AFH 007: Lessons from #NoEstimates Critics [PODCAST]

Hosts

Ryan Ripley, Zach Bonaker, Tim Ottinger

Discussion

We opened the discussion by defining the critic’s view against #NoEstimates:

Estimates are natural, ubiquitous, useful, and unavoidable in practical life and in business. Estimates are an important part of the process of collaboratively setting reasonable targets, goals, and commitments within an organization. The process of estimating, in and of itself, has by-products and benefits. Given that a rational estimating process is an integral part of making decisions in the presence of uncertainty, it is hard to understand why anyone would state that a desirable goal is to push forward into limiting estimates; down to zero where possible. –Peter Kretzman

From there the discussion flowed in multiple directions as we discuss the many areas of agreement that we have with some of the #NoEstimates critics:

  • The process of estimating, in and of itself, has by-products and benefits far beyond the sheer “number” or other indicator of sizing that emerges at the end.
  • When asking for estimates, management is acting benevolently and is looking to have a need met.
  • Abuse of estimates is poor management and a sign of dysfunction
  • The word “estimates”, as used in our debates, has been interpreted in a wide array of possibilities – from guesses to predictions.
  • It’s reasonable to feel a sense of shock here. It sounds like people are saying to “just stop giving estimates” and leave managers hanging.

Throughout the conversation we shared what we have learned from the interaction with the critics and worked on clarifying many of these areas:

  • Ambiguity around the #NoEstimates tag and the lack of civility demonstrated at times by those supporting NE has damaged the discussion and limited collaboration.
  • The connection between agile principles & values and #NoEstimates is not clear.
  • The role of systems thinking in the approach to minimizing the role of estimates is also not clear.
  • The field of “professional estimating” is highly complex, sophisticated, and dedicated to continuously improving the quantitative practice of estimating software development.

The remaining time was spent on systems thinking, pre-conditions necessary to question estimation processes and value, and the role of excellent engineering practices in reducing the role of estimates in a software delivery system. Then, we called it a night.

Resources, Plugs, and More

Ryan – http://agileanswerman.com

Zach – https://www.linkedin.com/in/zachbonaker

Tim – http://agileotter.blogspot.com/

Please note: I reserve the right to delete comments that are offensive or off-topic.

  • Pingback: Agile for Humans EP 12 - #NoEstimates | Agile Answer Man()

  • galleman

    The notion of a NE skeptic seems inverted.

    Estimating is the basis of decision making in the presence of uncertainty.

    Not estimating is skeptical. Critics of NE ask a simple question – can decisions about future outcomes in the presence of uncertainty be made without estimating (in some form). If the answer is Yes, then those conjecturing that need to address not only how, but how the gets around the microeconomics (opportunity costs), decision analysis (DA) principles, and the core principle of the underlying of all SWDev work – statistical variance – reducible and irreducible and the resulting variances. Since these variances are probabilistic they have not deterministic value – rather they’re random variables.

    NE to date has yet to address how to decide when the variables involved in the decision process – choices – are random without making estimates about their value needed to assess the choices.
    So rather than the Critics being classified as doubters. Invert the relationship.

    There are several misconceptions – professional estimating is Not highly complex. As one who practices in software intensive systems domain, the estimating process is straight forward, aided by tools and guidance.

    The notion that “systems thinking” (undefined) some how reduces the need to estimate has not evidence, it’s conjecture with any testable hypothesis.

    The original conjecture – decisions can be made without estimates – has yet to be tested outside personal anecdotes. Until that can standalone, #NE will remain a conjecture without a foundation in mathematics, decision analysis, or the principles of probabilistic decision making – all well developed and documented in many domains.

    NE appears to make a conjecture and then fails to back that conjecture with any principles or practices for making decisions in the presence of random variables of project work. Independent of ANY SWDev method, those random variables are still there. Connecting NE with agile does’nt remove those random behaviors.

  • Attila

    This podcast offends my sensibilities! It doesn’t link to my blog nor was I invited on as a guest. My opinion counts too! /s

    On a more serious note, I too would really prefer to have both critics and proponents be represented on a single podcast.

    • We’re open to another #NoEstimates show with skeptics included. Just need to find some willing to join us. Know any? Thanks for listening and for commenting! We truly do appreciate the feedback. 🙂

    • Paul Boos

      I’ll also point folks to the Agile Dialogs unconference (http://agiledialogs.org) being put together for these types of discussion – the first will be 13 November 2015 at the Arlington Technical Exchange in the DC Metro area.

      I do tire of the ‘debate’ aspect around the tag. When we make any decision, whether it be to estimate or not for example, we make it based on a set of assumptions. Dialog stems from elevating the assumptions of each side so that a more informed discussion can occur and perhaps each side will find uses in the approach taken on the other side. And perhaps the best outcome is a blend of the two.

      One example, when management asks for estimates, an assumption is they need to know the cost in order to develop a budget to fund the project. Once we have explored the assumption (and this is only one), we can ask ourselves, what other ways can we develop a budget? Under what conditions should we use these other methods over taking the time to estimate? If we are going to use estimation, how precise should our estimate be?

      If we decide that not estimating is a good method, what assumptions are we making? One I can see is that the value for what we are doing outweighs the cost of performing the work – value at risk as Glen Allemann would say (I think I paraphrased that correctly). This is important, perhaps we need a method that estimates value better rather than cost/schedule (where most of the focus is when not estimating). How do we quantify some elements of value that are aesthetic in nature – an example is the Sydney Opera house was way over budget and very late, yet can anyone think of the amount of value as a draw to Sydney and even all of Australia without it? If you used standard value estimation on number of seats sold per year for actual performances it would never capture that iconic value.

      Hmmm… perhaps this should have been its own blog post…

  • Curiously, and consistent with the general NoEstimates approach, this podcast focuses on the critical reaction to NE, yet (obviously quite intentionally) contains NO links or even name references to any of those critical reactions. This is a disservice to your listeners, and an odd choice.

    If you’re going to quote verbatim an 87-word excerpt, it’s common web courtesy to supply a link for your readers. Why wouldn’t you? Why don’t NoEstimates people want full visibility to both sides? Similarly, the lists of links they promulgate repeatedly as “articles on NoEstimates” (e.g., http://noestimates.org/blog/links/) cover only one side, THEIRS. As I wrote in a blog post on this, “As professionals, it’s basic table stakes that we be able to weigh all reasonable sides and perspectives in a business discussion. When I see reluctance or outright refusal to do so, I have two reactions: one, it must be a very weak argument indeed, to not be able to tolerate even linking to an opposing view; two, the argument’s advocates obviously don’t realize (or care) that these actions are reinforcing the sad stereotype of IT professionals as dogmatic, one-sided, theoretical rather than practical, disconnected from broader concerns, and unable to soberly weigh the business pros and cons of an issue.”

    I’d strongly recommend that you rectify this. It’s an easy edit to the show notes, if you want to.

    • Nick Zdunic

      I tend to be an advocate for NE but I do agree with you. Embracing a Skeptic helps with your own thinking and presents new information.

      • Thanks Nick for listening! I appreciate your follow-up on Twitter. Have a great night!