I never wanted to doubt you
but I'd be better off without you
I'm no good at lookin' the other way
It wasn't what you said
it's what you didn't say
Mary Chapin Carpenter, "What You Didn't Say," Shooting Straight in the Dark
I find myself lately quoted in Forbes by the esteemed Jason Bloomberg, covering this year's ITIL debates. There have been some fireworks, no doubt.
Jason gives a good flavor for the controversies. However, I think more can be said about the challenges faced by the framework-based practices:
- Project Management with PMBOK/PRINCE2
- Enterprise Architecture with TOGAF
- Security and Governance with COBIT
- IT Service Management with ITIL
Many organizations base their IT operating model on some combination of this guidance. And many are finding this approach wanting, and challenged by Agile teams who have little patience for any of it.
Why the challenge? I think, per the title of this post, it's not what the guidance says. It's what it doesn't.
In general, the concepts promoted by each of the frameworks are conceptually OK - they mostly have their place. But what is missing? One word:
In my same post that Jason quotes, I observe:
ITIL requires an improved foundational model of IT delivery ... centrally concerned with execution, feedback, and flow.
That's what ITIL doesn't say. Nor the rest. And that is what is needed, as digital transformation accelerates.
Without throughput -- without flow -- none of it matters. It seems that our frameworks have lost sight of the forest, in their cultivation of their trees. (For the record, I said this ten years ago.) For them, it's all about governance, about slowing things down and "doing them right."
But how do we know we're doing the right thing, if it takes us months to deliver a first cut product? The need for that fast feedback trumps so much else.
Yes, change comes with risk. Yes, technical debt should be avoided, where possible. Yes, there are security threats. But we are developing new perspectives on all of those things. The risk of change is not reduced by slowing it down and adding more approvals. That is a placebo. Theater to pacify the uninformed.
We have to measure and understand throughput and the obstacles to it. That is the first priority.We need to quantify the cost of delay that any governance practice or control imposes on that value stream. And then, we reduce risk through fast feedback, automation, and anti-fragile techniques (e.g. the Simian Army).
Can the frameworks step up to this new world? Or will they fade, with their partisans mumbling in denial about "cowboys" and "anarchist millenials"? Saying "it's not the frameworks' fault" while their Centers of Excellence are de-funded and staff re-allocated to product-centric teams?
What do you think?