Strategic Assumption Surfacing and Testing (SAST)

Mitroff’s ‘operationalization’ of Churchman’s systems approach, part 1

In table 12-3 (p. 301 of Mason’s and Mitroff’s ‘Challenging strategic planning assumptions’) major approaches to business problem solving are compared, including the systems approach and SAST (strategic assumption surfacing and testing), but also analytic modelling (typical of operations research), the case method (widely used, but lacking in objectivity), structured approaches (e.g. PIMS and its many derivatives, often failing to look at key non-quantifiables). The problem with the systems approach is that it is difficult to operationalize (although it could be argued that Wicked Solutions solved that problem). The problem of SAST may be the unwillingness of participants to lay bare their assumptions. This is a general problem in all approaches where we want to leave no stone unturned (as assumptions, e.g. about people’s motivations, lurk beneath them). In this post I will argue that SAST can be combined with the systems approach to improve both. By the way, Mitroff and Barabba wrote a 2014 update of “challenging strategic planning assumptions”, which was entitled “Business strategies for a messy world”).

[This post has been syndicated by The Systems Community of Inquiry to, the global network of systems thinkers, scientists and practitioners.]

Organized complexity     Mitroff and Mason start their book by describing something they call “organized complexity”. “Organization is usually considered the route to the solution of a complex problem. In reality, however, organization in complexity can become an insurmountable barrier to the solution of a problem” (p. 5). “Three factors – separability, reducibility and one-dimensional goal structure – mean that simple problems can be bounded, managed, and as Horst Rittel (1972) put it, ‘tamed.” Rittel was the first to identify and describe the ‘wicked’ problem in the mid-1960s. Mitroff argues that in the past complexity was much more disorganized and could be managed statistically. However, people, organizations, and facilities have become more and more tightly woven, increasing the risk of disastrous crises. This line of thought explains Mitroff’s later and continuing preoccupation with crisis management (see previous post). Mitroff and Mason add 6 characteristics for “wicked problems of organized complexity” to the 10 identified earlier by Rittel (see elsewhere in this blog): (1) interconnectedness; (2) complicatedness; (3) uncertainty; (4) ambiguity; (5) conflict; and (6) societal constraints. One fine day I will combine all 16 factors in a single concept map. To deal with these challenges, new problem-solving methods are required that are (a) participative; (b) adversarial; (c) integrative; and (d) managerial mind supporting. The last requirement means that managers and policymakers need to achieve a better insight into the nature of the complexity. This also means that managers and policymakers must be deeply involved in the process.

The SAST planning process      In chapter 3 (p. 35-57) Mason and Mitroff describe the essentials of SAST. At the end of it the management or policymaking team is “well informed about the strategy they are following and the assumptions that support it.” They have sufficient reason to believe that they have constructed an effective (model of) reality from which to proceed. A decision with regard to a plan or policy has been made. Key assumptions were not ignored but rather surfaced, challenged, and monitored over time. This is the purpose of the SAST process. In the example the SAST process involved five steps: (1) group formation; (2) assumption surfacing; (3) dialectically debating and ranking these assumptions; (4) further debating and results and negotiating the statements of assumptions; and (5) synthesizing results, arriving at a consensus, establish information requirements and guidelines for the final decision. I created a little concept map to summarize the steps, which I will very briefly discuss one by one, with some references to the systems approach of C. West Churchman, on both of which I have posted a lot earlier in this blog.

Group formation     No single mind is able to adequately grasp complex problems fully, so key stakeholders must be swept in to benefit of their insights. A single group of stakeholders is unlikely to develop the dynamics necessary to accommodate complexity, so multiple groups are needed (this is perhaps the most obvious way of unpacking Ashby’s Law of Requisite Variety, see also VSM or the CDE model). Mason and Mitroff propose a statistical clustering technique called MAPS (Kilmann’s Multivariate Analysis, Participation and Structure) to cluster people into several groups (minimum 2, maximum ??) based on their affinity for tasks and for co-workers with a maximum number of 6-8 people per group to find the best possible combination of intra-group and inter-group effectiveness. Other approaches can be used, e.g. based on the personality types of Carl Jung. The groups are further assigned basic strategic orientations and asked to adopt a name to reflect their orientation. The idea of multiple groups reflects Churchman’s principles of deception-perception. Churchman was early in recognizing the value of Jung’s personality theory to business planners and planning.

Assumption surfacing        Each group is asked to surface as many assumptions as possible. One of the best ways to do that is first to identify as many stakeholders as possible. The next step is to sort out the assumptions to find the 6 or 8 most important ones. The group members do that – in a dialectical debate – in several substeps, including: (a) winnowing out those that are irrelevant to the problem or topic or issue at hand; (b) testing the assumptions for certainty and importance. Mason and Mitroff suggest plotting an importance/certainty Assumption Graph, e.g. using Tom Saaty’s Analytic Hierarchy Process. In Churchman-inspired Wicked Solutions the assumption surfacing uses rich picturing, while the testing for relevance and importance is conducted in two stakeholder tables. What could be said to be lacking is testing for certainty, but this is dealt with during the debate. Preliminary agreement is found by way of ‘framings’, which serve as a basis for outlining an ideal situation, which in turn is used to surface more assumptions using Churchman’s dialectical heuristic (which is similar to Ulrich’s critical heuristic, because the latter is based on the former). Part of Churchman’s heuristic is the mapping of clients and decision-makers, which by and large overlaps with the stakeholder listing of Mason and Mitroff. The underlying principles and general approach are the same. This shows that the choice of method is a matter of design and suiting the characteristics of the planning problem or planning group. The two methods could also be combined in a variety of ways.

Stakeholding development     In the next step the debate is carried to all the groups. One of the steps is to engage in an assumption negotiation process. This involves proposing alternative assumptions to resolve controversies at large. At the end an intermediate synthesis is formulated with two types of assumptions: those over which consensus is reached and those that present an issue for further investigation. In the first edition of Wicked Solutions an effort was made to include stakeholding development as part of the method. In the second this idea was dropped. It seems worth considering to reinstate stakeholding development using elements suggested in ‘Challenging strategic planning assumptions’, especially those related to stakeholding development, information requirements and the argumentations scheme.

Information requirements analysis      “Each issue and key assumption [is] subjected to further analysis in order to adduce the data and warrants that underlie its claim. Where the available data [is] inadequate, business intelligence and management information systems activities [are] undertaken to acquire the specific data necessary to resolve the strategic issue.” (p. 53). “Brainstorming is one of the most effective ways to identify possible information-producing activities for each issue.” Most information activities fall into three approaches: (1) research approaches, e.g. market research, operations research, financial analysis, competitive analysis, PIMS; (2) judgmental approaches (dialogue and reflection to assess values , beliefs and judgmental dimensions), e.g. position papers, delphi panels, focus groups, scenario writing; and (3) monitoring approaches, e.g. using key performance indicators, social indicators, forecasting sources.

Final debate    “A final debate is held to ferret out additional implications, insights, and understanding. A judgment is made on the best set of assumptions from which to proceed. Finally, an appropriate policy is chosen based on the new information and the synthesis that emerged.” In the end the management or policy-making team is well informed about the strategy they are following and the assumptions that support it. A decision is made. Key assumptions were not ignored but surfaced and critically or dialectically debated.

Policy argumentation      Claims are the supporting foundation of any policy, plan, or strategy. Every assumption that underlies a policy is a claim. Once a policy claim is articulated it is proper, even mandatory, to call it into question, which in turn calls for an argument, resulting in argument chains. These are anchored at the bottom by two things: facts and judgments. Policy argumentation is a conceptual framework for dealing with claims, facts and judgments in a dynamic, innovative policy-making environment, thus bringing order out of the chaos. The SAST methodology and the Toulmin argumentation scheme are linked. Any policy statement is based on a series of assumptions made about stakeholders. One implicit warrant in any major policy argument is that all the stakeholders are properly identified and their assumptions properly specified. The backing for this warrant is the theory of teleological systems (the systems approach) and that the SAST method has been properly conducted. More about this in a next post.

The road to hell       … is paved with wrong assumptions. The original proverb is of course “the road to hell is paved with good intentions.” But there are two major links with SAST. First they mask ulterior motives, which SAST seeks to surface. And secondly good intentions often have unintended consequences, which SAST (and the systems approach, and therefore Wicked Solutions) seeks to to bring to light and avoid as well as possible.

About csl4d
This entry was posted in General. Bookmark the permalink.

1 Response to Strategic Assumption Surfacing and Testing (SAST)

  1. antlerboy - Benjamin P Taylor says:

    Reblogged this on Systems Community of Inquiry.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s