🌱 Example of a system with feedback loops

A business analyst might be asked to look into a sluggish purchase requisition process with the following symptoms:

On the surface, this may like a simple efficiency problem and might traditionally be addressed with solutions such as “streamline the approvals” or “automate the form.” But systems thinking digs deeper, asking:

"Why is the process behaving this way? What loops and incentives are in place? What is being reinforced or stabilised?"

This approach might reveal:

1 A Balancing Loop Protecting Against Overspending

This is a stabilising loop (also called a balancing loop). Its job is to prevent excessive or unnecessary spending:

This loop may not be consciously implemented but it has a desired purpose - it slows spending to protect the budget. So the sluggishness is not random - it’s stabilising the amount of spending.

2 **A Reinforcing Loop

This is a positive feedback loop, but in a destructive direction:

  1. Delays frustrate users who need to buy things.
  2. So they develop workarounds: buying without approval, using personal cards, inflating requests to get through faster, etc.
  3. These workarounds erode trust - Finance or management see staff circumventing rules.
  4. In response, they add more controls - more documentation, extra sign-offs, system restrictions.
  5. These new controls make the process even slower.
  6. The cycle reinforces itself.

The “sluggishness” of the process isn’t just bad design. It’s the result of:

If you only fix the surface (e.g. by speeding up the form), you may undermine the balancing loop (risking overspending) and / or ignore the root cause of the reinforcing loop (mistrust, misaligned incentives).

Instead, redesign the process to preserve the stabilising function (e.g. through real-time budget visibility) andreak the reinforcing loop (e.g. through better communication, delegation, or trust-building)