To make a decision, as much information as possible must be obtained about each of the alternative courses of action available, and about the possible consequences of each, in terms of the targets set. The data compiled acts as a primary resource, and if used correctly, the information it contains offers a clearer view of what actions have to be taken.
The course of action followed will produce new data, which should be added to the resources already gathered for analysis and interpretation, leading subsequently to a new decision, thus establishing a continuous feedback loop.
The process of selecting from among the various options available to be adopted in relation to any administrative function can be divided into eight stages:
- Identifying a problem: the first step needed in taking a decision is to have detected a difference between the current situation and the desired situation. This discrepancy, or problem, exerts pressure on the managing director, forcing him/her to take action, whether it be in such fields as company policy, deadlines, financial recession, or concerning future job evaluations, among other possibilities. For a situation to be considered a problem, the managing director must have the authority, the money, the information, and as many means as necessary in order to take action. If this is not the case, our expectations may prove unrealistic.
- Weighting the criteria: correct prioritization of the criteria selected in the previous stage - not all criteria will have the same importance when it comes to making the final decision. There is normally one key criterion, with the rest being weighed up against it or against each other.
- Considering the options: this means being capable of grasping and presenting all the feasible options that might successfully resolve the problem.
- Analysing the options: the person responsible for decision-making must study the proposed options in minute detail. The strengths and weaknesses of each one should become clear when compared against the criteria selected and ranked in stages two and three. While it is true that some assessments can come close to being objective, it must be clearly recognized that the majority of them are subjective because they are by nature value-judgments.
- Selecting an alternative: once all the options have been identified and put forward, and after the person responsible for making the decision has assessed them according to the hierarchy of criteria set out, the moment arrives when just one option must be selected: the best of the options put forward according to the procedure set out.
- Implementing the alternative option: with the selection process over, the implementation of the course of action decided upon now also becomes vitally important. The first thing that needs to be done in order to enact the decision is to communicate it to those affected and to ensure that they are committed to it. This is easier to achieve if those responsible for carrying out a decision participate in the decision-making process. Decisions like these are carried through to a conclusion by effective planning, organization and management.
- Assessing the effectiveness of the decision: finally, the results achieved as a result of the decision taken and the solution adopted must be assessed to see if they have corrected the problem. If, however, the problem still persists, the process will have to be looked at again to identify at which point any mistake was made and to then make a new decision: either dismissing the original idea completely, or making changes to one or more of the steps along the way.
Depending on the type of decision to be taken, models can be used to simplify an overly-complex situation. The aim is to highlight the most significant aspects of the situation to be analysed and, thanks to the model, to reach a better understanding and description of the reality it represents. Models can be classified as:
- Objective or subjective: when events cannot be expressed objectively and there are no formal models available with which to make a study, they have to be looked at in a less structured way, based on subjective views and intuition.
- Analytical or simulation: analytical models are used to reach solutions and must be resolved; simulations are simplified representations of reality, created in order to study the effects of different options.
- Static or dynamic: static models do not incorporate time as a variable, whereas in dynamic models, time is a fundamental parameter.
- Deterministic or probabilistic: in a deterministic model, all data are known with certainty. If this is not the case, then it is a probabilistic, random, or stochastic model.