── ── Decision-making
Map Is Not the Territory
Map is not the territory, a principle from Alfred Korzybski, holds that every model, metric, or representation is a simplified and aging abstraction of reality, never reality itself. Maps are useful precisely because they leave things out, but treating the map as the territory means acting on a picture that is incomplete, out of date, and shaped by whoever drew it.
How it works
Every dashboard, financial model, persona, and strategy deck is a compression of a far richer reality, made by discarding detail. That compression is what makes it useful and what makes it dangerous: the things left off the map are invisible, and the map keeps looking current long after the territory has changed.
Use maps without mistaking them for the ground. Ask what this model leaves out, who drew it and why, and how old its assumptions are. When the metric and direct reality disagree, trust reality and fix the map, rather than optimizing the number until it no longer means anything.
When to use it
- Interrogating a metric or KPI before steering the company by it
- Pressure-testing a financial model or forecast's hidden assumptions
- Noticing when a customer persona has drifted from real customers
- Avoiding optimizing a proxy metric at the expense of the real goal
When not to use it
When you need to act decisively and the available model is good enough, and endlessly questioning every abstraction becomes an excuse to never decide.
Worked example
Vanity metrics and the real business
A startup tracking total registered users can watch that number climb while the actual business, paying and active users, quietly stalls. The registration count is a map; the territory is who keeps using and paying. Founders who steer by the flattering metric can feel they're winning right up until the territory, real revenue and retention, asserts itself.
Why it matters for founders
Founders run on dashboards and decks, and the danger is steering by a confident number that quietly stopped describing the real business. The map is indispensable and never the whole truth, so the discipline is checking it against the ground. deciqAI's agents treat their models as approximations to be verified before acting, not as reality itself.
Install this skill (free, MIT)
npx skills add deciqAI/knowledge-skillsFAQ
Does this mean models and metrics are useless?
Not at all. Maps are essential; you can't navigate without them. The principle warns against forgetting they're simplifications, so you keep checking them against the reality they represent.
How does this relate to Goodhart's Law?
Closely. When a measure becomes a target, people optimize the map instead of the territory, and the metric stops tracking the reality it was meant to represent. Remembering the map isn't the territory is the antidote.