Lets Get Out Of The Prediction Business
Consider the photo above. Did you really have to think and process it and determine what you were looking at this Monday morning? Hold that thought for a few minutes.
I work with some really gifted professionals in product development and software engineering. We use the best practices available to us and try to do the impossible, or so it seems. We work in a ruthlessly competitive environment where Teams are asked to commit to delivering features to market hoping customers haven't changed their minds and are still willing to pay us.
To do that we need to work to a schedule. Predicting when you are ready to release a finished product to the market of isn't just very hard, it's mostly impossible.
This problem isn't just limited to product development. This is a universal challenge resulting in poor decisions in hiring and firing too. Ouch! That's going to resonate on a social media platform.
Whether we care to admit is or not, we are not very good at probability. Our ability to estimate outcomes and make decisions is really unreliable. Brilliant psychologists and cognitive science specialists who understand how we think have proven our fallibility.
Bottom Line
We suck at estimating and our decision making process is flawed. Deeply. We are human so we can't really help it though. So why do we do it? What are the alternatives?
It is very difficult for us to override our intuition, even if you may be an expert in your field, especially if you are an expert.
We might have very well exceeded the 10,000 hours milestone that Malcolm Gladwell shared in his very readable and compelling book "Outliers: The Story of Success," where he notes you can become very good at something once you've met that requirement.
But that achievement and expertise still won't matter because you brain just doesn't function the way you think it does.
Dr. Daniel Kahneman, Professor of Psychology and Public Affairs Emeritus at Princeton University, author of "Thinking Fast and Slow," explains in that marvelous book that our intuition, what he calls System 1, is an incredibly fast and powerful part of our brain that causes us to believe we know what we really don't know.
By the way, Dr, Kahneman won a Nobel Prize in 2002 for human judgement and decision making and practically invented Behavioral Economics so while you may never have never heard of him, just go with the fact that he is on to something!
Danny Kahneman spent a lifetime of research in psychology, much of it with a brilliant partner, Amos Tversky, a cognitive and mathematical psychologist. Maybe you've heard of their published study on the psychology of prediction and the judgement heuristic known as representativeness? If you missed that it is understandable as it was published in 1973.
Consider looking into my theme because for any recruiter or hiring manager who uses LinkedIn and used their intuition today to make a decision on a candidate, you may have made a mistake. In fact, you probably did. You have a cognitive bias that causes you to overestimate your ability to predict accurately.
Even if you stop right now to consider my nifty little article's theme, you won't reconsider your decision. Kahneman and Tversky understood that in 1973 and called that the illusion of validity.
Take some time to look at a video where Danny Kahneman illustrates several important Natural Assessments.
If you've read this far, try experimenting with a systematic approach to making your next important decision. For a backup, use a Simple Combinations of Variables to keep that brilliant intuition of yours honest.
About the picture, I am recommending you start paying attention to the forest and the trees.
Senior Advisor
7 年Agree Jim. It is necessary to have dates to targe release and deploy and coordinate the "non-functional/enabling "features" needed to go to market. Estimating via Planning Poker and like methods are just flawed. Anchoring, Framing and flawed intuitive decision making sucks and isn't reliable. A "simple model" that you constantly improve and update based on simple and relevant variables works better. In The Undoing Project by Michael Lewis, he discusses how Daryl Morey of the Houston Rockets built a model and maintained it carefully to predict NBA draft choices better than humans have ever done. Like Moneyball for baseball. As I mentioned in my article and provided a link to a video of a Kahneman lecture, simple models are better predictors, if you "inspect and adapt" the variables frequently.
Learning-Oriented and Results-Driven Product Leader | Strategist | Technical
7 年so what's the alternative?