Unfinished Article: Optimization and Fragility
Have you ever found something you started writing but never finished? Here's something I started in October 2014. I've just rediscovered it in July 2020.
Stop unnecessary reviews! Eliminate redundancies! Break down barriers! Streamline processes! Trim the fat!
These are all great ways to raise efficiency, increase speed, and reduce cost... especially if your organization and processes have had a heavy "Prevent+Protect" culture for a long time. However, there comes a point where optimizing systems and processes further will actually harm the organization's performance. When such problems do arise, root cause analysis teams need to be looking for the signs of over-optimization.
Imagine a process that has many built in checks, reviews, handoffs, hold points, etc... all basically barriers intended to prevent or mitigate problems. Maybe the process was designed that way, or maybe it evolved over time. In either case, such processes often result in long leadtimes and large backlogs.
that the organization starts avoiding it or modifying it further by adding exceptions, bypasses, streamlined subprocesses, etc.
When a process gets too fat, it's probably high time for it to be re-engineered from top to bottom, input to output, supplier to customer... basically the whole value stream. This is a great opportunity to create massive efficiencies. Unfortunately, it's also a...
And that's where it ends! Where was I going with this?
I remember that I was thinking about resilience quite a lot back then. This reads like an intro to a discussion about lack of resilience brought about by eliminating things that seem redundant or unnecessary at the time... without considering how necessary they might be when the spit hits the fan.
And now, somehow, this stub of an article becomes linked to the present. I can hear a voice in my head, that voice that always turns my stomach: "I’m a business person. I don’t like having thousands of people around when you don’t need them. When we need them, we can get them back very quickly."
Oh really. And how's that working out?
by Bill Wilson