Making Decisions – Vol. II: The Rational Model

     The rational model of decision-making feels familiar, intuitive, even obvious to most of us.  This is true despite the fact that few of us follow a well-defined process consistently.  Inconsistency in the process is reflected in poor decision quality, failure to achieve objectives, or undesired or unexpected outcomes.
     Versions of the rational model are available from various sources, though many do not identify the process by this name.  Ranging from four to eight steps, the description of each varying significantly, these sources offer a wide variety of perspectives on the classic sequential decision-making process.  Fundamentally, however, each is simply an interpretation of the rational model of decision-making.

     In an attempt to consolidate and clarify, a seven-step process will be defined here.  The aim is to provide a more serviceable explanation than is typically available, resulting in a clearer understanding of the process that encourages practical application and facilitates consistency.  The JayWink Solutions version of the rational model of decision-making is as follows:

Step 1:  Clearly define the situation.  Describe the problem or opportunity with which you are faced as succinctly as possible without omitting critical information.  Goals and objectives to be achieved and constraints on implementation, including outcomes to be avoided, should be defined.  Consider using “is/is not” statements to add clarity and prevent distraction by peripheral issues.

Step 2:  Define decision criteria and relative influence of each.  List all characteristics of potential solutions that will influence the decision.  For example, when choosing a component or raw-material supplier, cost, distance, delivery performance, and reject rate are all important factors; others could also be considered.
     When all relevant characteristics have been identified, rank the importance of each relative to the others.  The final result will be an ordinal ranking, from most-important to least-important, of the decision criteria.  It is important to establish criteria rankings early in the process to minimize the effects of bias in alternative development and selection.

Step 3:  Develop alternatives.  Potential solutions can be developed by refining previous strategies, modifying existing systems, brainstorming new approaches, or a combination of techniques.  Strive to strike a balance between gathering as many perspectives and alternatives as possible and expending an appropriate amount of resources.

Step 4:  Evaluate alternatives.  Rate each alternative developed in step 3 on each criterion identified in step 2.  Estimates, forecasts, or other forms of prediction may be necessary to assign values for comparison.  Qualitative or pseudo-quantitative comparison may also be needed, such as creating a Likert-type scale on which each alternative’s expected performance is rated.
     When concrete values are not available, substitutes must be generated as objectively as possible.  Allowing bias to influence intermediate steps will lead the process to a foregone conclusion – that which would have been reached without the pretense of a defined process.  The source of this bias is often a person in a position of authority seeking a predetermined outcome s/he views favorably.  Excessive bias at any stage can invalidate the process, causing practitioners to question its value and abandon it.
     A lack of concrete values also increases the risk of unexpected outcomes or failure to meet objectives.  While beyond the scope of this post, decision-makers should always be cognizant of the level of risk one is assuming.

Step 5:  Select alternative.  Compare alternatives according to the criteria rankings established in step 2.  If one alternative is found to be superior in all relevant characteristics, it is a “clear winner” or the dominant option.  If this is not the case, some judgment may be needed to finalize the selection.
     Consider the following scenario:  of four criteria, alternative A is rated best on the most important, while alternative B rates better on the remaining three.  Does alternative B’s higher rating on three criteria outweigh alternative A’s higher rating on the most important criterion?
     This determination cannot be made in the abstract.  Familiarity with the situation, the magnitude of the ranking preferences (cardinal ranking), the magnitude of each alternative’s “superiority” in each criterion, and other information is necessary to make an appropriate decision.  The example is provided to reinforce understanding that this process will not always “make the decision for you” by making the best choice obvious.  It will, however, structure the relevant information in such a way that enables a logical and defensible – that is, rational – decision to be made efficiently.

Step 6:  Implement the decision.  Some decisions may be implemented by completing a single action.  Others may require extensive planning and expenditure of resources, justifying the assignment of a project manager and use of the detailed procedures and record-keeping that a trained PM imposes on projects.  The application of judgment will guide implementation at a level of complexity appropriate to the situation and solution chosen.

Step 7:  Evaluate results.  Conduct a “lessons learned” exercise to improve future decision-making performance.  Consider all aspects of the process, including the individuals that were and were not involved, the amount of research and time dedicated to it, the effectiveness of brainstorming sessions, the pervasiveness of bias, and myriad other factors that influence the decision-making process.
     Also, review the outcomes generated by the decision, including the achievement of objectives, occurrence of unexpected or undesired outcomes, organizational support of the decision, and any other impacts related to the decision under review.  Correlate each outcome with the characteristics of the process that were most influential.  Identifying the connection between the process and the outcome will encourage repetition of activities that generated positive outcomes and facilitate improvement of less-successful activities and the process as a whole.
     Neglecting this final step is a “deadly sin” that permits team members to abandon structured processes and return to “shooting from the hip.”  The review helps practitioners recognize the causes of suboptimal performance and opportunities for improvement.  The ability to visualize – and effectuate – improving performance in both process execution and outcomes is the driving force behind consistent application of the model.
 
Return to “Why”
     The seven steps described above explain how to use the rational model, but only begin to explain why it is advantageous to do so.  To understand why one would choose to use the model, we will consider the consequences of an undefined decision-making process, including:

• Relevant information may be overlooked or undefined, leading to an uninformed (“random”) decision.
• Available information may not be thoroughly analyzed for accuracy, objectivity, and relevance.
• Extraneous information may not be filtered out, leading to an irrational decision.
• An individual decision-maker, or influential member of a group, may “anchor” on a predetermined solution, rejecting valid, but unquantified, arguments for competing alternatives.
• Unclear or undefined responsibilities may cause progress to stall as confusion about the status of the decision permeates the group.
• Instead of moving forward as a coordinated group, several people may each make an independent decision about the same situation, using different information and methods of analysis.
• Consensus may be impossible due to differing perceptions of the situation, objectives, constraints, or feasibility of proposed alternatives in the absence of shared definitions.
• No logical progression will be traceable to defend a decision or to learn from it.

 
In Times of Crisis
     It is a common refrain that standard procedures go “out the window” in times of crisis.  This does not have to be the case and it should not be.  In fact, it is in times of crisis that structure is needed most; emotions run high and decision-makers are more prone to irrationality.  It is true, however, that this process will be compressed in a number of ways.  Fewer people will be involved, less data analyzed, fewer alternatives developed, and to a lesser degree.  The decision criteria and ranking may exist exclusively in the mind of a single decision-maker who rapidly processes the available information to select an alternative.  Implementation may also be subject to less-rigorous planning and documentation requirements.
     The first six steps of the decision-making process may be abbreviated, but the seventh should not.  After the crisis, the “lessons learned” activity should take place as it does for other decisions.  Decision-making in time-critical and crisis situations can be improved by the same method used to review and improve routine decisions.
 
     The rational model is just one of many approaches to decision-making.  Future installments of “Making Decisions” will present other models, tools, and techniques to enhance your decision-making skillset.  Among others, accounting for the strength of preferences and addressing risk and uncertainty are anticipated topics of discussion.
 
     Send specific questions to JayWink via the Contact page or in the comments section below.  We look forward to assisting you and your organization.
 
     For a directory of “Making Decisions” volumes on “The Third Degree,” see “Vol. I:  Introduction and Terminology.”
 
References
[Link] “8 Steps to Making Better Business Decisions.”  Richard Lannon, BA Times, July 31, 2012.
[Link] “An Overview of Decision-Making Models.”  Hanh Vu, ToughNickel, February 23, 2019.
[Link] “The different decision-making models you need to know — and their pros and cons.”  Gazprom Marketing & Trading.
[Link] “7 Steps to Effective Decision Making.”  UMass Dartmouth.
 
Jody W. Phelps, MSc, PMP®, MBA
Principal Consultant
JayWink Solutions, LLC
jody@jaywink.com