If you’re not familiar with the #NoEstimates “debate”, try searching Twitter for the #NoEstimates hash-tag. You have been forewarned.
It was probably silly to re-enter the fray, but I did…
Risk Management
Risk management, compliance, and governance are extremely important. Spending people’s money irresponsibly is fraudulent. When you’re paying for something, it helps to have some sense of what you’re getting, when you’re getting it, and how it will perform (even if you don’t have all the answers upfront). All of this is true.
Underperforming Organizations
Many software development organizations (and the organizations they serve) are not realizing their potential for value delivery. We’re not talking 8/10. No… they performing at 1/5th or 1/10th of their capabilities (or more). They do not deliver value continuously. Their practices are unsafe. And they fail to learn and experiment. I’m not one for the 10x engineer thing. But I do believe in (and have experienced, and measured) 10x organizations … and 1x organizations.
So you have risk management on one side. Important. And you have poorly performing organizations on the other side. Important.
Take a dysfunctional, under-performing organization and ask them what is important and they’re likely to talk about “accountability, predictability, hitting deadlines, and execution”. When you take a step back, what you often see is that the org is battling a local maximum. Yes, those things are symptoms of the problem, but the real problem runs deeper.
Chances are they’re battling a significant amount of organizational and technical debt. There’s a lack of safety, humanity, passion, engagement, and motivation. Sure, you can do a stellar job of risk management — and I truly admire people who can do this well — but it belies the fact that you are eking out a downhill 85mph run in an old Pinto, and there’s an 80% likelihood you’ll make it from California to New York in 57.7hrs provided you pee quickly and run the fuel tank to the red-zone.
The problem is that the exact same practices — in different contexts — can constitute either 1) humane, reasonable conversations and a hearty effort to help out the people involved, or 2) theatrics, hand-offs, command-and-manipulate, and pass-the-buck. The exact same practices. One org’s governance and compliance is a source of competitive advantage and employee pride. Another org’s governance and compliance is their achilles heel, and will lead to their disruption.
From my perspective, the #NoEstimates crew is focusing on the 80% of orgs that have far bigger issues than estimates. They are trying to get a heartbeat going … some flow, some focus, some breathing room, and some / any value delivered. They are asking “how do you go from a 2x org to an 8x org?” and stripping away the excess to answer that question. Trying to define estimate is a rabbit hole in this particular conversation, and is almost beside the point.
#NoEstimates detractors find this utterly irresponsible, as they are risk management / project management pros and they live/breath this stuff. If I needed a guidance system for my missile developed, I’d go to a #NoEstimates detractor. If I was looking to turn around my company early in the game, I’d go to a #NoEstimates advocate. Manage an outside contractor or assist my teams with project work (vs. product work)… #NoEstimates detractor. We are talking about two fundamentally different problems that happen to share the word “estimates” occasionally.
The sad thing is that there are many skilled folks on both sides (of this accidental, word inspired debate) that have a ton of great information to share. And they do share! They contribute to the community in their respective domains. But when I hear from people who are too afraid to join the discussion for fear of being beaten down and harassed … then we have a problem. It’s not productive and/or effective.
Sad really. I’d love to learn more about risk management, continuos value delivery, and defeating the local maxima in under-performing orgs.