Outcome Intelligence: Four Primitives That Turn Audit Data into Market Position Signals

There is a gap between having a score and knowing what it means competitively. A Machine Trust Score of 62 is meaningless without context: is 62 strong, median, or weak for domains in your query cluster? Is it trending up or down? Are competitors above or below you? How much of your projected change is within your control versus driven by market dynamics? Outcome Intelligence answers all four questions — not with a single confidence label, but with four typed primitives that preserve uncertainty and give you the information needed for strategic decisions.

Primitive 1: Position

Position answers the question: where do you stand in the competitive distribution for your primary query cluster? It is expressed as a percentile interval — never a single number — because the distribution itself is uncertain. A position of [38th, 57th percentile] means you are likely in the lower half of the competitive set but the uncertainty is wide. A position of [71st, 79th percentile] means you are confidently in the top quartile. The width of the interval tells you how much you know versus how much is estimation. Position is always relative to a specific query cluster, not the web as a whole.

Primitive 2: Trajectory

Trajectory answers: where are you headed? It reports two scenarios at two time horizons — do-nothing and act (complete this week's actions) — as percentile intervals. The gap between do-nothing and act is the quantified value of action. The gap between the two time horizons (30 days vs. 90 days) shows how fast the opportunity decays. Trajectory is always conditional: "under the baseline competitor assumption at 90 days, completing this week's actions moves your position from [38th, 57th] to [52nd, 71st]." It names the primary driver explicitly.

Primitive 3: Displacement

Displacement answers: are you gaining or losing ground, and why? It decomposes the mechanism: is the change driven by trust decay in your own signals (reversible by fixing Fix Plan items), competitor improvement in the competitive set (only partially reversible — requires improving faster), or market saturation (the cluster is reaching equilibrium, not addressable by action alone)? Displacement tells you which portion of your projected share change is within your control. If 80% of the decline is competitor improvement and only 20% is your own decay, the strategic response is different from the reverse.

Primitive 4: Decision

Decision answers: what should you do next, expressed in competitive market terms? Not "fix your entity schema" but "fixing entity schema affects the commercial query cluster where you are currently in the bottom quartile and three competitors have improved their entity signals in the last audit cycle — this is the cluster where relative improvement produces the largest citation share gain." The Decision primitive connects the Fix Plan item to the competitive context, the affected cluster, the alternative action you are forgoing, and the tradeoff between them.

The Narrative Layer

Outcome Intelligence generates five plain-language statements from the four primitives: a position statement, a trajectory statement, a displacement statement, a decision statement, and an uncertainty statement. These are template-driven and deterministic — no language model inference for the narrative itself, which means they are reproducible and auditable. The uncertainty statement always names the dominant unknown and explains what data would narrow the interval. Uncertainty is a first-class output, not a footer disclaimer.