Process Improvement is Trickier Than You Think - Commoncog

Most companies skimp on process improvement. But the surprising thing is that they do so not because they're bad or lazy — but because there are system dynamics that prevent them from doing so. We take a look at what those are.

This is a companion discussion topic for the original entry at

It strikes me that many of the organisational dynamics in this piece overlap pretty well with the Commoncog Guide to Burnout.

Management blames people, so they force them to work harder, and then at some point in the vicious downward ‘work harder’ cycle, job demands outstrip job resources, and everyone burns out massively.

1 Like

I enjoyed reading this and agree with your conclusions. As I thought about it, I realized one issue is learning to apply process improvement to yourself before applying it to other things.

It is just as “tricky” to implement process improvement in yourself due to time constraints and making excuses.


Glad to see you exploring this space! This is at the core of what I have been doing for a living for a long time. Your observations definitely match up with my experiences trying to improve processes within large organizations.

One thing I’ve learned trying to do this is the importance of making the size of the improvement smaller. A lot of the history of productivity improvement can be viewed as ways to reduce the time between action taken and result observed. This yields many benefits:

  • We get value from the effort faster
  • We reduce the amount of “noise” in the results that are driven by other factors
  • We reduce the number of people whose behavior has to change in order to get better results
  • We reduce the amount of resources at risk of being applied to something that will not yield better results
  • We can more quickly apply the lessons learned to the next improvement effort
  • When high-priority items come up after we have started, smaller efforts allow us to face less painful choices on whether to abandon the current effort or wait to start the new one
  • Smaller efforts are at less risk of triggering the organization’s immune system regarding changes
  • Smaller efforts are often more easily replicable if they turn out to be even more successful than we thought they would be

I look forward to seeing more on this topic!


@Roger Would it be okay if I shared your observations with others outside of this forum? The management of one mid-sized company I know would benefit from your insights.

1 Like

@cedric My business partner, Laura Black, introduced me to system dynamics some 20 years ago. I’ve learned from her the value of visual representations in developed a shared understanding of a complex system. Even “simple” business processes tend to be complex (i.e., characterized by interdependencies, feedback, delays, and noise). It’s no surprise that we often fail to agree on process improvements when we can’t even agree on what the process is.

1 Like

That should be fine, as I am not saying anything there that I don’t say in my day job anyway, but it doesn’t reach the level of specific guidance that people would pay for that would irk my employer.

1 Like

@dbayless @Roger I wonder what you guys think of John Cutler’s ‘better experiments’ tool, which he details here.

I quite like the visualisation he uses:

Where the basic idea is the more properties on the left your proposed experiment/change has, the safer it is to execute.

1 Like

As usual @cedric, you have a knack of finding excellent thinking I have not seen. This is a great resource. It is a bit surprising to not see “cheap <-> expensive” explicitly stated, but perhaps that is thought to be part of “small shift <-> large shift”.

One other related thought on that point is that since shorter interventions are often cheaper, yet another reason to favor them is that they reduce the pressure for returns. If we are spending a tiny amount of resource relative to our sponsor’s budget, they are more tolerant and patient than if a lot of chips have been put into the pot, so to speak.

1 Like

@cedric I like it!

As a rule of thumb, I characterize risk as being equal to ambiguity x “skin in the game”. If ambiguity is high and skin in the game is low, I’m facing a valuable real option. Cutler’s tool offers a quick way to gauge both parameters.

This has been a valuable thread, thank you both, Cedric and @Roger

1 Like