Causal Loop Diagram

Purpose

A Causal Loop Diagram (CLD) is a visual notation for mapping how variables in a system influence each other through feedback loops. By drawing these loops, teams see why a problem persists (reinforcing loops that amplify) or why the system resists change (balancing loops that stabilise). CLDs are essential for understanding complex organisational, market, or ecological dynamics.

At a glance

How to run

1. Define the focal variable (3 min): What behaviour or metric are you trying to understand? (e.g., employee turnover, customer churn, product quality).
2. List influencing variables (8 min): Brainstorm what affects the focal variable and what the focal variable affects.
3. Draw causal links (15 min): Connect variables with arrows. Mark each arrow: + means "same direction" (A goes up, B goes up), – means "opposite direction" (A goes up, B goes down).
4. Trace loops (8 min): Follow circular paths. Count the minus signs: even = Reinforcing (R), odd = Balancing (B). Name each loop.
5. Identify leverage points (4 min): Where could an intervention shift the loop's behaviour?
6. Debrief (2 min): What surprised the team? What does this change about the approach?

Materials needed

• Large whiteboard or digital canvas
• Sticky notes for variables
• Markers (two colours: one for + links, one for – links)
• CLD notation guide (printed handout)

Pitfalls & common mistakes

• Too many variables: Start with 5–8 and expand only if needed.
• Confusing correlation with causation: Arrows must represent causal influence, not mere correlation.
• Ignoring delays: Mark significant time delays with a double line on the arrow.
• No loop closure: If you have no closed loops, you are drawing a flow chart, not a CLD.

Inclusion & accessibility

• Provide a brief tutorial on CLD notation before the session.
• Use both colour and text labels (+ / –) on arrows for colour-blind participants.
• Allow participants to suggest variables in writing before the drawing phase.

Variations

• Stock and Flow Model: Adds accumulations (stocks) and rates (flows) for quantitative modelling.
• Behaviour Over Time Graph: Sketch how the focal variable changes over time before drawing the CLD.
• Multi-layer CLD: Separate CLDs for different system layers (operational, strategic, cultural).

When NOT to use this

• For simple, linear problems where cause and effect are obvious.
• When the team lacks basic systems thinking vocabulary — train first.
• When a quick decision is needed — CLDs are for understanding, not urgency.

Source & further reading

• Donella Meadows, *Thinking in Systems* (2008)
• John Sterman, *Business Dynamics* (2000)
• The Systems Thinker: Causal Loop Diagrams