Clarity through modeling

Systems thinking become clearer once you have a credible model of the system you're investigating.

A model is a relation bewteen inputs and outputs, or a visual aid, that seeks to simplify and make introspectable a real system. A model is fake: it is not the real system. But a good model can close enough to the truth for your purposes such that you can perturb, query, and example it and gain insight into the real system you're modeling.

Althought the model won't be a perfect representation of the system you're seeking to understand,¹ you might be surprised how a model can make tangible what was previously a hazy concept, and provide a greater degree of intuition around how the system acts. It can be used to predict the response of a real system to a hypothetical set of inputs.

A sensitivity analysis of the model (i.e. change an input of the model and observe the modeled output) can help you understand the dynamics of the system, build intuition for the system, and gain some familiarity with the most important points of leverage or risks.

Models can be powerful arguments for or against something. A debate will be moved from the abstract to the concrete. They are tools for communication, and subsequently for decision making.

In my work, I've used simple linear models to make the case for investment in certain areas (e.g. ramping up hiring of a critical role), or to argue against the pursuit of a particular line of business. I've used novel visualizations to demonstrate how diseases are represented in various populations, suggesting care management capabilities to invest in. I've leveraged more complex models, predicting the behavior of the immune system in response to a foreign protein, enabling the production of (hopefully) effective therapeutics.

In all of these cases, a model enabled decision-making that wouldn't have been possible or well-advised otherwise.