The VSM Guide

An introduction to the
Viable System Model
as a diagnostic & design tool
for co-operatives & federations

Jon Walker

Version 3.0 (2006)

Section 0: Cybernetic Eyes
Section 1: The Quick Guide to the VSM
Section 2: Case Studies
        Hebden Water Milling 1985
        Triangle Wholefoods 1986
        One Mondragon Co-operative 1991
Section 3: Preliminary Diagnosis
Janus interlude
Section 4: Designing Autonomy
Section 5: The Internal Balance
Section 6: Information Systems
Section 7: Balance with the Environment
Section 8: Policy Systems
Section 9: The Whole System
Section 10: Application to Federations

Appendix 1: Levels of Recursion
Appendix 2: Variety


I began with HWMC because at that time it was the enterprise that I knew most about.

I had been involved in setting it up in 1980, and had seen it grow and prosper in its first five years without any formal structure.

The task I had set myself was to see if the VSM could describe the mechanisms which enable a small loose co-operative to function in an undeniably viable fashion.

Or ... if the VSM had said "this just can't be viable as you don't have a System X, and that's a fundamental of all viable systems" ... I would have had serious doubts about its usefulness to co-operatives.


HWMC is a small co-op that was formed in 1980 to blend and package a large range of wholefoods using fairly complicated machinery.

For the period 1980 - 1985, HWMC was successful both in its financial performance and in the working environment it provided for its members. The production processes became extremely efficient, crises were dealt with effectively, challenges were met and discharged, relations between the members were excellent, profits were good, and in general the system worked beautifully.

All decisions were made by consensus, and weekly meetings (when necessary) usually lasted about ten minutes.

Working procedures evolved during the day: after a 30 second planning session ("Let's do the Basmati rice first"), everyone would work around each other, without formal planning, and the days production would proceed. The process is much more in sympathy with the operation of a Jazz band or a football team: basic rules and constraints are understood, but the specific actions performed by the participants are dictated by the conditions of the moment.

[As an aside, I should mention that working in this way has been one of the most satisfying experiences that I have had, and goes some way to explain why co-ops are full of graduates doing apparently boring manual jobs.]

Preliminary Diagnosis

The initial diagnosis began with the five members positioned as the Operational units. The Metasystem was also composed of all five members, but in a different role: when the work was being done they were Operation, when planning was necessary they articulated the Metasystem. The fundamental co-operative principle of self-management means that there is no clear division in the roles of people working within the group: everyone is obliged to be both "manager" and "worker."

Generally, as the diagnosis proceeded it became obvious that as a function was identified (for example a System 2 stability function whereby the available worker-power was effectively allocated to the various tasks) the same principle was in operation: all members identified the need to articulate a particular Metasystemic function - often without verbal acknowledgement - and shifted into the appropriate mode to deal with the situation.

In many cases, the Metasystemic functions could be done while the Operational work was proceeding (we're nearly at the end of this run ... has anyone got the next one ready?) whereas other discussions required the temporary suspension of manual work, or in unusually complicated situations a few hours put aside to generate plans and strategies.

The Mechanics of Viability at HWMC

  • Most Metasystemic activity (systems 2 and 3) is present continuously.

  • Work proceeds as described above. As long as nothing unexpected occurs very little Metasystemic activity is needed. Most Metasystemic activity (systems 2 and 3) is present continuously.

  • The model of the Operation, located in System 3 and crucial to viability, is continuously being upgraded in the minds of the members. Thus the System 3 model is up to date and virtually complete. (It even knows if someone changes his shirt.)

  • Instabilities in the production techniques are usually dealt with immediately for two reasons: firstly, all members share a compatible System 3 model; and secondly, the performance of the Operational elements is based on co-operation and not competition.

  • .
  • The Metasystem is alerted immediately it is needed: any crisis (for example a sudden large demand for muesli, or a machinery breakdown) suspends the Operational activity briefly and the Metasystem springs into action - that is everyone goes into a huddle; ["Can we cope?" "I can do a couple of hours after work." "Jack is free tomorrow lets get him in."] and so on.

  • The operation of System 4 is firmly based on the System 3 model, and as long as long as everyone has their wits about them , the whole thing works very effectively.

Conditions for Viability in a Small Group

Having experienced this process for a number of years, it is apparent that three conditions are necessary for viability. They are as follows:

  1. Everyone must be capable of, and involved in, both Operational and Metasystemic functions: if this is not the case, the capabilities of the Metasystem will not be sufficient to deal with the Operation.

  2. Efforts must be made to ensure the completeness and availability of the System 3 model. In HWMC, everyone was present most of the time, so there was no problem.

  3. The mechanics of viability are completely dependant on thorough discussion. Effective discussion is essential in the generation of the System 3 model, the workings of the necessary interaction between System 3 and System 4, and the successful discharge of all Metasystemic functions.

Usually, in a group of four or five people, there is no problem in satisfying these three prerequisites, and viability is relatively straightforward provided that everyone works as a team for the majority of the time and the need for all the Metasystemic functions, especially future planning, is recognised.

There would seem to be nothing to prevent a non-structured organisation of this kind working Viably.

The Collapse of Viability in a Small Co-operative

The weakness of the kind of structure exhibited by HWMC is that viability is entirely dependant on the people involved. There is no formal structure to ensure effective viability. Consequently it is not uncommon for the group to degenerate into a non-viable form.

1. Problems with the Metasystem

There are many ways in which the capabilities of the Metasystem can be affected. For example, if a new member lacks confidence he may not feel able to enter into a Metasystemic function, as this may be seen to be the province of more experienced members. Thus vital information may not be forthcoming in a System 3 meeting.

Exactly the same problem may emerge if the "old hands", by their attitudes, discourage new members from becoming involved at all levels.

There also seems to be a potential danger that in a co-op which involves much concentration at the Operational level (say a co-op of computer programmers) very little brain power will be left to deal with Metasystemic issues. In this case it may be necessary to appoint a member to deal mainly with Metasystemic issues, and in this case a different organisational structure will be needed.

A further problem could emerge in a period of continuous crisis when the Metasystemic functions need such a large amount of the time that the Operational functions are neglected (that is, no production gets done). This could be overcome by extending production time, but again the division of the Operational and Metasystemic functions is a possible solution.

2. Problems with the System 3 Model

One of the more common complaints from people who deal regularly with co-operatives is that essential information is not available when it's needed.

This may be caused by movement of members (the relevant information was given to Jill who is out driving today ...), by feelings of unimportance (Jack doesn't think he's valued enough to make a contribution, although he may know the crucial element ...), and so on.

Recognition of the absolute necessity of the System 3 model for viability may be one of the more important contributions of VSM theory to co-ops. It may also avoid the usual knee-jerk reaction to the lack of the System 3 model which is: appoint a manager. Although this would allow a complete System 3 model to be generated (the Manager would act as a reference point for everyone and would thus accumulate the necessary information), other ways may be more appropriate such as a computer model or a large blackboard or magnetic shapes on a sheet of metal.

Summary: Conditions for Viability

A small group may demonstrate viability as long as:

  1. All members are capable of both Operational and Metasystemic functions
    (i.e. they can do the basic work and discuss optimisation, policy, future plans ...)

  2. All members are present all the time. Metasystemic functions need to refer to a real-time complete model of the Operation. If some people are not present, and they have the crucial information needed for a decision, then that decision will be made in ignorance.

  3. All decisions must be based upon thorough discussion by all members.

  4. All Metasystemic functions must be recognised and performed.

  5. The need to keep an eye on the external environment must be recognised, and this information should be used to generate plans to adapt to the future.


Having worked in this kind of co-operative for many years and experienced just how efficient and rewarding it can be, it is clear from the VSM analysis that two conclusions may be made:

  1. This kind of unstructured small group is able to demonstrate viability in the terms used in VSM studies.

  2. This viability is fragile.

The first three conditions for viability are not impossible to meet, but viability can collapse as membership changes, members' personalities clash, some people feel they can take on Metasystemic jobs without consultation, and so on. Condition 2 is obviously difficult due to holidays and sickness, and thus the Metasystem will occasionally have to cope with an incomplete model.

Two recommendations may be offered

  • That daily performance indicators are measured and displayed in meeting places. This will complement the model of the Operation which accumulates during work and discussions, and if viability does begin to collapse, there will be immediate alerting signals.

  • That there are regular slots on the agenda to discuss internal optimisation and future plans to ensure these two conditions for viability are met.