Reading time: 7 minutes
Any experienced professional knows that time management represents a fundamental skill in their job. Being able to meet deadlines and plan activities is the key point to the management of one’s own energies and reducing stress. Yet, we often make predictions that are too optimistic about the time necessary to complete a task. Even more frequently, then, the same mistake is repeated, ignoring past experiences.
This phenomenon is called planning fallacy, the tendency to believe that a project will go exactly as planned, even if it is known that the vast majority of similar activities had required more time than expected. There is the tendency then to be too optimistic about the planning of the work.
A famous example of this bias is represented by the initial project for the Sidney Opera House. As a matter of fact, in 1957, it was estimated that the building would have completed in six years, at a cost of seven million Australian dollars. At the end of the construction works, a scaled-down version of the building was unveiled, in 1973, ten years after the first prediction, at a total cost of 102 million dollars. But, as we have mentioned, this cognitive error does not occur only for mega-projects, rather in daily activities too.
The underlying processes to wrong predictions
Kahneman and Tversky, which in 1979 have defined this phenomenon, suggest that people can use two different types of information when formulating a prediction. Who uses a singular information bases their prediction on specifical aspects of the activity that can influence the time necessary to complete it. On the contrary, who uses a distributive information takes into account how much time was required to complete similar activities. The former, the two researchers suggest, adopts an internal perspective, which leads them to fall victim to the planning fallacy more easily.
In addition, Kahneman and Tversky propose three obstacles to taking into account past events in the formulation of the prediction:
- The intrinsic focus on the future.
The act of prediction, by its definition, focuses the attention on future, rather than the past.
- The difficult definition of “similar” experiences.
In some cases, the experiences can seem so different that we are not able to compare them in a significant way.
- Attributional processes that give less relevance to the past.
Even if we are able to compare past experiences with the present, this information is not taken into account. This can happen because we look for a cause-effect relationship for the failure of previous predictions, or we excessively tend to deny the relevance of past experiences when they implicate negative consequences on the prediction. We deny the relevance of some experiences if they imply the fact that the project will require more time than we hoped.
In a study conducted in 1994 by the researchers Buelher, Griffin and Ross, the effects of this bias in the planning of some students’ activities have been presented.
The data emerged by their experiments are incredible. In a first test, less than one third of the participants (29.7%) completed the activity in the time they considered “the most accurate prediction”. In addition, only 10.8% completed the work before their most optimistic prediction, in other words if everything went as good as possible.
In another experiment, this phenomenon was reduced if the researchers suggested to the participants to consider past experiences, especially comparing them to the activities of the present. Without this suggestion, in fact, the participants tended to remember their past without comparing it with their actual work.
In the entirety of the five experiments conducted, less than half of the participants had concluded the activity before their initial prediction. As hypothesized, the predictions were based on past experiences that confirmed their optimism, tending to ignore negative experiences. In addition, the suggestion to remember the past led the students not to make balanced predictions, neither too optimistic nor too pessimistic. Instead, it made the participants fall to the other extreme, predicting too pessimistic deadlines.
How to limit this phenomenon
Comparing the past to the present represents the main strategy to limit the planning fallacy, but there are other techniques that allow to diminish its effect. Many studies have showed how the estimation of the duration of any event is tied to its size. Short stimuli are generally overestimated, while long ones are underestimated. For this reason, dividing a task in more sub-tasks can increase the time that is considered necessary to complete an activity. This phenomenon is called segmentation effect and it has been demonstrated how it is capable in reducing the planning fallacy. In addition, we tend to round off time in multiples of five, therefore we will have many sub-tasks by five or ten minutes. In this way, the predicted time will be much higher when compared to the formulation of the prediction based on the entire activity. This can lead to the opposite error, which is overestimating the necessary time to complete a job, but it can be considered less problematic than the underestimation, especially if the error is smaller in size.
Curiously, it seems that the planning fallacy disappears if we formulate a prediction on the activity of others. Therefore, when we want to ask for a deadline, it is recommended to ask a question such as:
“If another person had to do it, how much time would they require?”.
- Kahneman, D., & Tversky, A. (1982). Intuitive prediction: Biases and corrective procedures. In D. Kahneman, P. Slovic, & A. Tversky (Eds.), Judgment under Uncertainty: Heuristics and Biases (pp. 414-421). Cambridge: Cambridge University Press. doi:10.1017/CBO9780511809477.031
- Buehler, Roger & Griffin, Dale & Ross, Michael. (1994). Exploring the “Planning Fallacy”: Why People Underestimate Their Task Completion Times. Journal of Personality and Social Psychology. 67. 366-381. 10.1037/0022-35184.108.40.2066.
- Forsyth, D.K., Burt, C.D.B. Allocating time to future tasks: The effect of task segmentation on planning fallacy bias. Memory & Cognition 36, 791–798 (2008). https://doi.org/10.3758/MC.36.4.791