What we mean by "decision intelligence"

A short definition of the term we use most — and how it differs from analytics, dashboards, and AI that simply answers questions.

“Decision intelligence” is a term worth defining plainly, because we use it constantly and because it is starting to mean everything and therefore nothing.

Here is what we mean by it: the discipline of supporting a specific decision, made by a specific person, with evidence that person can inspect, challenge, and stand behind.

Three parts of that sentence carry the weight.

A specific decision

Analytics describes the world. A dashboard tells you what happened; a report tells you what is happening. Decision intelligence starts from the other end: someone has to choose — approve or reject, fund or defer, act now or wait — and the choice has a deadline and consequences. Everything the system produces exists to serve that choice. Information that doesn’t change the decision is noise, however interesting.

A specific person

Decisions are not made by organizations, in the end. They are made by people inside them — a director who signs, a committee that votes, an official who approves. Decision intelligence keeps that person in the chair: it widens what they can consider, and it leaves the deciding — and the responsibility — with them. A system that quietly makes the choice itself is something else, and we think it should be called something else.

Evidence they can stand behind

This is the part most current AI skips. A fluent answer is not evidence. For a person to stand behind a decision, they need to see why it was recommended, what it rests on, what it assumed, and what remains uncertain — and they need a way to push back before they commit. That is the difference between being informed and being persuaded.

Why the definition matters

Definitions sound academic until you watch what happens without them. When “decision intelligence” means any AI near a decision, systems get adopted because they are impressive rather than because they are checkable — and the person who must answer for the outcome inherits a conclusion they cannot explain. Under our definition, that system fails, no matter how capable it is.

This note is the yardstick. When we say Enpath builds decision intelligence, this — a real decision, a named person, inspectable evidence — is the thing we are claiming to build, and the thing we invite you to hold us to.