Systems come in a variety of sizes, and they change as they scale up or down. Sometimes it is more effective to stay small because you anticipate that the changes growth will require of your system are not ones you want to make. When scaling up, it’s important to be aware that growth in systems is often nonlinear, which means that changing a single component might have effects that fundamentally alter your system and create both new opportunities and new dependencies. In baking, for example, it’s common that doubling a recipe doesn’t always work. You often need less yeast; because of the geometry of the bread you’re baking, double the yeast might cause the fermentation to be too fast. Scale as a model gives us insight into how size might impact our system choices.
When we study a complex system, it’s beneficial to consider how its functioning behaves differently at different scales. Looking at the micro level may mislead us about the macro, and vice versa. In general, systems become more complex as they scale up. Greater size means more connections and interdependencies between parts. Thus, it’s important to combine scale with bottlenecks. As systems become larger, different parts might struggle to keep up. Imagining your business scaling up in some areas faster than others lets you anticipate breakages and bottlenecks.
« If you do not look at things on a large scale, it will be difficult to master strategy. »
Miyamoto Musashi1
To give an example of how things change as they scale, consider a company at two different sizes. As a small company with a handful of people with close personal relationships working together in a garage, there’s no need for an HR department or management consultants. They can work together and solve problems face-to-face. Proximity can discourage them from letting too much tension build up. No one is going to steal a coworker’s lunch from the fridge because it’s a tight-knit group and the culprit would be obvious.
Fast-forward a few years when the company is larger, with 600 employees in several offices. Many have never met, and few are friends. Scaling up means the system has completely changed. It’s now necessary to hire people whose entire job is to organize and make sure everyone gets along. To avoid communication bottlenecks, the company divides into teams, meaning they are better able to manage social dynamics. Provided links remain between parts, systems can safely scale in this fashion: by dividing into parts. But things will always be different as a system scales, and a collection of teams within a company will never be able to communicate like a small company. The larger the company grows, the more work it takes to ensure information flows to the right places.
As companies increase in scale, parts of the system break because what works for ten people doesn’t typically work for a thousand. As changes to the system are implemented in response to growth, the question always is: How will this system fare in the next year? Ten years? A hundred years? In other words, how well will it age?
As growth occurs, resilience can be increased by keeping a measure of independence between parts of a system. Dependencies tend to age poorly because they rely on every one of their dependencies aging well.
In economics, production processes change as they scale. The more of something that is produced, the more the marginal cost of each additional unit tends to shrink. As more people can afford a product, demand tends to increase. Owning it may become a norm or habit. Economies of scale work because they enable cost-cutting measures, such as purchasing materials in bulk. Systems do not scale indefinitely; economies of scale begin to break after a certain point. Eventually saving any more money becomes impossible, or there may be no more possible customers. In addition, limitations exist when there are dependencies on finite resources, such as energy, raw material, or computing power.
Success often sows the seeds of its destruction. Sometimes getting bigger means becoming more vulnerable, and some things are most apt to survive if they stay small. After all, the majority of species on this planet are insects—tiny, simple creatures.
In business, scaling is often seen as inherently good. The bigger a company gets, the more successful we consider it. We hear laudatory stories of how fast new companies grow—hiring more people, opening new offices, and spreading their products or services to vast new audiences. But getting bigger can make companies more fragile. During difficult economic times, companies that scaled too fast can struggle to sustain themselves. Sometimes, when longevity is the goal, staying small and simple can be a superpower.
Most businesses fail in the first few years. The largest companies around at any given point in time, however mighty they may seem in the moment, don’t last long. The average life span of an S&P 500 company is 24 years, and this number is decreasing over time.2 In most parts of the world, a company lasting a few decades is remarkable. Yet in Japan, that’s not the case. The country is home to an astonishing number of incredibly old companies, known as shinise.3 Over 50,000 Japanese companies are more than a century old, with nearly 4,000 dating back over 200 years.4
Why are long-lived companies more common in Japan than the rest of the world? It’s impossible to know for certain. But most of the oldest companies have something in common: the way they scale. Or rather, the way they don’t scale.
Long-lived Japanese companies tend to be small. They’re owned and run by relatives and people with close relationships. They usually have fewer than a hundred employees and trade within a small area inside Japan. Durable, loyal customer relationships are integral to their business models. Also, they are driven by a strong internal philosophy that goes beyond their products and services, enabling them to adapt to changing times.
By staying small, long-lived Japanese companies can hold on to their traditional values. Being no larger than necessary benefits them during less favorable economic conditions. In a small team where a job may last a lifetime, diffusion of responsibility is less of a problem as there’s nowhere to hide. Employees may be more invested and take their work as a point of pride.
Take the case of perhaps the most famous long-lived Japanese company, Kongo Gumi. A construction company specializing in high-quality Buddhist temples, it operated independently from 578 AD to 2006. Today it exists as a subsidiary of a larger company. At the time of Kongo Gumi’s liquidation, it was the oldest company in the world, having built Japan’s first-ever Buddhist temple. Throughout the entirety of that time, it remained in the hands of the same family—40 generations of them. Each owner passed the company on to his oldest son.5 However, to ensure this close-knit succession system worked no matter what, it had some flexibility. If the oldest son didn’t have the right leadership potential, a younger son would take over. If none of the sons were suitable or an owner had no male children, they would select a suitable husband for a daughter, then adopt him. Adult adoption for business purposes is a common practice in Japan even today, enabling companies to stay within a single family for many generations.6 At one point, the widowed wife of an owner took charge of Kongo Gumi.7
To give context for the length of time Kongo Gumi remained in operation building iconic temples, at the time of its founding, the Roman Empire had just collapsed. The prophet Muhammad was not yet a decade old.8 The world changed a great deal between then and 2006. Kongo Gumi survived numerous wars, periods of immense political upheaval, economic crises, and other disasters. It managed this by adapting to the times. For instance, during World War II, demand for Buddhist temples was low. The company switched to making coffins.