Designing Processes That Don't Need Babysitting |
|
20 December 2025 - Erlenbach
Today I watched Antonia Scheidel's excellent talk on "Effortless Execution". She details falling into the trap of trying to automate engineering management and ending up creating processes that require more time than they save. I have been guity of this sin both in my professional and personal life. Fortunately Scheidel also details how she dug her way out of it. Being an engineer who loves process she, of course, created a process to analyze processes. Behold the Process Pentagon:
Scheidel's assertion is that, if a process doesn't have all five of these attributes, then it is likely to be abandoned or worse, take up more time than it saves. TPS reports come to mind... If a process is not transparent then its value won't be communicated to those using it and force will be required instead. If a process isn't socially reinforcing then some will be inclined to shirk using it to lay the work on someone else's doorstep. If a process is not worth it, then why do it? (e.g. it takes more time than it saves.) If a process is not self-healing then it will become some person's job to take on that role and waste their time. If a process is not minimally invasive then in the face of competing priorities for time it will get dropped. Analyzing processes under this rubric can lead a manager to realize how to fix (or drop) them.
