← How it works

Causal Graphs (DAGs)

Each AICompass module uses an explicit causal structure. These directed acyclic graphs (DAGs) show which factors influence which outcomes — and where we apply intervention logic.

product fit

Strategic alignment, technical fit, and board readiness influence implementation success. We apply causal adjustment to vendor ROI claims based on your institution's confounding factors.

Priority MatchMember RelevanceTimingBoard ReadinessCore IntegrationStrategic AlignmentTechnical FitImplementation SuccessVendor Claimed ROIAdjusted ROI

Edges (from → to)

  • priority_matchstrategic_alignment
  • member_relevancestrategic_alignment
  • timingstrategic_alignment
  • board_readinessstrategic_alignment
  • core_integrationtechnical_fit
  • strategic_alignmentimplementation_success
  • technical_fitimplementation_success
  • vendor_claimed_roi [confounding adjustment]adjusted_roi

investment

Fit, ROI trajectory, and risk combine into a venture signal. We model how these factors affect investment success and exit potential.

Product FitROI TrajectoryRisk PostureVC DemandExit PotentialRound RiskInvestment Outcome

Edges (from → to)

  • fit_scorevc_demand
  • roi_trajectoryvc_demand
  • roi_trajectoryexit_potential
  • risk_postureround_risk
  • risk_posturevc_demand
  • vc_demandinvestment_outcome
  • exit_potentialinvestment_outcome
  • round_riskinvestment_outcome

ma

Scale, cost synergy, and revenue synergy drive deal value. We model what happens to efficiency and ROE if we intervene on merge, headcount, or branch consolidation.

Merge (do)Headcount ReductionBranch ConsolidationScale EconomiesCost SynergyRevenue SynergyEfficiency RatioROENPV

Edges (from → to)

  • mergeheadcount_reduction
  • mergebranch_consolidation
  • mergescale_economies
  • headcount_reductioncost_synergy
  • branch_consolidationcost_synergy
  • scale_economiesefficiency_ratio
  • cost_synergyefficiency_ratio
  • revenue_synergyroe
  • efficiency_ratioroe
  • cost_synergynpv
  • revenue_synergynpv

portfolio

Macro factors (rates, VIX) drive asset returns. CPA decomposes observed covariance into regime-invariant interventional component (Σ_do) and confounded component. Most valuable for regime robustness, not static optimization.

Interest RatesVolatility (VIX)InflationAsset ReturnsΣ_obsΣ_do

Edges (from → to)

  • ratesasset_returns
  • volatilityasset_returns
  • inflationasset_returns
  • asset_returnssigma_obs
  • asset_returns [causal decomposition]sigma_do

Honest framing

  • CPA (Portfolio) — Genuine causal decomposition (Σ_do); most valuable for regime robustness.
  • M&A — Causal attribution + do-calculus (E[Efficiency | do(merge)]).
  • Product Fit — MCDA (multi-criteria decision analysis) + causal ROI adjustment for confounding.
  • Investment (CVC) — Composite heuristic with explicit decision structure; not formal causal inference.