The hashtag #NoEstimates was originated by Woody Zuill for "exploring alternatives to estimates [of time, effort, cost] for making decisions in software development." What began as a conversation between a few people (Woody, Neil Killick, and Vasco Duarte) grew into a much larger discussion. As the year 2015 is now over, let's take a look back at what we learned from the discussion.
A lot of different people have participated, each bringing their own unique views and experiences. There is no single "#NoEstimates view"; what one person says doesn't represent the rest of the group. Some argue that estimates should be used more sparingly, others never at all, and yet others that suggest that only under certain circumstances.
Despite the differences, there seems to be consensus on the following key points:
A lot of different people have participated, each bringing their own unique views and experiences. There is no single "#NoEstimates view"; what one person says doesn't represent the rest of the group. Some argue that estimates should be used more sparingly, others never at all, and yet others that suggest that only under certain circumstances.
Despite the differences, there seems to be consensus on the following key points:
- Software project often fail. Perhaps the use (or misuse) of estimates play a role in this.
- Estimates are often turned into commitments. This mechanism is used to put additional pressure on developers, often with questionable results.
- Compromising product quality to meet scope and date commitments is usually a bad idea.
- Making small bets is better than big bets.
- Story slicing is a great way break down user stories/requirements.
Comments