The notion of a balanced discussion of the topic of Not estimating work on agile projects has yet to produce any principles on which this can occur. The #NoEstimates advocates have not provided the principles by which a decision about how to best spend the customers money in the presence of uncertainty of the outcomes (Microeconomics of decision making) without estimating the impact of the decision.
The one principle based suggestion is Slicing and is considered #Noestimates. But of course slicing is the basis of making an estimate by reducing the variances of the work into a narrow range. Then assessing the effort to produce similar work in the future. This depends on that future work being like the work in the past. This also assumes there are no emergent uncertainties that create risk to that work. Either naturally occurring uncertainties (aleatory) or event based uncertainties (epistemic).
As well, slicing is also standard practice in all credible Basis of Estimate processes used in all domains where a non-trivial amount of money is at risk.
Without a foundational set of principles to support the conjecture that decisions can be made without knowing the outcome of that decision, #Noestimates is a solution looking for a problem to solve.
And there are many problems, starting with Bad Management, a variety of biases, cooking the books for political reasons, lack of experience, skill, and knowledge in estimating, poor understanding of risk management, and numerous other root causes of project difficulties and many times downright failure because the knowledge of who much something will cost, when it will be ready for use, and what actually will be produced is not available to the decision makers.
I'm headed to Florida this week for a conference where Agile and Earned Value Management Systems are the topic. I'm reminded by a NASA Cost Director colleague there are three reasons projects get over budget, behind schedule, and don't produce the expected outcomes:
- We couldn't know- it's a science project and we're inventing new physics
- We didn't know - we didn't do our homework
- We don't want to know - if we knew, we would cancel the project before it starts
To date no evidence has been put forward to show how to make a decision when spending other peoples money in the presence of uncertainty without estimating the impact of that decision so an informed choice can be made on how to proceed.