Topic · Structure and governance

Viable System Model (VSM)

The Viable System Model (VSM) describes which functions an organization needs in order to remain viable over time in a complex, changing environment. It looks not at the organizational chart but at the “nervous system” of an organization: how decisions and information flow. The VSM makes it possible to check which functions are well developed, which are missing, and where things get backed up.

Schedule an initial conversation

On this page

What the VSM describes

The Viable System Model (VSM) comes from the British cybernetician Stafford Beer. Cybernetics is the science of control and communication in complex systems. Beer applied it to social systems: to companies, public administrations, or associations. Its model is the living organism.

A vivid image comes from Martin Pfiffner, who adapted the VSM for practice. In his view, an organization has three dimensions:

Anatomy

The organizational chart

The division into departments and positions

Physiology

The processes

The workflows through which the work gets done

Neurology

Control and communication

The dimension that makes it possible to cope with complexity. This is exactly what the VSM describes.

Behind this are four basic assumptions:

Viability

A system is viable if it can adapt to a changing environment under its own power while preserving its identity.

Self-organization

Viable systems regulate much of what they do themselves. Above all, control creates conditions and feedback.

Information

The capacity to steer comes from functioning communication and feedback. A system can only steer itself as well as it knows about itself.

Balance

The goal is a balanced relationship between the autonomy of the parts and the cohesion of the whole, neither pure centralization nor pure decentralization.

The five systems of the VSM at a glance

The VSM distinguishes five interacting functions, called “systems”: the actual work, its coordination, control in the here and now, the view outward and into the future, and identity. In addition, there is an independent view of reality, System 3*.

The five systems. Each operational unit (System 1) is itself a viable system with the same five functions, indicated by the small stacks.

The systems in detail

System 1

The operational units

They do what the organization exists for: producing, advising, caring, selling. This is where value is created. Each unit has its own environment and management and acts as independently as possible. Support functions such as HR or accounting are not part of it, even if they appear on the same level in the organizational chart.

Guiding question: Is the focus on what customers pay for?

If it is missing: Units are not allowed to make their own decisions, everything goes up the chain, bottlenecks form.

System 2

Coordination

Where units share customers, resources, spaces, or schedules, System 2 dampens the oscillations: standards, agreements, joint planning, a shared calendar. It does not give orders; it makes self-organization run with little friction.

Guiding question: Can the units coordinate with each other where they depend on one another?

If it is missing: Friction, duplicated work, back and forth. Or the opposite: too many rules and reports.

System 3

Control in the here and now

Looks inward and at today. Sets priorities, negotiates goals and resources with the units, uses synergies, and resolves conflicts the units cannot resolve themselves. Intervenes when self-regulation and coordination are not enough.

Guiding question: Do operational decisions serve the strategic goals?

If it takes over: Management meddles in day-to-day business and stifles personal responsibility.

System 3*

The independent view

Supplements System 3 with unfiltered information straight from the work: spot checks, quality audits, site visits, conversations across levels. This is not about mistrust but about a realistic picture.

Guiding question: Does management know what is really going on, and in time?

If it is missing: Leadership only sees embellished reports and learns about problems too late.

System 4

Environment and future

Looks outward and ahead: markets, technology, society, laws. Recognizes opportunities and threats and connects knowledge about the environment with knowledge about the organization's own capabilities.

Guiding question: Does the speed of perception keep up with the speed of the environment?

If it is missing: The organization is completely absorbed by today and misses changes.

System 5

Identity and direction

Determines who the organization is and what it stands for: purpose, values, principles, guardrails. Makes normative decisions that do not follow from numbers alone.

Guiding question: Is there a clear purpose and a final authority that is effective?

If it is missing: Purpose and values are unclear, or the top loses itself in operations.

System 3 wants stability and efficiency, System 4 change and renewal. Both are right. The most important ongoing task of System 5 is to keep the balance between today's business and tomorrow's.

Variety: Only variety can absorb variety

Variety is the measure of complexity: the number of possible states or behaviors of a system. The environment always has more variety than the organization, and operational work more than management. Behind this is a fundamental law of cybernetics, formulated by W. Ross Ashby: Only variety can absorb variety. Anyone who wants to control something needs at least as many options as what is being controlled. In the VSM, this shows up at every point where the environment, units, and management meet.

Because this never fully succeeds, viable systems work with two levers:

	Amplifiers	Attenuators
Effect	increase your own impact	filter complexity
Examples	training, tools, delegation	standards, rules, metrics, summaries
In the team	The team lead hands over responsibility.	The team lead sets clear guardrails.

For design, this means: The vertical axis of control must be strong enough to carry the horizontal variety of the operational units.

Is control equal to the variety of the units?

Three ways: strengthen control, reduce the number or diversity of the units, or add another management level. The third way is best avoided.

Yes

Then the organization may even be flattened.

Recursion: Systems within systems

The VSM is recursive. Every viable system is embedded in a larger one and itself consists of viable systems. The same five functions are found at every level.

Group

National subsidiary · one level up

Region · system in focus

Site · one level down

That is why any work with the VSM begins with choosing the system in focus: Which unit are we looking at, what lies one level above it, what one level below? From this follows the principle of subsidiarity: The higher level decides only what the lower level cannot decide better itself. Because a department is itself a VSM, the VSM can also be applied to a division or a team with its own task, as long as it is viewed as a unit of an organization.

Channels and feedback loops

The systems are connected by communication channels, and the VSM thinks in closed loops, not one-way streets. The most important loops:

Loop	Systems involved
Self-coordination of the units	1 and 2
Prioritization and synergy	3, 2, and 1
Negotiating goals and resources, with a realistic picture	3, 1, and 3*
Linking strategy and operational planning	4 and 3
Balancing today and tomorrow	5, 4, and 3

Four questions apply to every channel:

Capacity

Does it have enough capacity for what has to flow through it?

Clarity

Is the message understandable and “digestible”?

Rhythm

Does its cadence match the pace at which things change?

Reliability

Is it robust against failures, and does the loop close?

A special feature is the alarm channel, Beer's “algedonic signal” (from pain and pleasure): a direct path from the operational level to the identity level that skips all intermediate levels in the case of serious danger or great opportunity.

Typical patterns

When a function of the VSM is missing or takes over, recurring patterns appear. In addition to the patterns of the individual systems above, there is one that affects all of them:

Open loops

Information gets lost along the way, decisions do not arrive, committees multiply without anyone being clear about what for.

Ten common mistakes in organizing, according to Martin Pfiffner

starting with the organizational chart
organizing without regard to customer value
copying the competition
optimizing weaknesses instead of using strengths
retreating into complicated matrix structures
starting with people instead of tasks
overlooking the nesting
organizing the new within the old
not distinguishing between support units and operational units
neglecting the “neurology”

Quick check along the five systems

For a first look with the VSM, a few questions per system are enough. Recorded on a single page, they show what is well developed, what is missing, and where things get backed up.

System	Guiding questions
5 · Identity	Is there a clearly formulated purpose? Are the principles within which leadership acts clear? Is there a final authority for fundamental decisions, and is it effective?
4 · Environment and future	Is the strategy known and embedded? Are gaps and opportunities perceived and changes initiated? How quickly does the organization respond to its environment?
3* · Independent view	Does management receive unfiltered information straight from the work, and in time?
3 · Control today	Are priorities set deliberately, changes implemented consistently, synergies used, resources allocated for the good of the whole?
2 · Coordination	Can the units coordinate where they overlap? Is coordination effective and well run?
1 · Operational units	Is the focus on customer value? Can the units act as independently as possible, with the necessary resources and competencies?

Seven steps to design

Pfiffner organizes work with the VSM into seven steps. Implementation follows.

Operational units: Which units fulfill the actual purpose? They should be designed so that, in principle, they could also exist on their own.
Manageability and flattening: Balancing variety. Is control equal to the variety, and can a level be eliminated?
Critical success tasks: the permanent tasks that determine future success, derived from the environment, the work, overlaps, and dependencies
Centralized or decentralized: Can we afford to hand the task off? Does it affect what matters to customers? Does bundling bring real synergy? Customer value comes first, then synergy.
Designing Systems 2 to 5: among other things, the committee landscape. Each meeting is assigned to a system, with the question of what would be missing if it did not exist. Add a responsibility matrix: who carries out, is accountable, supports, advises, or is informed.
Communication channels: channel check and meeting rhythm across levels
Representation: the target picture of clear responsibilities, lean committees, and closed communication paths

The attitude behind it: structure before person, meaning you look for problems first in tasks, channels, and committees. As much autonomy as possible, as much cohesion as necessary. Think from the purpose. Loops instead of chains of command. And you work on the system instead of in the system. What matters is that leadership is involved from the start, because holistic design cannot be delegated. In Pfiffner's experience, a redesign without leadership that guides the change remains a technical exercise and does not hold.

Common misconceptions

Common assumption	In fact
“The VSM is a new organizational chart.”	It describes functions and communication paths, not boxes and hierarchy levels. One person can fill several functions.
“System 5 is the boss.”	Not in the sense of authority to give orders. System 5 holds identity and balance; most decisions are made elsewhere.
“VSM means: decentralize everything.”	It seeks a balance between autonomy and cohesion.
“System 3* is mistrust.”	It ensures a realistic picture, so that control is based on reality rather than embellished numbers.
“This only applies to large corporations.”	The VSM can be applied at any level, even in a small business.

Limits

Not a model of people. The VSM says nothing about the character or abilities of individuals.
Not a recipe. It cannot be applied mechanically and does not replace conversation with those involved.
Not the whole of reality. Culture, relationships, and experience matter, but lie outside its view.
An organizational model. It can be applied to teams only through recursion, as a unit of an organization, not to individuals.

Further reading: Martin Pfiffner, The Neurology of Business. Implementing the Viable System Model (Springer, 2022).

Let's get to know each other.

The first step is clarity: an initial conversation.