Solving Organizational Complexity With Simplicity
Measured in employees, the fastest organically growing organization on our Bucket List is Dutch health care organization Buurtzorg. In 10 years, they have grown from 0 to more than 10,000 nurses. Such rapid growth is better explained by the laws of biology rather than those of management. Let me take you on a journey of simplicity, scaling and complexity.
This graph shows how Buurtzorg has scaled rapidly since their founding in 2006. (I received this data directly from Buurtzorg.) They have a workforce of >10,000 nurses throughout the Netherlands. Remarkably, they scaled without actively recruiting, and without employing any middle-managers in the organization. Is it possible they scaled so fast because they have no active recruiting and middle-managers? Who knows?
Buurtzorg is best described by what it doesn’t have—managers, unnecessary policies or marketing staff. It has few rules, no promotion policy, no complicated titles and no long job descriptions.
Buurtzorg is about simplifying matters. The company has hundreds of small self-managing teams but only one small office with 50 people, plus 20 coaches and 2 directors who support the teams. There is little supervision and a minimum of meetings.
The aim is to simplify procedures, rules, and communication, and to give the best possible care. Little time/effort is spent on anything not relevant to that focus. The self-managing teams work independently. They can always contact coaches or head office for help but they bear responsibility.
Nevertheless, over 14 years Buurtzorg has become one of the largest homecare companies in the Netherlands. There are now ~900 self-managing teams, each up to 12 nurses, responsible for their own districts. They are autonomous. They plan their work and hire their colleagues.
Through their founding and rapid scaling, Buurtzorg operated with simple principles. They decided not to scale a traditional hierarchy, and instead chose to scale a network of self-managing teams.
These teams handle their own regions. A team is not allowed to get bigger than 12. When it gets to 12 nurses it splits, and each team has their own region. The two new teams scale until they reach the magic number of 12, and then split again.
The organization does not set up new teams. It waits until four interested nurses show an interest in starting a team in a region not yet served by Buurtzorg. After a short training period, the new team hooks up to Buurtzorg's IT system, and is ready to go.
Buurtzorg’s growth resembles the growth of an organism more than the growth of a traditional company. This insight led me to the research of Geoffrey West (British theoretical physicist, former president of, and distinguished professor at, the Santa Fe Institute). His quest has been to find the unifying principles and patterns that connect everything from cells to businesses.
His landmark book, 'Scale: The Universal Laws of Growth, Innovation, Sustainability, and the Pace of Life in Organisms, Cities, Economies, and Companies', says that "most of the literature on companies has been developed from the vantage points of economics, finance, law, and organisational studies, though in more recent years ideas from ecology and evolutionary biology have begun to gain prominence."
West argues we should "search for fundamental principles that govern how the complexity of organisational life emerges from its underlying simplicity." He says "Size and scale are major determinants of the generic behaviour of highly complex, evolving systems." And, within “the daunting complexity of the natural world lies a surprising simplicity and unity—when viewed through the coarse-grained lens of scale."
A coarse-grained lens of scale?
Why focus on a 'coarse-grained lens of scale'? West says: "A major challenge in constructing theories and models is to identify the important quantities that capture the essential dynamics at each organisational level of a system. What is irrelevant at one scale can become dominant at another. The challenge at every level of observation is to abstract the important variables that determine behaviour of the system."
As a first step in this, physicists like West use so-called 'toy models'. These "simplify a complicated system by abstracting its essential components, represented by a small number of dominant variables, from which its leading behaviour can be determined".
He argues that while these highly simplified models are not strictly correct in detail they can capture and explain macroscopic, coarse-grained features of a complex system, even complex systems like Buurtzorg. They provide, says West, a point of departure for developing a more precise understanding of complex systems by refining the first, basic model.
By creating a toy-model of Buurtzorg we aim to understand the emergent laws, patterns and underlying principles of this unique organization. Just as organisms are constrained by the emergent laws of biology, organizations like Buurtzorg emerge from, and are constrained by, the underlying emergent dynamics of social interactions.
West argues "We should aim to understand these laws", because "their systematic regularity and universality provide a window onto the emergent laws and underlying principles." He adds: "As external environments change, all of these various systems must be scalable in order to meet the continuing challenge of adaptability, evolvability, and growth."
Can we construct a toy-model of Buurtzorg? Can we construct a coarse-grained zeroth order theory for understanding the organization based on principles that capture the essence of organizations like Buurtzorg?
At least we can try.
Three Organizational Design Principles
For the first model I propose to use three design principles directly derived from West & his coworkers' research on life's universal scaling laws. These laws nicely suggest there might be a simplicity underlying what we observe in complex systems—and complex organizations like Buurtzorg.
To understand these universal scaling laws, West, Brown & Enquist presented a model in Science suggesting that all organisms use fractal-like architectures of hierarchical branching networks that distribute resources within the organisms. This idealised representation uses three unifying principles or assumptions:
- "First, in order for the network to supply the entire volume of the organism, a space-filling fractal-like branching pattern is required."
- "Second, the final branch of the network (such as the capillary in the circulatory system) is a size-invariant unit."
- "Third, the energy required to distribute resources is minimized; this final restriction is basically equivalent to minimizing the total hydrodynamic resistance of the system."
West and coworkers argue that life's universal scaling laws "arise from the interplay between physical and geometric constraints implicit in these three principles."
It is exactly these universal scaling laws that I use to explain the rapid growth of Buurtzorg, and are the basis for constructing a coarse-grained zeroth order theory for their rapid scaling.
This way of thinking is certainly not flawless. But before pointing out its limitations, it is worth considering the findings.
1. Space-filling, branched organizational network
West & Brown wrote in Physics Today that "organisms use space-filling networks that service all local, biologically active regions in both mature and growing biological systems." They add that the tentacles of these network must extend everywhere throughout the system that it is serving; "It must service all active subunits of the subsystem."
That is, at all scales, organisms are sustained by the transport of energy and essential materials throughout the organisms' whole volume to supply all local part of the organism. They call these networks 'space-filling'.
If we translate this into an organization design principle, then we could argue that firms should have space-filling networks that service all commercially active regions in both mature and growing markets. Moreover, these networks should be branched everywhere throughout the market(s) it is serving.
This is exactly what you conclude when you see the presence of Buurtzorg in its home market. If you plot the presence of their ~900 teams over The Netherlands (and if you are familiar with the typology of The Netherlands) then you see Buurtzorg is utilizing such a 'space-filling and branched network' in the market it serves. (See below).
In the left-most image you see how Buurtzorg teams are distributed over The Netherlands to cover the market they serve. The two right-most images (close-ups) show how teams cover their own dedicated regions and then branch organically throughout the country.
You can also see more teams active in 'mature', more densely populated regions, and fewer in less 'mature', less populated regions. When all these self-managing teams combine in an emergent whole, we see the organization called Buurtzorg.
2. Size invariant teams
West & Brown write that the final branch of the network is a size-invariant unit: "The terminal units of a given network design all have approximately the same size and characterizes, regardless of the size of the organism. Terminal units are critical elements of the networks because they are points of delivery and transmission where energy and resources are exchanged."
If we translate this to an organization design principle than we can argue that teams of firms should have about the same size and characteristics, regardless of the size of the firm. And, all teams must exchange some sort of resource: they must add value to the firm as a whole.
Which means that as the firm grows teams are not reinvented, nor are they significantly reconfigured or rescaled. Teams should be 'ready-made' building blocks of the corresponding organization network. So, the nature of teams should not be reinvented every time we design a firm, regardless of where or how big it is.
This is also something you could conclude when analyzing the nature of teams at Buurtzorg, and how they have scaled rapidly from founding till today (see graph below). Where Buurtzorg consisted of a few teams in 2007, they had 890 in 2019. In all those years, nothing changed the nature of the teams. They were always small (max. 12 nurses), self-managed, multidisciplinary and responsible for a dedicated region.
Such rapid scaling is often to thought to go hand in hand with evolution from simplicity to complexity. However, Buurtzorg has scaled from small to large while keeping small self-managing teams as the basic building blocks. They show that the simple building blocks of an organization do not have to change significantly as they scale from a few people to thousands.
Buurtzorg teams, however, do not act completely independently, but interact in a coherent fashion to ensure that customers get the care they need. That means teams need to be organized such that resources flow based on demand and supply. All activities support Buurtzorg's mission of delivering the best care possible.
3. Minimizing time and resistance to deliver resources
West & Brown write that "Organisms evolve towards an optimal state in which the energy required for resource distribution is minimized." This should not be a surprise. Optimization principles lie at the heart of all fundamental laws of nature. Think about the Theory of Relativity, Quantum Mechanics or Electromagnetic Theory. All can be derived from the 'principle of least action'.
Roughly speaking, this means that of all configurations a system can have, it evolves over time to one physically realized to minimise its actions. In other words, the system evolves towards a state which minimizes time and resistance to delivering resources. This happens via continuous feedback and adjustment mechanisms, leading to system performance being 'optimized'.
West writes: "Highly complex, self-sustaining structures require the close integration of enormous numbers of their constituent units that need efficient servicing at all scales. This has been accomplished in a system optimised by the continuous 'competitive' feedback mechanisms implicit in natural selection."
In other words, West & Brown say that organisms have evolved towards a best state by natural selection to minimize the energy needed to distribute resources in the network, and to maximize the scaling of their area of interface with their environment.
If we translate this to an organization design principle, we could argue the structure of a firm must evolve such that the energy needed for an average team to perform their mundane tasks is minimized, so as to maximize the amount of energy available to serve the customer. It also means firms should evolve to a state which maximizes the number of employees facing customers (internal or external).
We could also argue that the futures of firms depend on the maintenance, reproduction, and integration of all teams in the structure, and the level of efficiency in serving those teams, to regulate activity, supply resources and remove waste.
West writes on this: "The smooth and efficient functioning of social networks requires good communication in which information is faithfully transmitted between groups and individuals. When information is dissipated or 'reflected', such as when one side is not listening, it cannot be faithfully or efficiently processed, inevitably leading to misinterpretation, a process analogous to the loss of energy when impedances are not matched."
Organization structures like Buurtzorg can be regarded as social networks. These determine the rate at which resources and information are delivered to individuals and teams. This rate can be regarded as the 'metabolism of the firm', with levels dependent on the integration of energy, resources and capital via the exchange of information.
That is, it is the network system in the firm that conveys and distributes information, resources, and capital to support, and scale the firm. The scale of firms is then ultimately, constrained by the rate at which resources and information can be supplied through its structure. In other words, the scale of firms depends on the optimization of its 'metabolism'.
This raises the question as to whether organization structures and the metabolisms of firms are the result of optimization principles? More fundamentally, what are firms optimized for?
I would argue that most traditional organizations are optimized to maximize profits. And when traditional firms scale, they tend to optimize to meet the inevitable administrative and bureaucratic needs perceived necessary for operating a traditional firm with many hierarchical levels.
Buurtzorg, on the contrary, seems not to be optimized for maximum profit, but rather for their mission of delivering best possible care. Taking a closer look at their structure we can also conclude they do not seem to be optimized for bureaucracy and hierarchy reasons, but rather for interaction and resource allocation. Let's take a look at their organization chart (below).
We see that the vast majority of Buurtzorg employees are nurses. In 2019, they made up >99% of Buurtzorg's workforce, and all worked in local teams. They are in direct contact with patients. They deliver the value to the customers. Moreover, all are highly autonomous in decision-making—to serve customers as efficiently as possible. The back-office support groups of ~75 people make up less than 1% of the workforce—and are mainly in coaching and director roles.
Buurtzorg has minimal hierarchy, and very few middle-managers. This maximises the time available to deliver patient care. They aim to keep bureaucracy to an absolute minimum to optimize the organiational ‘metabolism’. Buurtzorg's self-developed IT system (BuurtzorgWeb) is crucial here. It was built to support Buurtzorg’s way of organizing. It joins all the teams, and provides access across the community. It supports teams in care-giving, teamwork, and communication. It is claimed to reduce bureaucracy, increase productivity, and improve care quality.
Plus it gives teams real-time insights into results, sick days, client satisfaction, and specific care requirements—plus an overview of workload and holidays. All this information is compared to averages. Each team has access to all data. This is not to create competition, but to allow teams to offer and acquire support and insights. Communication within a team and with other teams is controlled locally, by the teams.
While the software was built for the metabolism of Buurtzorg, it has since been used by 30+ other Dutch organizations who want to adopt Buurtzorg's way of organizing. It is claimed that in all these cases, the software improved the adoptee’s metabolism by improving quality of care and productivity, and reducing bureaucracy.
To be clear, these principles describe an idealized view of Buurtzorg. In reality, Buurtzorg is neither completely space filling nor precisely optimized. And while the Buurzorg teams did not reconfigure as they scaled from few people to thousands, there are some variations in the exact size of all the teams. However, these are minor effects. Such variations only lead to minor corrections to the model.
That's why we say our three principles describe an idealized first-model for structuring and organising the dynamics of fast scaling organizations. In this ‘toy-model’, the scaling of all firms of this kind is constrained by the same three principles. They are space-filling (i.e. a network of team structures), they feature teams of invariant sizes (i.e. small self-managing teams), and they minimize the energy needed to pump resources through the organization (via IT solutions and a lean back office).
I cannot conclude without pointing out some limitations. When we examine the scaling dynamics of Buurtzorg, we see the organization recently hit a plateau. But this doesn't reveal the complete truth. The plateau exists because we only considered the number of nurses working in neighbourhood care in The Netherlands. The origins of Buurtzorg are in this Dutch niche market, and they have >10% share. But they haven't been idle since.
Indeed, they have expanded organically into a variety of areas like domestic help (Buurtdiensten), youth services (Buurtzorg Jong), hospices (Buurtzorghuis), pensions (Buurtzorgpension), psychiatric services (BuurtzorgT), and maternity care (Buurtzorg Kraam). IPlus, they expanded beyond Dutch borders. There are Buurtzorg-inspired initiatives in Germany, Sweden, India, Taiwan, and the UK.
Interestingly, just as the organization itself doesn't set up new teams, Buurtzorg doesn't initiate these expansions. They wait until entrepreneurs knock on their door with a new opportunity. When Buurtzorg shares their enthusiasm, they just go ahead. And although these utilize Buurtzorg's way of organizing, and its brand, they are separate legal entities.
I can't comment on the size and success of these initiatives because I have no scaling data (yet). Nor am I aware of their individual success rates. This is worth pursuing in the future.
West & Brown write that "scaling as a manifestation of underlying dynamics and geometry is familiar throughout physics. It has been instrumental in helping scientists gain deeper insights into problems ranging across the entire spectrum of science and technology, because scaling laws typically reflect underlying generic features and physical principles that are independent of detailed dynamics or specific characteristics of particular models."
They argue: "Therefore, scaling can be used as a potent tool for revealing universal behaviour and its corresponding underlying principles in any physical system.”
It’s time to import this kind of wisdom from the natural sciences to business and management. I attempt that in this post—via a toy-model of the rapid scaling of Buurtzorg. Hopefully, this idealized toy-model can develop into a model for future organizing based on the scaling laws of life, rather than the scaling laws of business and management.
What are your thoughts?