get acquainted and eventually speed up; like the A-Team. However, implementing this, I was blocking the technical leader to grow the team, someone who wants to grow needs to change environments. From then on I made sure to be transparent about my vision and ideas and onboard all leaders before execution and consider more aspects. Things improved immediately.

Running the Chengdu office

Being a GM and managing the office has been a fun job. Every Monday at 9 AM, I would gather the team and give corporate updates and on Friday end of days we drank beers and sometimes had guest speakers/interesting presentations, for example from a game presenter with an interesting new game that hit the market. Besides that, my role involved all sorts of tasks; Building a portfolio of vendors, for example audio, UX and game vendors was also part of my job. I ran the back-end (Java) and web (AWS) team, expanding from a website to a full suite of managed services. (worthy of a post on it's on). But I was also purchasing furniture, arranging stable internet, raising fire safety issues with the building management, and taking action when the pollution from outside became too much. Meeting press, or people from the government. Needless to say I spend a lot of time in that office. I also got to run a couple of events together with the HR team, I remember a Halloween party where we had the whole office decorated in spooky attributes, and everybody dressed up. We ate pizza and played all sorts of fun games, that was great!

The office is one, but it’s all about the people; From the crazy effects, amazing Unity3D frameworks to cute characters, There was a constant need to hire the best, so I was always going through LinkedIn, game forums, Reddit to find new talent. I also worked with recruiting agencies like Randstad and localized recruiters. Honestly, it was tough. As the game industry in China is mature, it’s a great place to be to find talent but it’s also competitive. I assumed wrongly this game studio was looking for cheap labor, maybe once but not anymore. A General Manager of an AAA game company told me: “Don’t think China is about cheap labor, it’s comparable to Canada in terms of salary – it’s talent competing on a worldwide market.” That kind of woke me up. I could compare differences from other locations and indeed, China wasn’t cheap. Besides costs, hiring to an ‘isolated’ Chinese city like Chengdu (West side of China) wasn’t easy. I tried to maintain a mix of local and international artists and engineers to have international appeal with our products. Some Internationals took a good six months to hire and when they joined, they might not like living in China, or the company and leave again; However, some loved it and build their life here. After some time, I got some relations with recruiters in various locations that would send resumes my way. I must have interviewed over a hundred people; Hire, develop and retain the best was a constant big part of my job. I started by hiring a couple of producers that spoke English and knew a bit about Scrum, then build out with artists, developers and game designers.

Besides hiring, I was involved with the evaluations of the entire team. This was one of my big projects in a given year -, evaluating 50+ FTE fairly takes a lot of attention to details and a lot of support from the entire team. Evaluations meant to me: getting 360 feedback from all (at least five) peers. I’d create an overview of the entire team, assess and propose a salary plan to CFO and co founders. That sounds like a smooth process, but in reality, it was tough at times – especially around the salary talks and handling those that did not meet the bar. It was hard work in terms of recruitment and providing growth, however I take pride in seeing the team grow. For example, the hiring of a junior scrum master, who was quite ambitious. With coaching and taking some risks, this person in a few years became responsible for one of the top billing games in the company.

I wanted to know and measure how much a game cost to produce. For that I introduced time sheets. It was quite a hassle to get all the parameters right (overhead and costs per employee per hour) but after a while, it worked wonderful. I didn’t make friends with this decision – the team was used to work without time sheets and it felt like overhead to keep track of tasks. Games in 2016 were largely premium ($1 dollar). However, Apple decided IAP (free games) would be prioritized. Due to the game economy changing, it was essential that we learned how much development cost.

Work Life balance and culture

Now first off: There was no work from home mentality here in this office. Everybody works from the office, from Monday 9 to Friday 6, five days a week. In the evening, many teammates would eat together and hang out in the office. Being together the whole week and work on a single thing does help, especially creating new products. One cultural clash was time management. Locals told me that compared to fast life in Shanghai, Chengdu is known as a ‘lazy’ city and people like to sleep in. I’m not sure if that is true, but did notice most meetings would start 15 minutes late. It was the office culture – actually it felt like a home, people were eating, walking around on slippers. I wanted to avoid wasting time and started by ordering clocks for the walls at every meeting room. Secondly, at some mornings I would sit at the elevators and note down names of those who were late. (Kind of as a joke, but teammates became aware of the time) I know these things sound childish, but they did have impact to time awareness.

On the other hand, a Chinese game studio can be a pressure cooker, with expectations for everyone to do overtime, even in weekends. I had lots of debates on this, I wanted to reduce overtime and focus on healthy work during daytime, to frustrations to leadership. The reasoning was: You cannot create beautiful games on a 9 to 5 mentality. It takes relentless effort, which is true. This was tough, I did remember that during my startup days, I had times that I worked 7 days per week. However, the Chinese ‘996’ (nine till nine, six days a week –?wiki ) was a discussion at the time. My take: However; it can be done but not?all the time. I started quite strict and trying to reduce the overtime hours, but over time found that managing the expectation works best: We’ll do everything we can to let you work hard during work time, however: do expect overtime in the last two weeks of the project. This seemed to balance the discussion.

Scaling scrum training

In this section, I want to share a bit deeper about Scrum and the production of games for those interested.

Besides managing the office, I was an executive producer; To scale scrum and the successful roll out of the self steering team vision, the team needed training; Scrum wasn’t a thing in China at the time. If a candidate during a job interview could talk a bit about agile concepts, they were heavily favored. Training scrum – was directly addressing the hierarchical way of thinking. I showed a Dutch documentary about?‘The end of the manager‘ (added Chinese subtitles) – about the differences between measuring performance at Ford and the production street to creative work and self steering teams today.

I also gave workshops on Agile and Scrum. With a ‘quick course’ in the on-boarding package for new recruits. At one point, I flew one trainer located over (1800 km away) to deliver training. I also did a three-day training to the entire company at once. Finally, some teammates got Scrum Master or Product Owner certified. We managed to re-organize the entire organization to work with Scrum. Story points are an important part of that. A lot of time was lost going through the game scenes to give time estimates. This was a waste of time in my opinion. Instead, we started using story points.

First I showed the image above (without texts). At first, I would ask the weight of the Panda. 80?kg? OK and the Giraffe? 500?kg? These were tough questions to answer precisely. Then I introduced the concept of story points, where you have a Fibonacci scale and sizes are relative. I proceeded asking: Is the giraffe heavier or lighter than the panda? Suddenly it was an easy task. This visual way of explaining worked really well. We needed to do this for all our upcoming work. Stop guessing how many hours you would spend per task, just clarify to me what is more work. Based on velocity (previous completion time) I could get pretty good indicators of time spent on tasks and the total project. Delivery dates became easier to estimate, as programmers and artists estimating features became easier to estimate. The marketing team, under tight deadlines acknowledged that it was working.

On a Friday afternoon, I had the entire studio move from the original line by line floor plan, to a new floor setting. We moved to four large circles of desks. We called them?‘domes’?and teams were free to set organize themselves and decorate their dome, with each a producer (we doubled the producing team by then) responsible for production, but also some HR work; including recording sick leave; time and budget. Ownership on a ‘dome’ level. There was also a scrum-master in each dome to remove impediments. This helped reduce the central management style. Here is where Scrum really took off. The ‘islands’ structure, was a big contributor to ‘self steering teams’ that we aimed for, changing seats paid off. Teams got closer to each other, and we brought work to them, instead of the other way around. Seating arrangements might seem like a small thing but for us it was an enabler.

After some iterations, I proposed a renewed workflow. Remember, ownership issues led to frustrations (who gets to decide on the icon of the game?), so this needed to be tackled. In short: I announced that there can be only two actors that define new initiatives: The Game Designer or the Marketing team lead.

Marketeers would be able to manage a backlog of ‘Marketing Initiative’ tasks: To me, these were ‘distracting side issues’; a necessary evil.?“Could you revamp this game to be compatible with iOS 13?”?or?“can you make a build for Amazon as well”? All these projects would end up on the backlog, and because I counted, I was surprised there were over 100 (!) in a single year. I couldn’t believe this work wasn’t tracked before, was unplanned and essentially invisible to the wider org! These smaller ones usually ended up with a ‘bounty’, asking new recruits or offering extra pay to the veterans. No, the real joy is with producing brand-new games. (Called ‘Production‘ in the chart above) – Our studio could produce 2~3 and later 4 games at the same time. I remember super fun pitch sessions with the various game designers presenting sheets of paper with levels, and other creative visualizations of how a game could work. At times, we (the ‘pitch committee’) would agree to fund an idea to be worked out in a?‘pre-production’?prototype, to be tested with the audience. Once we were convinced of the games fun factor*, it was time for?post-production. Scaling of production team, sound engineering, special effects)

*Fun factor; So how do you make a fun game? That’s a tough question. Unique games are usually certain input (tapping/ swiping) with a novelty theme. Difficulty is another important aspect, too difficult is not fun. We never copied games or concepts, we took pride in original contents, and noticed that other studios would copy our games in just a few months after launch.

We’d gather often at the ‘scrum of scrum’ epic board for company planning. (One priority for the whole org). Once per quarter I'd organize a session with leadership to do 'planning poker' with all companies epics. It's really tough to set one priority for the company, we working with our annual theme's with SMART goals to prioritise.

As were started planning the team as one leadership, and with the domes in place, we finally entered the dogma of bring work to one team/ ‘single team’ approach. Then I suddenly had to scale one game agile plan to multiple, how could that work. We called it:?The Mother ship.?(The idea was that small teams would be autonomous and eventually be able to go back to the mother ship) Teams were stable groups, sitting with each other and working together. We planned work for the teams, instead of revamping teams constantly. We would have multiple teams work on various games on one ‘global’ backlog (viewable by everybody), where we only picked up projects that were initiated as explained above, and had a clear owner. This led to deadlines becoming more realistic and ‘time’ wasn’t a negotiation piece anymore. Technical debt and innovation needed to be build in. We created the notion of ‘slack’ in sprints to allow time for that. However, in retrospect this wasn’t enough. I think we’d better had more attention to these topics and really call them out more explicitly on our backlog.

Lastly, I put these giant screens at the canteen and put the ‘mother-ship’ epics on the screen with a giant countdown. Projects would at first be green and slowly turn red towards the deadline; At first they would pop up as green, and slowly turn red if time passing by. They surely led to lively discussions during lunchtime. For example, QA would realize that two project would finish around the same time (problem as all games had to be tested on all relevant iPhone’s, iPads and Android devices, which led to company wide discussions. The aim was to be super transparent with our pipeline and avoid surprises down the line.

Life in China and our decision to leave

During the first years we lived in a small apartment, close to the office. I actually miss this apartment today. It was a simpler time. We walked a lot, went to the local gym and kids to/from the kindergarten and such. There is lots of excitements around the city, like new restaurants and ‘happening’ places.

The main downside of living in a big Chinese city for me was the air quality. During winters, the air got so bad that we breathed small particles at a dangerous level (> 500 AQI) inside the office and our house. We took action and made it bearable by isolating the house and purchasing air purifiers for every room. Our indoor AQI went lower than ‘clean countries’. Outside, during bad days we wore masks (before COVID-19 made this common) most of the time. In The Netherlands we often talk about the weather, some (conscious) people in China always talk about the air quality. Over time the pollution seemed to be getting less though, and during summers blue skies aren’t uncommon.

Later (After resigning), I was offered to move to the vacant house of the co-founder and could enjoy a more luxury lifestyle. It was right next to the Korean school of my daughter. I remember nice moments like Saturdays afternoon when we would go to Korean school and buy stick bread on the way home. Cycling around town on those Chinese WeChat lending bikes. Life at home was good. The weekends we would spend in malls, or in a swimming pool. Chengdu is a fast changing city, just in one year, I have seen the environment change completely, including new tunnels being build to handle all the traffic. Impressive. Chengdu is relatively close to the mountains. With an hour drive you can be in the mountain range which have clean air all the time, whenever I could, I would take trips outside the city. Visiting?Shangri-la?and Lhasa Tibet with my dad were an absolute highlight, those were far. But with just a two-hour drive one can be?plucking cherries in Wenchuan?or visit places like Songpan (blog post).

In the summer of 2018 – after three years – things were running smooth! Actually, the game company was just bought by a huge company. My wife and I saw my children grow older in a third country and started to wonder what the future would bring. We decided it was a good time to leave. Once I wanted to resign, I had a talk about it with the co-founder. They wanted to keep me aboard for a smooth handover, so I stayed for another year. At that time the HR manager quit as well, so this was an excellent opportunity to find someone new. – I said goodbye with a?hotpot dinner. I still think about my time at this studio often. It was such a whirlwind, with stress, emotions, overcoming sudden problems and generally a stressful feeling. It was a time that I had to work relentlessly, but I also felt very much alive, However, looking back - I’m glad we did it.

After note: Thanks for making it this far! I’d like to hear from you! For those working with me, is this how you experienced it as well? And if you want to learn more, reach out! And if you want to know more about the peculiar world of making games, I recommend reading ‘Blood, Sweat and Pixels ‘.