Systems Thinking Models
Much of the way we have been taught to think is in parts - we break things down into more easily understandable chunks. But, as Aristotle would say, the whole is more than the sum of its parts.
According to https://youtu.be/VUXeQGsVbqU?si=jP-b-vHEu84RUjjB, there are six mental models that will help us see problems as a whole.
1. Non-linear organisation
We tend to be most familiar with linear thinking: A leads to B which leads to C. A sequence. But in reality, the bigger picture of the world is not like that - things are cyclical or a web of cause and effect. Things are interconnected. And this is important, because if we don't realise this then we can get unintended consequences of changing any of the elements.
2. Stocks and Flows
- Stocks β things you can count (e.g. cash, customers, inventory).
- Flows β actions or events that change the stocks (e.g. purchases, refunds, leads).
3. π± The iceberg model
Four levels of realities:
Level 1: Events β What happened? (e.g. "Sales dropped last month")
Level 2: Patterns/Trends β What keeps happening over time?
Level 3: Systemic Structures β What processes or relationships drive the pattern?
Level 4: Mental Models β What beliefs or assumptions shape those structures?
We tend to think at the level of events, but this can lead us to treating symptoms and not the causes.
4. Bottleneck
The system is only as strong as its weakest part. Bottlenecks are where value gets delayed or lost
Look for:
- High delays.
- Overloaded teams.
- Recurring complaints.
The 80 20 rule might be applicable here to focus on the most limiting factor first..
5. Second order thinking
Most people think in first order - if A, then B. But second order thinking considers the implications beyond this - if I do A, what else might happen because B happened?
Before acting, ask:
- What are all the likely consequences?
- What donβt I know?
- How confident am I in this prediction?
6. Feedback loop
Feedback loops allow a system to self-regulate or evolve.
Two types:
- Reinforcing loop: action amplifies further action (e.g. positive reviews β more customers β more reviews).
- Balancing loop: action limits itself (e.g. sales rise β stock falls β delivery delays β sales drop).