|
Elements of a Good Decision-Making Process
In this series of articles we are establishing a profile for what
is required of a process for good corporate decision-making. While
we are principally referring to collective decision-making at the
executive or management level, the processes, methods and techniques
are applicable for all personnel and at all levels of the corporate
chain.
We regard the following as essential properties of a good decision-making
process.
Consistent
Obviously, the process must be used consistently. But more than
that, it should guarantee logical consistency. Decision making requires
that we express preferences that do not allow situations where A
is preferred to B, B is preferred to C and C is preferred to A.
Being logically consistent is not a sufficient condition for excellence
in decision-making but it certainly is a necessary one.
Appropriately Rigorous
Rigour is the exactness and care with which we analyse issues. The
approach is selective because we ought to use our time and energy
efficiently and this means using appropriate levels of care where
it matters. Determining where it matters is part of the process.
Different parts of the decision will have different levels of impact
on the outcome. We need to understand these sensitivities ahead
of time so that we can apportion our efforts intelligently.
Consultative
No individual can claim to carry the totality of information required
to make good decisions in a complex environment. What is needed
is balance and diversity of views that can only arise from collective
experience. We therefore need to canvass a sufficiently wide range
of opinion in order to control the subjectivity that would otherwise
colour our decision. The depth of consultation should again be guided
by the degree of impact the issue is likely to have on the decision
outcome.
Documented
A good decision process ensures that collective experience becomes
collective memory so that future deliberations can benefit from
past ones. Therefore details, issues, debates, research, analysis
and opinions must be recorded and filed in a way that facilitates
later access.
Transparent
The process must be open and transparent, avoiding the potential
for mis-interpretations by those whose opinions do not eventually
prevail. The reasoning and evaluations that led to the conclusion
should be clear, placing the integrity of the process beyond suspicion.
Justifiability
A good process should facilitate the justification and defence of
the decision. It does this by being transparent, rigourous and consultative,
allowing it to be scrutinized with impunity.
Objective
The process should enhance objectivity by making use of recognized
standards and benchmarks wherever possible. This provides credibility
and objectivity.
Scaleable
A scaleable process can be applied to a variety of decisions both
large and small, simple and complex. The full force of the process
can be spared for cases where to use it would be overkill. However,
even in the context of small decisions, certain basic steps ought
to be taken such as to define the objective, identify criteria,
establish alternatives etc.
Utilised
There is no point having a carefully documented process that nobody
bothers to use. It should be a living, breathing mechanism whose
benefits are obvious and so becomes inseparable from the ordinary
operation of the organization.
The Analytic Hierarchy Process
Methods like the Analytic Hierarchy Process (AHP) exhibit may of
these qualities. AHP was developed in the seventies by Professor
Thomas Saaty in the United States. The process provides the means
of identifying and analyzing decision criteria, options and attributes
and quantifying their relative value. One of its strengths is that
it can synthesise data arising from many information collecting
instruments and methods. These can include
· Focus Groups Results
· Surveys
· Multiple Choice Questionnaires
· Decision Trees
· Decision Tables
· Expert Advice
· Pair-wise Comparisons
The results of these can always be processed in a way that informs
the hierarchy that lies at the heart of the method. We shall discuss
the AHP method in some detail in future articles. |