Decision Intelligence Layer: From Twelve Scores to a Competitive Action Roadmap
Scores tell you where you stand. The Decision Intelligence Layer tells you what to do, in what order, and what it is worth. The Decision Orchestrator and Competitive Reality Simulation transform audit data into a market-position intelligence system.
Most intelligence platforms produce scores. SiteNexis v4 produces decisions. The Decision Intelligence Layer sits above the scoring stack and transforms twelve module scores, hundreds of issues, and the competitive context into a sequenced action roadmap — telling you not just what is wrong but what to fix first, why that ordering maximizes impact, and what you give up by choosing a different sequence.
The Decision Orchestrator
The Decision Orchestrator takes the output of the Fix Plan — a ranked list of all issues across all twelve modules — and applies a sequencing algorithm to produce the Decision Roadmap. The ranking key is the Sequencing Value Score (SVS): impact total multiplied by an unlock multiplier (each issue that unlocks other downstream issues receives a bonus) multiplied by a critical path bonus (1.25x for issues on the highest-impact dependency chain), all divided by effort hours. The result is an ordered queue of actions that front-loads high-unlock, low-effort items.
◆Quick Win Override: if the top-ranked action requires 6 hours and a lower-ranked action is within 15% SVS and requires only 0.5 hours, the quick win takes position 1. The first action in your roadmap should always be completable in a single session — momentum matters more at the start than marginal SVS optimality.
Roadmap Bands: This Week, This Sprint, Backlog
The Decision Roadmap groups sequenced actions into three bands by cumulative effort: This Week (up to 8 hours), This Sprint (8–56 hours), and Backlog (beyond 56 hours). Each band shows the projected score impact of completing all items in that band — so you can tell a stakeholder exactly what "fixing everything this week" would do to each of the twelve module scores before touching a single file. Parallel tracks surface items with no inter-dependency so multiple team members can work simultaneously.
Competitive Reality Simulation
The Competitive Reality Simulation reframes AI citation probability as what it actually is: a relative position in a competitive distribution, not an absolute score. For any query cluster, citation probability is modeled as a softmax allocation across all competing domains — if you improve from 60 to 70 but all competitors also improve by 10 points, your citation share is unchanged. The simulation shows three scenarios (conservative, baseline, aggressive competitor trajectories) across three time horizons (30, 90, 365 days) — so you can see how much of your projected performance change is attributable to your own actions versus competitor movement.
Outcome Intelligence
Outcome Intelligence compresses the full simulation output into four decision primitives: Position (where you stand in the competitive percentile distribution), Trajectory (where you are headed under do-nothing versus act scenarios), Displacement (whether you are gaining or losing ground and why), and Decision (the next recommended action expressed in competitive market terms). Every primitive is expressed as a probability interval — never a point estimate. The interval itself carries the uncertainty, which is always named and decomposed so you know whether it is reducible with more data or irreducibly inherent to market unpredictability.
Get a Decision Roadmap for your domain — see what to fix first, in what order, and what each band of work is worth in score impact.
Get your Decision Roadmap