Alfred Korzybski's famous observation -- "the map is not the territory" -- captures one of the most important truths about decision-making. Every model, framework, metric, and mental map is a simplification of reality. The moment we forget this, we begin making decisions about our map rather than about the territory it represents.
Why All Models Are Wrong
Models are simplifications by design. They omit information to make reality tractable. A road map omits elevation, vegetation, building heights, and countless other features because including them would make the map unusable for navigation.
The same is true for decision models. A financial model omits human behavior, competitive response, and tail risks because including them would make the model unmanageable. A strategic framework omits operational complexity, timing uncertainty, and interpersonal dynamics for the same reason.
These omissions are not bugs -- they are features. A model that captured every detail of reality would be as complex as reality itself and therefore useless. The decision-making scenarios at KeepRule help you develop intuition for when models help and when they mislead.
The Three Dangers of Map-Territory Confusion
Misplaced precision: Models generate specific numbers that feel precise. A financial forecast predicting revenue of 3,247,891 dollars looks more trustworthy than "roughly 3 million" -- but the false precision masks enormous uncertainty. The core principles emphasize expressing uncertainty honestly rather than hiding it behind precise-looking numbers.
Excluded middle: Models define categories with clear boundaries. In reality, most things exist on a spectrum. When your model says a customer is "high value" or "low value," the reality is that customer value is continuous and context-dependent.
Reification: We begin treating the model's abstractions as real things. "Market segments" are useful fictions that help us think about diverse customers. They are not actual groups of people who behave consistently within their assigned segment.
Using Models Without Being Used by Them
Hold models lightly: Use models as thinking tools, not truth claims. A model that helps you think about a problem is valuable even if (especially if) it is not perfectly accurate.
Use multiple models: No single model captures reality. Using multiple models, each highlighting different aspects of the territory, gives a more complete picture. The decision masters were notable for their ability to switch between models fluidly.
Test models against reality: Periodically check whether your model's predictions match actual outcomes. When they diverge, update the model -- do not explain away the divergence.
Know what is excluded: For every model you use, explicitly identify what it leaves out. These exclusions are your blind spots, and they are where surprises come from.
Prefer simpler models: All else being equal, simpler models are better. They have fewer parameters to go wrong, they are easier to understand, and they make their assumptions more transparent. Complexity in a model is a cost, not a benefit.
Common Map-Territory Confusions
Financial models: Treating a spreadsheet projection as a prediction rather than a structured guess. The spreadsheet is a map of possible financial futures, not the future itself.
Organizational charts: Treating the org chart as a description of how information flows and decisions are made. The actual network of relationships, influence, and communication rarely matches the formal structure.
Strategic plans: Treating a strategic plan as a commitment to a specific future rather than a framework for navigating uncertainty. Plans are maps; execution happens in the territory.
Performance metrics: Treating KPIs as direct measures of performance rather than imperfect proxies. The metric is a map of performance, and it always omits something important.
The Meta-Model Problem
The "map is not the territory" principle is itself a model -- a map of the relationship between models and reality. It too simplifies. The question is not whether to use models but how to use them wisely.
The wisest approach is pragmatic: use models when they help, discard them when they do not, and never forget that the map is an aid to navigating the territory, not a substitute for understanding it.
For more on mental models and decision frameworks, visit the KeepRule blog and FAQ.
Top comments (0)