Pre-mortem analysis

Abstract

The concept of pre-mortem analysis consists in identifying the possible causes of failure of a project. The peculiarity is that the analysis takes place in the very early stages of the project, immediately after the conception of the idea. In fact, we all know the meaning of post-mortem analysis, which in medicine is the process that allows to identify the causes of the patient's death. The same happens with projects. After the failure of a plan, the team tries to understand what went wrong, what may have been the causes of the crash and what led to the "death" of the project. The results of this analysis will certainly be useful in the future because they will allow us not to repeat the same mistakes but one thing is certain: the project died. The time spent, the resources used and the efforts made are now lost. The purpose of the pre-mortem analysis is to prevent this. The basic concept is to assume that the project will fail and, starting from this assumption, to research the causes of death. This method requires negligible application costs and is also inexpensive in terms of time. Furthermore, it has been proven that a psychological factor can bring benefits not obtainable with the more classic post-mortem analysis. This tool is very useful in project management and allows you to have a greater awareness of the complexity of a project and to reduce the uncertainty factor that is often linked to new plans.

Relevance of the tool

It is important to understand why pre-mortem analysis is relevant in project management. The reasons are to be found in some concepts of social psychology, the so-called cognitive biases. Cognitive bias refers to a systematic (that is, nonrandom and, thus, predictable) deviation from rationality in judgment or decision-making. In other words, in project management, teams can be unconsciously influenced by certain factors that lead to a wrong perception of the project.

Relevant examples of cognitive biases that can influence decision making are:

• Confirmation bias: The tendency to selectively search for or interpret information in a way that confirms one's preconceptions or hypotheses.
• Conformity bias: choices of mass populations influence how we think, even if against independent personal judgments. This can result in poor decision making and lead to groupthink.
• Authority bias: favoring authority figure opinions ideas within innovation teams.
• Loss-aversion bias: once a decision has been made, sticking to it rather than taking risks due to the fear of losing what you gained in starting something and wishing to see it finished.

These phenomena occur in an unconscious and natural way when analyzing a project. In particular, these biases become more relevant when the project is at an advanced stage.

For example, let's say you've worked hard for six months on a project. The manager requests an update meeting and to ensure that the plan is progressing in line with the established objectives. During the discussion the manager proposes an idea that he believes can bring value to the project.

How many of those present would try to deeply analyze the idea to evaluate potential risks and how many would agree from the start trusting the manager's experience? (authority-bias).

As the meeting progresses, the team finds itself discussing a particular issue. Everyone has the same opinion except one member. It can happen that this person changes his mind conforming to the ideas of the other team members. This can happen either because it is influenced by "mass thinking" (conformity bias) or simply to avoid conflicts and maintain serenity within the team (groupthinking).

Once this situation has also been resolved, someone arises with a proposal: following some research it seems that a complete change of strategy can bring advantages to the project. Is it necessary to make a decision, continue the work done so far or change the plans? How many people would think of all the hard work they have done so far will be lost? And how many after all the efforts made are totally convinced that the initial approach was the right one? (loss-aversion bias and confirmation bias).

At this point the meeting ends, how many did not express all their ideas thinking they were silly, not relevant or for fear of being judged by the other members?

The peculiarity of the pre-mortem analysis consists in being carried out before the project is officially started. In this phase the influence of the cognitive biases listed above is strongly reduced or even canceled. Team members are asked to formulate hypotheses and any ideas are welcome and considered. Participants are encouraged to be proactive, the barriers dictated by the different roles in the company are eliminated. Everyone can apply their skills and experience by opening scenarios that others would not have considered. This allows the team to have a broader vision of the project, evaluate different types of possibilities and therefore reduce the probability of encountering unexpected situations in the future.

Problems such as planning fallacy (the tendency for individuals to underestimate the time required to complete a task ) can be prevented as it is very likely that the time constraint will be considered among the many causes. Moreover, of particular relevance is a research cited by Gary Klein himself in his article on pre-mortem analysis []. The study carried out in 1989 by experts from American universities showed that assuming that an event has already occured increases the ability to correctly identify reasons for future outcomes by 30%.[]

Downstream of all these considerations, the result obtained from the application of the tool is a clear reduction of overconfidence (The pervasive tendency for an individual to be more confident in his or her abilities or judgments than is justified ). in other words, awareness of the criticalities of the project increases. By doing so, it is possible to immediately evaluate adequate measures to prevent failure.

Application

A correct application of the pre-mortem analysis can be performed following five main steps:

1.Team Selection and Plan Understanding: The project manager selects the team members who will have to carry out the analysis. These do not necessarily have to be highly specialized, the important thing is that they are people involved in the realization of the project, from the conception, design, manufacturing or distribution phase. Then the manager proceeds by illustrating and explaining the project.

2.Declare failure: the manager declares that the project has failed. Based on the characteristics of the project and the objectives set, he can provide different time horizons (i.e. the project failed after three, six, twelve months of work)

3.Failure Hypothesis: At this point each team member is asked to list all possible causes of failure. this phase can take from 10 to 20 minutes and everyone works individually, noting as many hypotheses as possible. When everyone has finished, starting with the project manager, everyone reads one of their reasons and continues in the cycle until they are all read and transcribed in a place where everyone can read them.

4.Suggest fixes: now is the time to look for solutions in order to prevent each of the previously hypothesized causes of failure. First of all, priority is given to the causes that are considered most likely to occur and therefore riskier. It is important to take into consideration all hypotheses, even the strangest and, in the case of something that cannot be prevented (i.e. the destruction of the factory due to a natural catastrophe), for example, to declare that in any case there would be no resources. necessary to prevent the event from occurring.

5.Revise the plan: after discussing all possible countermeasures the plan is reviewed and the necessary measures are taken so that none of the causes of failure listed during the analysis can take place.

The result will be a more solid project with a limited level of uncertainty. Team members will also be prepared to deal with different types of contingencies and more aware of the importance of each stage of the plan. An experiment conducted by Beth Veinot, Gary A. Klein and Sterling Wiggins showed that the use of pre-mortem analysis compared to other risk analysis techniques decreases confidence in the initial phase, it makes team members more attentive to possible threats and more aware of the risks. Conversely, confidence increases in the next stage, when the causes and possible solutions to them have been discussed.

Limitations

The biggest limitation in applying this tool is the risk of spending so many resources such as time and money trying to prevent possible causes that may never actually happen.