Critical Thinking Tools and Techniques

Decision trees are one of the many tools that are used to fundament a decision and to ensure that each event outcome is properly analyzed and described. There are many definitions one can give for a decision tree, but basically it is a diagram with nodes and braches that "constitute a sequential path through a decision tree that reaches a final decision in the end"

Each path that one can take at a certain point is presented in a detailed form.

The diagram below describes the decision tree that is built when considering whether to launch a new product or consolidate the already exiting business. The decision tree generally starts with the decision that needs to be made, written in a square at the beginning of the tree. As we can see from the diagram below, there are two different paths of action we may decide to follow upon. Each has its own set of consequences and implications. The first branch would be a first decision, to launch a new product, while the second would be to consolidate the business. Each decision is written alongside the respective branch

We shall describe each path in part. If we decide to launch a new product, we will have two subsequent decisions to make. The new product can be produced either through thorough development, including beta testing and bug fixing (perhaps 6 months to a year to produce) or through rapid development, an initial launch not covering bug fixing and testing (less than 6 months). Each separate development decision will provide a certain market reaction, which we are interested in covering and describing, in order to be able to correct the eventual problems. In a decision tree, uncertain outcomes such as this one are represented through a circle. In this case, each type of development will result in a good market reaction (an increase in sales by up to 20%, for example), a moderate market reaction (an increase up to 10%) or a poor market reaction (no increase in sales or decrease, due to a decrease in customer confidence for the company's product). These outcomes are described on each outcome line in the decision tree, with the appropriate figures.

The second initial path is represented by the decision to consolidate. If we decide to consolidate, there are several courses of action related to the product portfolio that need to be considered. The example below mention two: reap products or strengthen products, by adding additional features, for example. Even if the example does not show it, each decision related to strengthening the product portfolio comprises several additional decisions. Strengthening the products can be done by adding additional features, by a powerful marketing and promotion campaign or by a testing period during which any problems can be solved. The outcomes are similar to the ones presented previously and consist of the market reaction to the measures applied.

The presentation of the courses of action the company may take, as well as the different outcomes that the decisions may bring about is followed in a decision tree analysis by the evaluation. Each circle, which represents the possible outcomes, should have an estimated probability that evaluates the odds with which that event will happen. Each outcome should have a numerical representation. Following the example I have presented, an increase in sales by 20% or by 10% (for moderate) will be the case here. Each uncertain node, representing the possible uncertain outcomes, will have a certain value that is calculated by multiplying each probability with each outcome and by adding them.

As such, in the evaluation phase of the decision tree analysis, we will have numerical data on each of the outcomes. If in the beginning, we would have had a probability that the outcome would be good or moderate, now we have a factual representation of this fact. If we consider the costs involved by the respective solution or respective path, then we will be able to compare the net profit that each decision is most likely to provide for the company.

After this definition of the decision tree and the accurate description of the processes and mechanisms that the decision tree implies, we are able to apply this technique to a Mental Research Institute (MRI) company that provides an alternative to traditional hospitals in terms of mental healthcare. The fact that the company provides an alternative to traditional hospital means that it needs to offer additional or better services. This it does, with state of the art technology, excellent service and excellent staff.

The decision we will be describing relies somewhat on the example provided previously. The company has decided that it needs to improve net profits in the next fiscal year. In order to do this, it has three decisions it needs to account for. The first one would be to (1) cut down costs. Cutting down costs would mean either (1) reducing purchases of technology, cutting down on staff's salaries and premium or (3) reducing overall administrative costs.

The second branch of the decision tree would be to increase revenues. This can be done either by (1) increase the prices that the clients pay, (2) increase the technological capabilities and the quality of service being offered or (3) increase the volume of clients (this would somewhat go hand in hand with the previous decision branch.