Culture Clash

(October 2007)

Weight Watchers

"…Reasonableness applied to processes that reduce elements to their base level and puts them back together eliminating the waste built into them." - Yoshiki Iwata, Shingijustu Institute

That's one definition of an industrial strategy known as Lean Manufacturing. Derived from and inspired by the Toyota Production System, Lean comes in many forms and offers a buffet of tools, best practices and techniques for optimization of process efficiency. If the definition seems unnecessarily complicated, we can pare it down by 20 words, leaving only “reduce waste.”

The systematic elimination of waste – defined in Lean philosophy as anything that consumes resources but contributes no value to the process or consumer – leads successful Lean organizations down a path of tidy production efficiency and often dramatically improved throughput. When properly implemented, Lean's ability to optimize the passage of materials and information through processes adds considerable value.

Alongside Lean is a discipline called Six Sigma, a statistical approach to quality maximization originally pioneered at General Electric and Motorola. The two concepts share many core philosophies, but where Lean's objective is waste-free speed and efficiency, Six Sigma is a quality tool. Lean/Six Sigma, therefore, means more efficient processes at superior quality.

Lean/Six Sigma is not without detractors, and it's worth noting that of major corporations adopting variants of the philosophy, more than a few of them aren't doing as well as they expected. Thus anyone adopting Lean/Six Sigma should weigh the risks and benefits carefully. I'd personally suggest that for disciplines intended to reduce waste and streamline systems, Lean and Six Sigma may have simply gotten bloated. But even if that's not the case, there is no doubt that despite many examples where it is proven successful, Lean isn't a bullseye every time. And it certainly hasn't been successful – or even tried – in every industry.

The videogame business has its own idiosyncrasies, and Lean philosophy cannot be whole-hog applied with much hope of success. Zero inventory would not, for example, work in a world of region-specific console launches, where it's impossible (and undesirable, in the short term) for initial supply to meet demand – sometimes for months. The unpredictability of new console attachment rates would similarly hobble a just-in-time production system.

But some things – slightly customized – would work. With hindsight we can speculate that they may have resulted in significant savings for the applying company in the form of reduced waste, shorter time to market, and superior customer satisfaction. For example, eschewing zero inventory while adopting visual mapping of Wii production might reveal key areas of waste and improve manufacturing speed, getting backlogged product into consumers' hands faster.

Going Lean isn't the only option, but its philosophies have broad relevance, and ignoring them altogether can be disastrous. Microsoft is currently reaping what it has sown by rushing the 360 into production rather than applying Lean/Six Sigma quality at the source mechanisms during design and engineering. In failing to do this, the company produced the Ford Pinto of the videogame industry and sealed its own fate in the console war. The result – a billion dollar mitigation fiasco, circular warranty return shipments of still-broken consoles, furious consumers, an increasingly clumsy and backlogged service center – is the definition of a wasteful system that benefits no one. If “waste” is anything that consumes resources but adds no value, that example is the Lean equivalent of a black hole.

The role of Lean in software development – a much less industry-ey industry – is more complex. Development is primarily creative. Emotionless statisticians and bean counters cannot build amazingly powerful game experiences, and Lean engineers are exactly that. In fact, it's best to separate these people and the chromatically talented artists who create games, bridged by those rare individuals gifted enough to speak both languages. You don't want some stooge messing up your opus by insisting that you're only budgeted for 5000 polygons per object; but that doesn't mean efficient processes need go out the window.

The culture of mature, stable business activities offend the bring-your-bird-to-work mentality of many game developers, people who are naturally independent and often suspicious to the point of xenophobia when it comes to ideas from outside. But structure and efficiency doesn't necessarily interfere with the uninhibited attitude that is so core to the development community. It depends on how it's presented and implemented. Frankly, any business process expert who tells you that khaki pants and a necktie are the only way to guarantee good work is a moron and therefore safely ignored. It's possible to wear hemp and make great games while still adhering to ordered, symmetrical models.

The application of Lean workflow methods can greatly enhance the development process. That's not meant to be another “the business needs more proven process models” argument we hear so much (because it's true); in fact, it's a way to improve games themselves, along with budgets and quality of life and so forth. A good game and a good Lean implementation have a lot in common. Ideal Lean systems depend upon the concept of flow: the seamless transition of information, products and services from origin to destination, uninterrupted, neither faster nor slower than explicitly necessary. Ideal game experiences also flow, while bad ones are recursive and convoluted. Good business and good games are inherently elegant – visible, predictable, adjustable, organized and repeatable.

“Waste” in development could be rework, unused assets, reversed design decisions, creative clashes, you name it. Largely it would be the product of poor planning, which, I'm sorry to say, does lead right back to the “business needs more proven process models” argument. Fortunately, the people who've been saying that haven't just been braying into the wind; they've been beaverishly productive. And the industry is beginning to listen.

Some argue that Lean/Six Sigma can be applied, unchanged, to any business model, and that one of the biggest mistakes an industry can make is to assume that it is unique among others. I prefer to believe there are shades of gray between commonplace and unique, and the games industry, like everything else, falls somewhere in the middle. Even if Lean philosophy isn't the right path, we can still learn from it. Recognition of its key elements – waste reduction, workflow optimization, increased quality, measurable results – would do the business a world of good. This philosophy, developed by outsiders, offers many commandments of great potential value.