System Dynamics Stock-and-Flow Modelling

Description

Builds quantitative simulations of stocks (users, trust, backlog) and flows with feedback and delays to explore long-term dynamics of products and services.

Causal loop diagrams show feedback qualitatively; stock-and-flow models make it quantitative. Stocks accumulate (active users, content, support backlog, trust); flows change them (sign-ups, churn, resolution); auxiliary variables and delays connect them. Simulating the model reveals non-intuitive behaviour: overshoot and collapse, tipping points, worse-before-better dynamics after interventions. Sterman's work shows that people systematically misjudge stock-flow dynamics. Rigorous practice involves calibrating to historical data, testing extreme conditions, and sensitivity analysis. For UX strategy, it shows how growth tactics, support capacity and user trust interact over months and years.

Attributes

Loss type
Unmigrated
Stage
Strategy
Data
Mixed
Effort
Weeks

Steps

  1. Define the problem behaviour over time (reference mode), such as growth then stagnation.
  2. Elicit a causal loop diagram with stakeholders and data, then convert to stocks, flows and delays.
  3. Quantify relationships from data, research and expert estimates; build the model (Vensim, Stella, PySD).
  4. Validate: reproduce historical behaviour, test extreme conditions and dimensional consistency.
  5. Run policy experiments and sensitivity analysis on uncertain parameters.
  6. Communicate insights with simulation runs and use them to frame research questions.

Sources

  • Sterman, Business Dynamics: Systems Thinking and Modeling for a Complex World (2000)
  • Forrester, Industrial Dynamics (1961)
  • Meadows, Thinking in Systems (2008)

Also known as

  • System dynamics
  • Stock and flow diagrams
  • Forrester modelling

Categories

  • Strategy & Collaboration
  • Systems & Complexity