This website covers knowledge management, personal effectiveness, theory of constraints, amongst other topics. Opinions expressed here are strictly those of the owner, Jack Vinson, and those of the commenters.

High WIP - break out the whip

I'm thinking a bunch about environments where we have opportunities to multitask. And modern business is rife with them. Today, I'm somewhat inspired by a report out of the DevOps Enterprise Forum 2017 by Domenica DeGrandis, Alanna Brown, Courtney Kissler, and Lisa Scheinkopf, The Cornerstone for Winning: How to Get Strategic Alignment. I'm also inspired by conversations with colleagues.


It's easy to say things like "stop multitasking" or "reduce Work In Process (WIP)" or many other solutions that merely address the symptoms we see in the workplace. But why do these things persist? In the above paper, they draw a version of a vicious cycle that I have seen many times. I'll write it instead:

We have a lot of work going: projects, operations, initiatives, favors, etc. In most cases, the work from these different sources has many different priority signals: bosses, friends, priority ABC, priority red/yellow/green, priority 123. As a result, individuals have plenty to do and yet no clear signal on which one is most important. So they pick the one that seems most interesting, easiest, loudest screaming. And they do a lot of switching. In the end, there are sure to be delays in finishing the work which then leads to unmet business needs. And then? Rather than stepping back and giving people space to finish what they've started, here come new projects, projects to fix what has previously been broken, projects that start even earlier (to overcome the inevitable delays). And this generates more WIP and more conflicting priorities and more chaos.  People get tired. Mistakes get made. Things deliver later. And tempers flare.

This cycle continues around and around. At some point people get tired of the trouble and take actions to correct it, but it never seems to go away. Or the corrective actions work for a while and then the system seems to revert back to the way it's always been.

This happens because there is an inherent conflict inside the system: Do more and more stuff in order to satisfy market demand (for features, products, etc.). And yet on the other side is the idea that we should be working on fewer things in order to actually get them done. The conflict isn't between getting things done and satisfying the market - these things are both valid needs of the system and support a larger goal of meeting business needs.  The conflict comes in the way we decide to meet the needs.  And the big question: can one develop a solution that breaks the conflict and delivers both of the needs in full?  

The general approach is to examine the assumptions buried in the conflict. One big assumption is that people can absorb more work without significantly damaging their ability to deliver. But a few exercises on multitasking or investigating Little's Law or deeper reading (such as Factory Physics) can help dispel that concept. And this is the place where we can look to develop a solution. But the solution must also include clear mechanisms for setting priorities and ensuring that the WIP levels remain low in order for work to speed through the system. This is where the organizational leaders come into play: they set the priorities (there is only one #1 priority) and they work to ensure that priorities remain clear as the world continues to change.  And the people responsible for operations set the WIP limits and monitor the system to find improvements to enable more and more flow.

Obviously, there is more to it than this. The DevOps paper describes a bit more detail. And there are many additional sources of information and further help.

Synchronizing on practice and how to get out of bad habits

Why is multitasking so attractive?