There is always lots of complaining about the biases introduced into managing projects and making the estimates needed to make project decisions. In Kahnemann's Thinking Fast and Slow there are three biases. †
- Cognitive bias - a mistake in reasoning, evaluating, remembering, or other cognitive processes, often occurring as a result of holding onto one's preferences and beliefs regardless of contrary information.
- Optimism bias - a cognitive bias that causes a person to believe that they are at a lesser risk of experiencing a negative event compared to others.
- Ingroup bias - a pattern of favoring members of one's in-group over out-group members. This can be expressed in evaluation of others, in the allocation of resources, and other ways.
These errors in reasoning, evaluating, or recalling produce results different from rational judgment. Cognitive bias results from a systematic pattern of people making rapid decisions. Cognitive biases are part of our automatic system of how we think. Because of that, we don't separate the thinking processes into a dual-system ...
- System One - automatic
- System Two - deliberative
System One is an unconscious system that is automatic, fast, efficient at using the data or experience it already knows. Or - and this is a critical point - it thinks it knows. System One requires very little energy or attention. System One is what we operate in the majority of the time we are making decisions. Some research shows 95% of the time we are in System One.
System One is prone to errors, with little awareness of how the decisions made. System One is gut feel or intuition. When System One works it is because of memories and past experiences inform the decision. This lays the foundation for errors since our memories and past experiences are filtered through what we believe the situation to be. And of course, those are biased, filtered are fed back into System One at a later date, creating more bias to the decision-making process in the future.
Research shows [1] that optimism bias is a primary source of project failure. There is no one fix for optimism bias or any of the other biases on projects. But there are specific actionable steps that can be taken to expose these biases and prevent or correct the undesirable outcomes [2].
- Start with an awareness that the bias exists. Train the people working on the project about optimism bias and how that bias is exhibited in their behavior as well as their decision making. This training includes recognizing and applying the Principles of decision making in the presence of uncertainty. This means understanding the principles of probability and statistics and how these principles are applied to those decisions.
- When decisions are required in the presence of uncertainty, provide sufficient time to make the decision. This means putting a formal process in place for making that decision. A process based on tangible evidentiary materials.
How Can We Improve the Probability of Project Success?
The first step is to recognize there are immutable principles of managing in the presence of uncertainty.
These principles originate in:
- Managerial Finance - the branch of finance that concerns itself with the managerial significance of finance techniques. It is focused on assessment rather than technique.
- Probabilistic Decision Making - a fundamental statistical approach to the problem of pattern classification. Quantifies the tradeoffs between various classifications using probability and the costs that accompany such classifications. The decision problem is posed in probabilistic terms.
- Microeconomics of Decision Making - a process by which businesses make decisions is as complex as the processes which characterize consumer decision-making. The business draws upon microeconomic data to make a variety of critical choices – any one of which could mean the success or failure of their enterprise.
With these business principles of software development and projects in general, we can ask and answer five principles of project success
- What dos Done look like in units of measure meaningful to the decision makers?
- What is the Plan to reach done at the needed time for the needed budget, with the needed outcomes?
- What Resources will be needed to reach Done?
- What Impediments will be encountered along the way to Done and what are the handling strategies for each impediment?
- What are the measures of physical percent complete needed to inform the decision makers of progress to plan in order to take corrective or preventive actions to stay on plan?
A final Thought
Software engineering economics is about making decisions related to software engineering in a business context. The success of a software product, service, and solution depends on good business, financial, and risk management. Yet, in many companies and organizations, software business relationships to software development and engineering remain vague. [6]
† This post was inspired by "De-biasing project management," Joshua Ramirez
[1] "What Causes Cost Overrun in Transport Infrastructure Projects?" B. Flyvbjerg, H. Skamris & S. Buhl, Transport Reviews, 24, 1, 3-18.
[2] "The effect of fast and slow decisions on risk taking," M. Kirchler, D. Andersson, C. Bonn, M. Johannesson, D. Västfjäll, E. O. Sorensen, M. Stefan, G. Tinghog, and D. Andersson, Journal of Risk and Uncertainty, 54(1), June 2017.
[3] The Microeconomics of Decision Making in the Presence of Uncertainty
[4] Microeconomics and Risk Management in Decision Making for Software Development
[5] "Decisions in software development projects management. An exploratory study," Ricardo Colomo‒Palacios, Cristina Casado‒Lumbreras, Pedro Soto‒Acosta, and Angel Garcia Crespo, Behavior and Information Technology, 32(11):1‒9, January 2011.
[6] "Chapter 12: Software Engineering Economics," Software Engineering Body of Knowledge,
