Soft Systems Methodology:Soft systems methodology

6.3 Soft systems methodology

Soft systems methodology (SSM) was developed as a response to hard systems thinking and its failure to address messy situations in which no clear problem definition exists. We have an understanding of 'systemic' through the systems fundamentals introduced in chapter 5, i.e., thinking with wholes, or holism. In soft systems thinking we organize our thinking about the world systemically, while recognizing that the perceived world is problematic. The word 'problem' can have negative connotations in some organizational settings, where people talk along the lines of having 'opportunities' rather than problems. In SSM the phrase 'problem situation' is not used pejoratively, in the same way that 'negative' feedback does not in itself indicate good or bad. To avoid this mental trap the negative feedback loop in chapter 5 is often known as a reinforcing loop. Similarly with SSM, the phrase 'problem situation' does not indicate good or bad and therefore encompasses 'opportunity' situations as well. However, custom dictates that we will continue to use the term 'problem situation' in the context of SSM.

SSM supports an enquiring process and is most commonly presented as a seven-stage model (Checkland & Scholes, 1990) (figure 6.2). A key aspect of figure 6.2 is the division between the real world and systems thinking about the real world, which we can think of as 'above the line' thinking and 'below the line' thinking respectively. Above the line is concerned with our perceptions of the real world problem situation; below the line is the province of conceptual thinking, using systems models that help us gain insight into the problem situation by comparison with the perceived, above the line, world. Possibly the most difficult aspect of SSM to grasp, but an essential one if the potential of SSM is to be leveraged – is that the below the line systems models are not models of the real world problem situation. The models are devices that serve to inform our understanding of the problem situation above the line, to foster discussion and debate, and to surface deep-rooted assumptions held by those involved in the intervention. The below the line models are not models of the situation and it follows that neither are they statements of what should be, ought to be, might be, or could be (see Checkland, 1995 for a discussion of the role of conceptual models in SSM). This is an important aspect of SSM and the point will be reiterated as we work through the basic SSM activities.

A further word of caution is needed before we look at the content of the 7 stages of SSM. The 7-stage model is helpful insofar as it makes visible a process for 'doing' SSM. It is also unhelpful in that it reduces SSM to a set of stages that suggests a step-by-step method that can be picked up and applied. In practice, experienced SSM practitioners use SSM ideas and frameworks as a guide to organizing an intervention and not as a recipe book. The experienced SSM practitioner will take account of the interplay between the situation, the people involved (including the self-awareness of the SSM practitioner), and the problem situation. It might not, for example, be appropriate to begin an intervention by the SSM facilitator introducing the project team to the principles and terminology of SSM, as this might not be perceived as meaningful. Systems thinking can still be applied, but not in a directive or prescriptive way. For the purposes of understanding we will use the sevenstage model to structure our description of SSM using the Barchester Playhouse case study as an illustration, but these concerns about the application of a stage model should be borne in mind.

6.3.1 Exploring the problem situation

Situation considered problematic [1]

The starting point is for someone to perceive a situation as problematical. The issue does not pre-exist in the real world; it must be perceived and recognized as problematic by a would-be improver of the situation. For example, a new manager joins the company and is concerned that there is poor representation of women and minorities on the staff. This might be the motivation for an enquiry into why this situation prevails. In the Barchester Playhouse case study one motivation is the Chief Executive's concern with the inability of the box office to cope during the last pantomime season. A further motivation might be the web site manager pushing for box office operations to be Internetenabled. In this situation there is a senior management top-down concern for a business issue and a bottom-up technology push from an Internet evangelist. Either or both of these concerns – one a 'problem' and the other an 'opportunity' – might be the catalyst for an intervention.

Problem situation expressed [2]

The second stage is to attempt to express the situation that is considered problematic. In traditional systems development this might take the form of the development of a process model and a list of problems associated with the current business process. In SSM the aim is to open out the situation, to avoid the temptation of closing the situation down into a neat box and line graphical representation. The rich picture diagram is one way of representing our mental models of a problem situation, helping us to surface and record our assumptions about the relationships and interconnections between the elements we perceive as being pertinent in the problem situation. The rich picture diagram is not a formal technique and each person will develop their own style (and work around their drawing skills as necessary!). Rich pictures can be created using graphics software, such as Microsoft Visio, but there is a danger that the result will be rather stiff and formal and the use of standard clip-art can make it clichéd.

Rich pictures develop over time as the intervention unfolds. This means that the original diagram will be elaborated or re-drawn entirely as the project develops. The rich picture is not an objective representation of an external reality; it says as much about the person(s) creating the diagram as it does about the problem situation. Rich pictures can be created collaboratively with the client or used as an internal thinking device by the SSM practitioner. What is appropriate depends, as usual, on the situation and on the characteristics of the would-be improvers of the situation. Our experience is that it is useful to develop rich pictures collaboratively in a workshop with members drawn from different areas of the organization. This would not work in all organizations in all interventions and the SSM practitioner must therefore be sensitive to the setting.

Although the technique is informal, a few conventions are seen as being useful and have therefore gained general acceptance. Crossed swords are used to represent conflict, an eyeball to represent a watching interest, and a hurdle to show a difficulty in communication between elements, such as departments in an organization. Human actors are often represented by stick figures and their opinions by speech bubbles; hidden or unarticulated views can be shown with thought bubbles. Practitioners are encouraged to create their own symbols that work for them and are relevant to the specific problem situation being explored.

The rich picture in figure 6.3 was developed by Dominic Leaver, an MSc student taking the IS Development course at the University of Bath. Dominic has used a magnifying glass to represent scrutiny, a lightning flash to represent 'future shock', and scales to show the trade-off between people issues and financial cost together with their impact on the project timescale. Dominic's background as an architect and his strong graphical skills have enabled him to produce a rich picture of complexity and aesthetic appeal. But, don't worry if your drawing and drafting skills are not up to this level of aspiration.

The rich picture is about the communication of perceptions of a problem situation and aesthetics are only part of the mix. If your abilities with a pen are limited try using a mixture of clip art and hand drawing to produce recognizable icons while retaining a sense of the situation as grounded, organic, and messy.

6.3.2 Below the line: root definitions and conceptual models

The dividing line between the real world and systems thinking about the real world is fundamental to SSM. Above the line activities are concerned with the real world; below the line thinking is concerned with developing systemic models that help us to think about and debate the real world. These models are not models of the real world. They are not models of how the real world ought to be. They are simply devices that are more or less relevant to the problem situation being investigated and offering more or less insight into the issues. A model can therefore be formally correct but offer very little insight into a problem situation. A successful model is one that promotes discussion and debate, surfaces hidden assumptions, and questions deep-rooted beliefs. There are two main components of below the line thinking: root definitions and conceptual modeling.

Root definition of relevant systems [3]

A root definition is a short textual definition of the purpose and means of the system to be modeled. The root definition should tell us what the system will do (X), how it will do it (Y), and why it is meaningful for it to be done (Z). This can be encapsulated in template form as:

A system to do X, by (means of) Y, in order to Z

Before we look at a root definition, we will consider the purpose of a system, i.e., the transformation that it might achieve. There are two key things to hold in mind here:

• The system is not a representation of the real world. Therefore there is an unlimited number of possible transformations that can be generated. These models will be more or less relevant and more or less useful in exploring the problem situation;

• A transformation is akin to a state transition; a system in one state is transformed into the system in another state (figure 6.4). For example, it is not a systemic transformation for a model to take in labour and materials and produce products.

Part of the spirit of systems thinking is in an opening out of a problem situation, the perception of a mess rather than a difficulty. To understand the role of the box office in the Barchester Playhouse it is likely that we will need to consider the box office environment. An understanding of the theatre may be necessary to gain insight into making change in the box office. What might be the purpose of a theatre? Some relevant transformations are developed in figure 6.5, where 'T' stands for transformation.

These are all possible systemic transformations that might be used to discuss the role of the theatre in Barchester. Some are obvious. Surely a theatre is there to entertain? Well, yes and no. Entertainment might be best served by taking off productions and turning it into a bingo hall. It might then be argued that the identity would change as the operation would no longer be recognizable as a theatre. What if entertainment is provided by putting on sexually explicit material? The organization might still be recognizable as a theatre but questions of who is entertained and what constitutes entertainment are raised, as are external constraints such as censorship. Although it is counter-intuitive perhaps to think of a theatre as a system to maintain the fabric of a building, the theatre happens to be a grade 1 listed building with a unique interior by a famous Victorian designer. Considerable insight into the theatre can be gained by creating a model that considers the theatre in this setting. For example, the historic nature of the building might constrain the ability to provide disabled access.

In figure 6.5 we consider primary task models. Typically, a primary task model reflects some real-world entity whose organized activity does indeed coincide with the notional human activity system described by the root definition. As figure 6.5 shows, there is an unlimited number of potential models that might be relevant to discussion about the purpose of the theatre. Each of those models can be sub-divided into a number of systems that together cooperate to achieve the transformation of the higher level of recursion. For example, a system model to entertain the local population might have a subsystem to allocate theatre time to different types of production, such as drama and comedy. An issue arising from this primary task activity might be disagreements concerning how much resource should be allocated to each type of production. Such a model formulation is known as an issue-based system. Further examples relevant to a theatre could include how to balance the competing demands of arts funding agencies and commercial sponsors, and how to ensure that disabled people in the local population are represented in the audience (figure 6.6).

A Weltanschauung (W) – or worldview – is what makes the transformations (T) in figures 6.5 and 6.6 meaningful. Why is it meaningful to entertain the local population? One W might be that it would improve quality of life. Another W might be that it demonstrates the good judgment of the Barchester Council in funding the Playhouse. It is, therefore, quite possible to associate a single transformation, e.g., local population entertained, with different Ws. The pairing of a T and a W form the heart of the CATWOE mnemonic (table 6.3).

In the CATWOE the Customer is the beneficiary or victim and should not be confused with the more everyday sense of customer as the purchaser or consumer of a product. If we posit a model of the theatre as an entertainment system then it is indeed possible that the Customer would be the local population and in this case there is congruence between the Customer of SSM and the customer of everyday usage. However, if we model the T as promoting tourism with a W that 'local businesses, such as restaurants and hotels, will benefit from the trade that results from the increased numbers of tourists visiting the town to see plays' then the CATWOE Customer is more appropriately local businesses than tourists. Tourists do have a role to play but they are not the direct beneficiary – to ascertain the beneficiary of a Transformation it is necessary to look at the T and W jointly.

The actors are those who would carry out the T. In the case of the theatre as entertainment system the Actors could be box office staff, theatre management, and tourists – the Actors could even be the stage actors! The Owners are those who could stop T. In an entertainment system this could be the management of the theatre. The environmental constraints are things that will be accepted as given. For example, the local council might put restrictions on the theatre with regard to safety or the police with respect to the laws of censorship. If we construct an entertainment model that allows productions to contain sexually explicit content or dangerous acts, then environmental constraints could be censorship laws and health and safety legislation. This exposes further Owners who could stop the T taking place – the local council and the police. If we shift the boundaries then environmental constraints can be brought inside the model. This might involve a subsystem to campaign for changes in the censorship laws in order to allow more explicit material to be presented in the interests of entertainment – however we might define it.

These fragmented examples can now be brought together into an example model that has a consistent and coherent root definition and CATWOE (table 6.4). The root definition (primary task) is as follows:

A theatre-owned and operated system to entertain the local population (X) by promoting and staging live theatrical productions (Y), in keeping with health and safety legislation and the laws of censorship, in order to improve the quality of life of the people of Barchester (Z).

A further example of a root definition is now given, but this time it is relevant to the box office level rather than the theatre as a whole. The root definition is formulated as follows (with accompanying CATWOE in table 6.5):

A senior management-owned system to shorten the queues at the box office (X) by the introduction of an Internet-enabled box office ticketing system (Y), subject to theatre-goers having Internet access and being willing to pay online, in order to increase the number of ticket sales (Z).

Conceptual models [4]

Now that the CATWOE has been produced we will model the purposeful activity using the minimum number of necessary activities to meet the requirements specified in the CATWOE (figure 6.7). Each of the activities is itself a system (the principle of recursion from chapter 5) and can be modeled

with its own root definition and CATWOE. Similarly, the whole model is itself just one activity in a wider system.

The arrows between subsystems in figure 6.7 represent logical dependencies. For example, it is not possible to 'schedule performances' until the productions to be staged have been decided. In this model it is allowable to promote a production in advance of the performance schedule being fixed.

The logical activity 'sell theatre seats' is preferred to 'sell tickets', which would be a rather physical representation of the purposeful activity; indeed, we might go further and replace the notion of ticket by a more abstract notion such as 'entry token'. This would help us to think about novel ways of communicating entry tokens to customers, such as by mobile phone.

In this model an appreciation of Health and Safety requirements and censorship constraints is needed when deciding which productions to stage; this appreciation is also needed when the performances take place. Monitoring and control is needed to regulate the system – is the Transformation being achieved? How well? Monitoring and control in the conceptual model is one of the characteristics of the systems approach identified in chapter 5. At the level of recursion shown in figure 6.7, monitoring and control are concerned with two of the three measures of success for a Transformation (in this case, the entertainment of the local population): efficiency and effectiveness. In a wider context there is a concern about whether the system contributes to the wider purpose, i.e., effectiveness. These three success factors are known as the Three Es (table 6.6).

The stated purpose of the system is to entertain the local population. Therefore, Efficacyanswers the question: 'is the local population entertained?' Efficiency is concerned with the amount of resource needed to achieve the output. The third E, Effectiveness, requires the wider context to be considered – is the expenditure of resource and effort worthwhile? The three Es for the Theatre system CATWOE that was defined in table 6.4 are shown in table 6.7.

According to the Weltanschauung in the CATWOE, it is meaningful to entertain the local population since it will contribute to the quality of life of the people of Barchester. The assessment of effectiveness requires consideration of the wider system in which the Theatre is embedded and awareness that

many other (possibly non-theatre) initiatives could contribute to the quality of life of the people of Barchester. In figure 6.8 there is an activity 'appreciate local population quality of life', which sits at a higher level of recursion from the entertainment system. Without a concern for the bigger picture it is possible for the Theatre system to be efficient in entertaining people, but to fail or only make a marginal contribution to the wider goal of improving the quality of life.

6.3.3 Back above the line: taking action in the problem situation

The conceptual models developed are not models of a real world theatre. The models help us to organize our thoughts about the problem situation in a systemic manner. It is therefore important to develop more than one conceptual model, where each model has its own root definition and CATWOE. Figures 6.5 and 6.6 show multiple Transformations that might be relevant to both primary-task and issue-based conceptual modeling in the theatre domain. If there is only one model then it will be very difficult for the SSM practitioner and others involved in the intervention to stop themselves from slipping into viewing the single model as a representation of the real world problem situation. Once this happens, then much of the value of the

SSM approach is in danger of being lost. For example, think about how the conceptual model might be structured if the Transformation were to educate the local population. What might be some of the Weltanschauungs that would give such a Transformation meaning?

Comparison of models with real world [5]

The rigorous, coherent and defensible conceptual models developed below the line can now be compared with the messy, real world problem situation. To support such an analysis, a matrix can be prepared listing both the activities and the logical dependencies in the model as rows. The columns are used to show whether the activity is done currently in the problem situation and if it is, then how it is done, how well it is done, and the criteria used to judge the activity (table 6.8). For example, how does a theatre know what constitutes entertainment for the local population? Perhaps it is just an assumption based on 'expert' artistic opinion, or possibly some market research might have been conducted.

Evaluating change [6] and taking action [7]

In hard systems thinking the rubric is that change must be systemically feasible and should be culturally feasible. This comes from a belief that the system model is a representation of a real world problem, that the situation can be addressed through problem solving. Cultural acceptance is nice to have but is of secondary concern to the systems engineering task at hand.

The soft systems view turns this on its head. The change should be systemically desirable, given the analysis conducted during the preceding stages of SSM. However, cultural feasibility is of paramount importance since if the change is not perceived to be meaningful by those involved in the problem situation then it will not succeed. The implication of this is that large scale change can be achieved if it is perceived as meaningful, while small-scale and seemingly insignificant changes will fail because they are not perceived as meaningful. Does this mean that there must be a consensus before the intervention can proceed? Not necessarily, but there needs to be an accommodation where people can agree on the ends, that is what can be said to constitute an improvement in the situation and then act in general accordance with the ends that have been agreed. In other words, there is give and take in which the parties are happy, to a greater or lesser extent, to go along with what has been agreed. Consensus is therefore best seen as a special case of accommodation, lying at the end of the spectrum where there are less demands on the affected parties to compromise as their ends and their understanding of what constitute reasonable means are congruent.

As a consequence of this there is much emphasis within SSM on the need to create shared understandings as the basis for change that will be perceived as meaningful in the problem situation. This is very much an exercise in the kind of organizational change that accompanies many information system development projects, but which tends to be overshadowed by the technical worldview.