This text was generated using AI and might contain mistakes. Found a mistake? Edit at GitHub
This text was generated using AI and might contain mistakes. Found a mistake? Edit at GitHub
Here’s a blog post based on the podcast transcript:
In software development, effort estimation has always been a controversial topic. Woody Zuill, an experienced software developer and advocate of the “No Estimates” approach, explains why estimates in software development often do more harm than good.
When you ask people what they understand by an estimate, about half mention the word “guessing” or similar terms like “prediction,” “assumption,” or “forecast.” At their core, estimates are conjectures about the time effort, costs, or scope of work that are meant to serve as a basis for planning and budgeting.
Estimates are mainly used to predict the future - when will something be finished, how much will it cost, how much can we achieve in a certain time? The problem: This “information” is not only inaccurate but often misleading. We make important decisions based on assumptions:
Zuill argues: “While estimates help us make decisions - a coin flip would do that too. We don’t need help with deciding per se, but with making GOOD decisions.”
The well-known Chaos Report by the Standish Group from 1994 showed alarming numbers:
Even 25 years later, these numbers have barely improved - despite better processes and shorter iterations. This suggests that the problem lies deeper.
Instead of relying on uncertain estimates, Zuill recommends:
“Becoming good at generating value is more important than being good at estimating,” says Zuill.
Software development is a complex, creative process of discovery - not a standardized manufacturing process. While efficiency and predictability make sense in mass production, these concepts don’t simply transfer to software development.
The “No Estimates” approach doesn’t mean categorically making no more estimates. Rather, it’s about questioning the prevailing system of effort-based decision-making and finding better alternatives.
Estimates give us a false sense of control in an inherently uncertain environment. Instead of improving predictability, we should develop our ability to deliver value quickly and learn from feedback.
As Eisenhower said: “Plans are worthless, but planning is everything.” The value lies not in the estimates themselves, but in our ability to make informed decisions when it matters.
The challenge is to bridge the gap between business requirements for predictability and the inherent uncertainty of creative knowledge work. The “No Estimates” approach offers important food for thought in this regard.