How to fix a Broken Work-System

A bad process will beat a good person every time.
W. Edwards Deming

The results of an individual contributor only depend in part on the individual. The results are a function of both the individual and the system they are working in! Leaders are responsible for these work-systems. These include both the flow of information and the tools for work. So fixing the work-systems is a chief concern of leadership! The trouble is, it takes great rigor to effectively resolve complex organizational challenges. And the pace of feedback is much slower that when working on ‘the work’.

When you inspect a work-system, start like you’re debugging; Don’t Touch ANYTHING! Resist the temptation to start moving wires! Every code change you make risks changing outcomes, and masking where the bug in the program is. In much the same way, work-system bugs move whenever you change the system. Changing the system before understanding it only serves to hide the root problem. Instead, start by reviewing the system. Check your expectations. And if possible, take real measures of flow within the system! Just be careful not to hover. We all know that uncomfortable feeling when the boss is looking over your shoulder.

The next step is to identify where the problem is. You can use visual flow diagrams on a whiteboard to help this along. But there will be bottlenecks that limit your system’s effectiveness. You might see several. Now, the Theory of Constrains holds that there is one and only one bottleneck. The rest are merely local constraints. So take a spell to check the probability that the root you’re looking at is THE root, rather than a constraint. You may find that several symptoms emerge from one bottleneck.

Next comes the hard part: Change. Specifically not fixing everything all at once, but just one thing. Remember: Changing the system at all, means the other bottlenecks may move! Change like addressing the highest probability root cause in your system. As a result, all the performance characteristics of the system we knew before, will change. Some may change very little, others more so. If we are not disciplined about our changes, we risk wrecking the whole engine! Allow your system time to reach it’s new steady-state after making a process change. Resist the temptation to act on ‘existing data’. Chances are that data reflects a system that no longer exists… because you changed it!

When faced with a systemic problem, debug it like you would any other complex system. Inspect, and resist the urge to tinker! Identify the highest-probability root-cause, and work on fixing that. Once you do, step back and watch the system run. A little discipline in your problem solving will go a long way to avoiding unnecessary change in your organization.