{
  "meta": {
    "title": "GTM World Model",
    "version": "3.2",
    "built": "2026-05-31",
    "language": "en-US",
    "purpose": "A computational world model of go-to-market that covers the full vocabulary (gtm_vocabulary.json v2, 195 terms) and explains how the field actually behaves: what is true-by-identity, what is causal-but-regime-dependent, and what is merely correlational. Designed so that an agentic GTM system (e.g. an SDR/account-intelligence agent like LangChain's) plugs into the execution tier while reading constraints and objectives from the strategy and economics tiers. The model is a LEARNING object: human overrides are treated as labeled corrections to behavioral coefficients.",
    "relationship_to_vocabulary": "Every vocabulary term is assigned to exactly one tier and one role. The vocabulary supplies the NODES; this file supplies the TIERS, the STATE VARIABLES, the typed EDGES (identity/causal/correlational), the EQUATIONS, and the AGENT INTERFACE. A glossary lists entities; a world model states how entities produce each other.",
    "core_thesis": "GTM is an exact accounting identity (the MRR walk) summarized by regime- and assumption-dependent ESTIMATORS, wrapped in a behavioral membrane whose conversion coefficients are CONDITIONAL on a buyer-state that is largely determined BEFORE the seller funnel begins. Revenue is generated piecewise: multiplicatively by product-market fit (Phi) where switching costs are low, and additively by a switching-cost moat (S) where they are high. A macro regime scalar (Psi) sets whether the market prices growth or free cash flow. The field rewards getting the STRUCTURE right -- which tier owns a decision, whether a link is identity/estimator/conditional, and where the model must refuse to predict -- far more than getting any single coefficient right.",
    "how_to_read": "Start with tiers (the three layers). Then state_variables (what the system tracks). Then edges (how variables move each other, each tagged with epistemic_type). Then equations (the exact identities). Then agent_interface (how an executing agent maps onto the model, using LangChain as the worked example). Then theses (the load-bearing claims) and failure_modes (how models of this field go wrong).",
    "changelog": [
      "v1.0: three tiers + substrate over vocabulary v2 (195 terms); state vars, typed edges, equations, agent interface (LangChain worked example).",
      "v1.1: aligned to vocabulary v2.1 (219 terms). Added the Sense-Reason-Act-Learn cycle as the atomic unit of Tier 1; added agent_properties (inverted cost profile + compounding failure); added the FinOps->CAC edge to the economics tier; added coefficient-compression as a distributional effect on r_i; added governance instruments (task-to-role matrix, escalation paths, data-quality gates, audit logging) to the agent interface; added the GTM-AI maturity ladder and phased-rollout as adoption structure. Sources: Apollo, Poexis, Wyzard (+ LangChain).",
      "v3.2: C6+ Agentic-GTM formal extension. Formalizes eight prose theses (T8-T28 cluster) into 10 new/refined state variables (alpha, G, gamma, p, R, C_a, lambda, theta, kappa, B_agent) and 9 equations (E-A1..E-A9) scoped to Tier 1 (Execution) and the Operations & Systems substrate; does NOT revise Tiers 0/2/3/5 or the 17 base equations. Adds theses T8.1, T29 (governance maturity is a precondition for autonomy to lower cost, not a safety tax — with a worked sign-flip example), and T30 (agent adoption is a category externality no firm-level governance internalizes). Extends the measurement-gap register with 6 agentic parameters, all tagged unmeasurable_hypothesis. Every added edge carries an epistemic_type; the model refuses to assert the agentic-CAC sign (E-A8) until a public holdout exists."
    ],
    "revision_note": "v2.0 is a structural rebuild following an adversarial audit, four independent construction reports, and ~70 documented agentic-GTM implementations. v1.1 was diagnosed as single-firm, seller-centric, steady-state, and overfit to a mid-2010s ZIRP sales-led-SaaS regime it never named. v2.0 (a) demotes Tier-2 'identities' to model-dependent ESTIMATORS, keeping only the MRR walk as exact; (b) makes the PMF thesis PIECEWISE (multiplicative in low-switching-cost segments, additive + a switching-cost moat S in high-switching-cost segments); (c) adds a macro regime scalar Psi above Tier 2 that re-weights the growth/FCF objective; (d) adds a buyer-state tier B and a brand-stock B_r that together form a pre-funnel 'Tier 0'; (e) adds explicit PLG-loop equations; (f) operationalizes PMF (Phi) ex-ante so T7 becomes (in principle) testable; (g) declares a ~25-40% irreducible-noise floor with a prediction-refusal rule. Each added term is either measurable today or explicitly marked unmeasurable in the measurement_gap_register.",
    "epistemic_honesty": "This model marks the boundary between specification and speculation on every added component. Established forms (the MRR walk; Fader-Hardie sBG/BdW survival; Nerlove-Arrow adstock; Bessemer Rule-of-X; the 6sense/LinkedIn buyer-state findings) are labeled as such. Proposed forms with no fitted data (the Phi x S x Psi interaction coefficients; the S regime threshold; brand alpha/delta for SaaS) are labeled 'proposed, untested'. Vendor-sourced magnitudes (6sense, LinkedIn B2B Institute) are flagged: the DIRECTION is well-supported, the exact percentages are commercially interested. The measurement_gap_register lists every parameter the model needs but cannot currently measure from public data.",
    "vocabulary_terms_count": 219,
    "prior_version": "3.1 (backed up at gtm_world_model_v3.1_backup.json)",
    "posture_v3": "v3.0 is the MAXIMAL synthesis: it folds in every substantive structural point raised across the twelve adversarial critiques as first-class modeled content, rather than excluding the unmeasurable to stay parsimonious. This is a deliberate reversal of the v2.x disciplined-minimal direction. The cost is explicit and accepted: v3.0 is MORE COMPLETE and LESS FALSIFIABLE than the minimal core, and a larger share of its constructs are hypotheses without measurement protocols. The single safeguard against the model rotting into an after-the-fact narrative engine is that EVERY added construct is tagged with both an epistemic_type and a measurement_status (see measurement_status_legend). Maximal breadth; honest labels. Treat the model as a MAP of the whole territory whose legend tells you which regions are surveyed and which are sketched.",
    "runnable_seed": "The one executable piece is gtm_engine_slice.py — the MRR-walk identity + an ex-ante, falsifiable Phi prior + an enforced refusal gate. v3.0 is the comprehensive map; that file is the only surveyed-and-walkable trail on it so far.",
    "v3_1_note": "v3.1 removes the hidden subscription assumptions from the 'general' spine. The ten-kinds stress test showed v3.0 was a recurring-revenue model wearing a universal costume: it assumed stock=revenue, one binding-gate type, a single stock, that an equilibrium exists, and that value lands in-period. Each kind that broke broke exactly one of these. v3.1 adds 8 generalizations that make the spine model-agnostic, plus 10 instantiable kind-templates. These are STRUCTURAL upgrades — they let the model REPRESENT each kind honestly; they do not make the unmeasurable cells forecastable. Richer skeleton, same ceiling.",
    "extended": "2026-07-05",
    "v3_2_note": "v3.2 is ADDITIVE, not a rebuild. It closes the gap the v3.1 posture left open: the agentic claims (T8-T28) were carried as prose theses without state variables or equations of their own. v3.2 gives them the same treatment as the rest of the model — typed edges, epistemic tags, and a measurement-gap entry per unmeasured parameter — so an agentic GTM system can read the execution/governance tier as structured constraints, not narrative. Richer execution tier, same epistemic ceiling: the six new gap-register rows mean most added coefficients remain unmeasurable_hypothesis and may inform risk-awareness but never a forward agent action.",
    "agentic_extension_c6plus": "See equations E-A1..E-A9, state variables alpha/G/gamma/p_step/R_chain/C_a/lambda_hitl/theta/kappa/B_agent, and theses T8.1/T29/T30. Runnable seed gtm_engine_slice.py is unchanged; the governance sign-flip worked example (E-A2/E-A3/E-A4) is illustrative, not vendor-calibrated.",
    "term_count": 219,
    "api_version": "1.0",
    "endpoint": "https://shalvisingh.com/api/gtm/model"
  },
  "tiers": {
    "tier_3_strategy": {
      "name": "Strategy / Configuration tier (slow-moving, human-owned)",
      "tempo": "quarters to years",
      "owned_by": [
        "Founders",
        "Product Marketing",
        "GTM leadership"
      ],
      "what_it_decides": "WHO you sell to, WHAT you claim, HOW you sell, WHERE you reach them, and WHETHER the product earns demand. These are largely FORCED MOVES given the economics, not free preferences.",
      "is_input_to": "The agent reads these as configuration; it does not get to change them. In LangChain's system, the ICP, the motion, and the warm/cold playbook are encoded in the agent's SKILL and do-not-send rules — decided by humans, handed to the agent.",
      "vocabulary_terms": [
        "Go-to-Market (GTM) Strategy",
        "GTM Strategy vs Tactics",
        "GTM as a System",
        "Product-Market Fit",
        "Go-to-Market Fit",
        "Minimum Viable GTM",
        "90-Day Plan",
        "Launch",
        "Product Marketing",
        "TAM / SAM / SOM",
        "Market Opportunity",
        "Market Segmentation",
        "STP",
        "ICP",
        "Buyer Persona",
        "Buying Committee",
        "Firmographic Data",
        "Technographic Data",
        "Positioning",
        "Value Proposition",
        "Competitive Positioning",
        "Category Design",
        "Messaging Framework",
        "Proof Points",
        "Why Now",
        "Jobs to be Done",
        "Pricing & Packaging",
        "Freemium",
        "Free Trial",
        "GTM Motion",
        "Product-Led Growth",
        "Sales-Led Growth",
        "Marketing-Led / Inbound-Led",
        "Outbound-Led",
        "Channel/Partner-Led",
        "Community-Led",
        "Ecosystem-Led",
        "Content-Led",
        "Paid-Led",
        "Event-Led",
        "Hybrid Motion",
        "Land and Expand",
        "Bottom-Up Adoption",
        "Self-Serve",
        "4 Ps",
        "AIDA",
        "GTM Operating System",
        "Sales Methodology",
        "Sales Qualification Framework",
        "BANT",
        "CHAMP",
        "MEDDIC",
        "SPICED",
        "The Challenger Sale",
        "Marketing Strategy",
        "ABM",
        "Named Accounts",
        "GTM AI Maturity Ladder",
        "Agentic ABM"
      ],
      "key_outputs_to_lower_tiers": [
        "ICP definition",
        "motion selection",
        "pricing",
        "messaging/playbook",
        "qualification criteria"
      ]
    },
    "tier_2_economics": {
      "name": "Unit-Economics ESTIMATOR tier (model-dependent; only the MRR walk is exact)",
      "tempo": "continuous; reviewed monthly/quarterly",
      "owned_by": [
        "RevOps",
        "Finance",
        "GTM leadership"
      ],
      "what_it_decides": "Whether the Tier-3 configuration is VIABLE and SOLVENT. CRITICAL v2.0 CORRECTION: this tier is NOT 'mostly identities'. The ONLY exact identity is the MRR walk (MRR_t = MRR_{t-1} + New + Expansion - Contraction - Churn). Every other metric -- LTV, NRR, GRR, CAC-payback, Magic Number, Rule-of-40 -- is an ESTIMATOR that summarizes the walk under contestable assumptions. v1.1 mislabeled these as identities, which let the model claim more certainty than its primitives support and propagated upward into a circular T7.",
      "is_input_to": "Sets the objective function and the guardrails the agent optimizes against. LangChain's agent implicitly serves this tier: by raising lead->opportunity conversion it improves the funnel term that feeds CAC efficiency.",
      "vocabulary_terms": [
        "ARR",
        "MRR",
        "ACV",
        "TCV",
        "ARPA",
        "ASP",
        "Average Deal Size",
        "Bookings",
        "Billings",
        "CAC",
        "CAC Payback",
        "CPL",
        "LTV",
        "LTV:CAC",
        "Magic Number",
        "Rule of 40",
        "NRR",
        "GRR",
        "Churn",
        "Retention",
        "Expansion Revenue",
        "Cohort Analysis",
        "Sales Velocity",
        "Sales Cycle",
        "Win Rate",
        "Closed-Won / Closed-Lost",
        "Pipeline Coverage",
        "Forecast Accuracy",
        "Sales Forecasting",
        "Deal Health",
        "NPS",
        "Time to Value",
        "Stickiness",
        "North Star Metric",
        "Leading vs Lagging Indicators",
        "Vanity Metrics",
        "Attribution",
        "Self-Reported Attribution",
        "Dark Funnel",
        "Booking Rate",
        "Win/Loss Analysis",
        "Bowtie Model"
      ],
      "key_constraints_imposed": [
        "LTV > CAC",
        "CAC payback < runway-bounded threshold",
        "motion-viability threshold on ACV",
        "Rule of 40 for scale-stage health"
      ],
      "the_one_identity": {
        "expression": "MRR_t = MRR_{t-1} + New_t + Expansion_t - Contraction_t - Churn_t",
        "status": "identity (exact on a consistent recognition basis)",
        "leaks_when": "Revenue recognition desyncs from cash (ASC 606 deferrals, multi-element arrangements, usage true-ups, mid-period amendments). Falsification: monthly invoiced-MRR-walk vs GAAP revenue delta on public 10-Ks; persistent >5% divergence => even the walk is convention-bound. (Proposed test, unrun.)",
        "provenance": "All four construction reports converge; Component A."
      },
      "estimators": [
        {
          "metric": "LTV",
          "naive_form": "ARPA*margin/churn",
          "hidden_assumption": "homogeneous exponential churn (constant hazard, no tenure dependence)",
          "corrected_form": "sum_t ARPA*margin*S(t)*(1+d)^-t with S(t) from Fader-Hardie sBG; expected lifetime E[T]=(alpha+beta-1)/(alpha-1)",
          "correction_magnitude": "naive overstates 2-3x on a ~10%-churn cohort because high-churn accounts sort out early (worked: alpha=2.5,beta=6 -> E[T]=5mo -> LTV $800 vs naive $1600)",
          "provenance": "Fader-Hardie 2007 sBG / 2018 BdW (established); worked numbers from construction reports A"
        },
        {
          "metric": "NRR",
          "hidden_assumption": "intra-account expansion only; ignores cross-sell attribution; no standard definition (Fastly published DBNR=126% and NRR=93% same period)",
          "note": "endogenous -- inflated by discounting and forced multi-year renewals; cannot be used to 'prove' Phi (the T7 circularity)"
        },
        {
          "metric": "GRR",
          "hidden_assumption": "stationary contract-level churn hazard"
        },
        {
          "metric": "CAC Payback",
          "hidden_assumption": "no expansion, no downgrades, constant gross margin; ignores TVM; numerator scope varies across Bessemer/KeyBanc/OpenView"
        },
        {
          "metric": "Magic Number",
          "hidden_assumption": "fixed one-quarter lag between S&M spend and ARR growth"
        },
        {
          "metric": "Rule of 40 / Rule of X",
          "hidden_assumption": "linear growth-vs-margin trade-off; Rule-of-X corrects this by weighting growth M~2-3x via the Psi regime scalar"
        }
      ],
      "provenance": "Construction reports A (all four converge); adversarial audit finding #1 (the single most damaging) + epistemic audit."
    },
    "tier_1_execution": {
      "name": "Execution / Behavioral tier (fast, increasingly agent-owned)",
      "tempo": "per-lead, per-account, daily/weekly",
      "owned_by": [
        "SDR/BDR",
        "AE",
        "CS/Deployed Eng",
        "Demand Gen",
        "GTM Engineer",
        "and increasingly: the agent"
      ],
      "what_it_decides": "The actual per-lead and per-account loop: detect signal -> decide whether to act -> gather context -> qualify -> draft/personalize -> route -> hand off -> follow up -> expand/retain. This is where the behavioral coefficients (conversion rates, response, reply rates) actually get produced, and where they DRIFT.",
      "is_the_layer_the_agent_runs": true,
      "langchain_mapping": "This entire tier is what LangChain's GTM agent automates: trigger on Salesforce lead -> do-not-send checks -> research (CRM/Gong/LinkedIn/Exa) -> load warm/cold skill -> draft -> human approval in Slack -> queue follow-ups; plus Monday account-intelligence for expansion/renewal/churn signals.",
      "vocabulary_terms": [
        "Funnel",
        "TOFU / MOFU / BOFU",
        "Critical User Journey",
        "Lead",
        "Lead Lifecycle",
        "MQL",
        "MQA",
        "SAL",
        "SQL",
        "CQL",
        "PQL",
        "Opportunity",
        "Lead Handoff",
        "Lead Scoring",
        "Predictive Lead Scoring",
        "Demand Generation",
        "Demand Creation",
        "Demand Capture",
        "Lead Generation",
        "Nurture",
        "Gated Content",
        "Content Syndication",
        "Thought Leadership",
        "Inbound",
        "Speed-to-Lead",
        "Intent Data",
        "Intent Surge",
        "Buying Signal",
        "Buying Intent",
        "Signal-to-Action",
        "SEO",
        "PPC",
        "Retargeting",
        "Co-Marketing",
        "Webinar",
        "VAR",
        "Partner Enablement",
        "Marketplace",
        "Channel Strategy",
        "SDR / BDR",
        "Account Executive",
        "Account Manager",
        "Customer Success",
        "Sales Engineer",
        "Discovery",
        "Proof of Concept",
        "Quota",
        "Sales Stages",
        "Sales Sequence",
        "Cold Email",
        "Email Deliverability",
        "Domain Warm-Up",
        "Sales Enablement",
        "Sales Playbook",
        "Battlecard",
        "Case Study",
        "Multi-Threading",
        "Mutual Action Plan",
        "Territory Management",
        "Account Expansion",
        "Upsell",
        "Cross-Sell",
        "Advocacy",
        "Word-of-Mouth",
        "Onboarding",
        "Activation",
        "Feature Adoption",
        "Product Analytics",
        "Economic Buyer",
        "Champion",
        "Identify Pain",
        "Decision Criteria",
        "Decision Process",
        "Critical Event",
        "Pipeline",
        "Field Sales",
        "Inside Sales",
        "Deal Desk",
        "Sense-Reason-Act-Learn",
        "Next-Best-Action",
        "Signal-to-Revenue",
        "Signal Taxonomy",
        "Coefficient Compression",
        "Outer Flywheel"
      ]
    },
    "cross_cutting_substrate": {
      "name": "Operations & Systems substrate (the rails all tiers run on)",
      "tempo": "always-on",
      "owned_by": [
        "RevOps",
        "Marketing Ops",
        "GTM Engineer"
      ],
      "what_it_is": "The data, tooling, routing, and governance that let the other tiers function. Not a tier in the causal stack — it is the substrate. The AGENT itself is the newest member of this substrate.",
      "vocabulary_terms": [
        "Revenue Operations",
        "Marketing Operations",
        "GTM Engineer",
        "Data Enrichment",
        "Lead Routing",
        "Stage Definitions",
        "Rules of Engagement",
        "SLA",
        "Operating Cadence",
        "QBR",
        "Sales-Marketing Alignment",
        "RACI / DRI",
        "CRM",
        "Marketing Automation Platform",
        "Sales Engagement Platform",
        "Clay",
        "Business Intelligence",
        "PLG CRM",
        "Agentic GTM",
        "GTM OS",
        "Task-to-Role Matrix",
        "Human-in-the-Loop",
        "Escalation Path",
        "Guardrails",
        "Data Quality Gate",
        "Audit Logging",
        "Compounding Failure",
        "Agent Cost Profile",
        "Agent FinOps",
        "Context Engineering",
        "Workflow Orchestration",
        "Subagent",
        "Learning Loop",
        "Phased Rollout"
      ],
      "langchain_mapping": "LangChain's agent IS a substrate component: it sits on Salesforce/Gong/BigQuery/Gmail (systems of record), enforces do-not-send Rules of Engagement, honors a 48h SLA, and its memory store is a new piece of RevOps infrastructure. Their lesson 'connect to systems of record from the start' is a statement about this substrate."
    },
    "tier_0_prefunnel": {
      "name": "Pre-funnel / Buyer-state & Brand tier (where the outcome is largely decided)",
      "tempo": "slow (brand: quarters-years) to medium (buyer-state: weeks-months); mostly INVISIBLE to the seller",
      "owned_by": [
        "Marketing/Brand",
        "Demand-gen",
        "Product (for PLG-led discovery)"
      ],
      "what_it_decides": "Whether you are even in contention before a seller engages. Encodes the empirical reality that ~95% of B2B buyers are out-of-market in any quarter (LinkedIn 95-5 Rule) and that winners are overwhelmingly on the buyer's Day-1 shortlist (6sense: ~85-95%, first seller contact at ~61% of journey in 2025). The seller funnel (Tier 1) is DOWNSTREAM of and CONFOUNDED by this tier.",
      "is_input_to": "Sets the buyer-state B that conditions every Tier-1 conversion coefficient. Brand stock B_r raises the probability of Day-1 shortlist inclusion. This is the 'Tier 0' all four construction reports converged on as the highest-leverage addition.",
      "vocabulary_terms": [
        "Brand Awareness",
        "Demand Creation",
        "Demand Capture",
        "Category Entry Points",
        "Mental Availability",
        "Share of Voice",
        "95-5 Rule",
        "Dark Funnel",
        "Buying Group",
        "Buying Committee",
        "Day-1 Shortlist",
        "Intent Data",
        "Signal-Based Selling",
        "Out-of-Market",
        "In-Market",
        "Thought Leadership"
      ],
      "key_constraints_imposed": [
        "Tier-1 win rate is bounded by Day-1 shortlist position; cold (non-shortlist) outbound converts at ~3-5% vs ~30-50%+ for shortlist/inbound (prospeo/6sense benchmarks).",
        "Doubling seller headcount into a fixed in-market pool causes GEOMETRIC conversion decay, not proportional pipeline growth -- the single biggest behavioral change from v1.1.",
        "Brand stock is a STOCK with its own equation of motion, not a lagged flow; it cannot be switched on quickly and decays slowly (~10-20%/yr in B2B)."
      ],
      "provenance": "Construction reports D+E (all four converge); 6sense BER 2024/2025; LinkedIn B2B Institute 95-5; Ehrenberg-Bass CEPs; adversarial audit blind-spot #4 (highest severity).",
      "evidence_caveat": "Buyer-state PRIMACY (direction) is well-supported across independent sources. Exact magnitudes (95-5, 85-95% shortlist) are 6sense/LinkedIn-sourced and commercially interested; treat percentages as indicative, not precise."
    },
    "cross_cutting_regime": {
      "name": "Macro regime scalar Psi (sits ABOVE Tier 2; re-weights the objective)",
      "tempo": "slow (shifts with the rate/capital cycle over quarters-years)",
      "owned_by": [
        "Board",
        "CFO",
        "Investors (exogenous to the firm)"
      ],
      "what_it_decides": "WHICH weighted combination of growth and free cash flow the market prices. Not a Tier-2 input coefficient -- it changes the OBJECTIVE FUNCTION itself, which is why the latent PMF multiplier Phi cannot absorb it. Resolves T3 (payback-vs-ratio) as REGIME-CONDITIONAL rather than a fixed law.",
      "definition": "Scalar index reducing a macro vector (10yr Treasury, Cloud forward multiples, VC deployment pace, median burn multiple) to a regime score in {ZIRP, tightening, high-rate}.",
      "functional_form": "Objective U = w_g(Psi)*growth + w_f(Psi)*FCF, with w_g/w_f = M(Psi). Bessemer Rule-of-X: X = M*growth + FCF_margin, M~2-3x by capital tightness. (Rule-of-X established; the w_g/w_f mapping is proposed.)",
      "measured_values": {
        "10yr_Treasury": "2021 ~1.45% (ZIRP) -> 2023 ~3.96% -> 2025 ~4.29% (Macrotrends)",
        "Cloud_100_multiple": "~34x peak 2021 (inferred) -> 26x 2023 -> 20x 2025 (Bessemer, -41% from peak)",
        "M_growth_weight": "~3.0-3.5 ZIRP 2021 -> ~2.0-2.3 tightening 2023 -> ~2.0-2.5 normalized 2025-26",
        "implied_w_g/w_f": "~0.8/0.2 in ZIRP -> ~0.3/0.7 in high-rate (proposed mapping)"
      },
      "resolves": "T3: longer CAC payback (24-30mo) is acceptable in ZIRP; high-rate forces <12-18mo. Payback is NOT a fixed king -- it is regime-conditional.",
      "falsification": "A segment (e.g. bootstrapped profitable vertical SaaS) whose growth/FCF weighting did NOT shift 2021->2025 despite the rate spike would show Psi is not universal but a VC-backed artifact. No public panel of firm-level objective weights exists; specified-but-unrun.",
      "provenance": "Construction reports C (all four converge); Bessemer Rule-of-X 2024; Macrotrends; adversarial audit blind-spot #1.",
      "evidence_caveat": "Rate/multiple values are measured (public). The w_g/w_f mapping and the exact M-by-regime are PROPOSED, consistent with Bessemer's Rule-of-40 -> Rule-of-X evolution but not independently fitted here."
    }
  },
  "state_variables": [
    {
      "symbol": "C",
      "name": "Customer/revenue stock",
      "tier": "economics",
      "type": "stock",
      "vocab": [
        "ARR",
        "MRR"
      ],
      "note": "The central stock the whole system grows or leaks. Everything is a flow into or out of C."
    },
    {
      "symbol": "a",
      "name": "Acquisition inflow rate",
      "tier": "execution",
      "type": "flow_in",
      "vocab": [
        "Bookings",
        "Win Rate",
        "Lead Generation"
      ],
      "note": "New customers per period. A LINEAR lever on the steady state."
    },
    {
      "symbol": "delta",
      "name": "Churn rate",
      "tier": "economics",
      "type": "flow_out",
      "vocab": [
        "Churn",
        "GRR"
      ],
      "note": "Fraction of C lost per period. Sits in the DENOMINATOR of steady state, hence nonlinear."
    },
    {
      "symbol": "g",
      "name": "Expansion rate",
      "tier": "execution",
      "type": "flow_modifier",
      "vocab": [
        "Expansion Revenue",
        "Account Expansion",
        "Upsell",
        "Cross-Sell",
        "Land and Expand"
      ],
      "note": "Revenue growth within existing C. Net retention = 1 + g - delta."
    },
    {
      "symbol": "V",
      "name": "Top-of-funnel volume",
      "tier": "execution",
      "type": "flow_source",
      "vocab": [
        "Demand Generation",
        "Lead",
        "Intent Data"
      ],
      "note": "Raw demand entering the funnel."
    },
    {
      "symbol": "r_i",
      "name": "Stage conversion rates",
      "tier": "execution",
      "type": "coefficient",
      "vocab": [
        "MQL",
        "SAL",
        "SQL",
        "Opportunity",
        "Lead Scoring",
        "Speed-to-Lead"
      ],
      "note": "The behavioral coefficients. THIS is what an execution agent like LangChain's moves (lead->oppty up 250%). They DRIFT and saturate. REFINEMENT (v1.1): r_i is not one number but a DISTRIBUTION across reps; agentic augmentation compresses its variance (raises the floor) rather than only shifting the mean — Coefficient Compression."
    },
    {
      "symbol": "p_bar",
      "name": "Average price / ACV",
      "tier": "strategy",
      "type": "coefficient",
      "vocab": [
        "ACV",
        "ARPA",
        "ASP",
        "Pricing & Packaging"
      ],
      "note": "Set in Tier 3. Determines which MOTION is viable (the threshold)."
    },
    {
      "symbol": "m",
      "name": "Gross margin",
      "tier": "economics",
      "type": "coefficient",
      "vocab": [
        "CAC Payback",
        "LTV"
      ],
      "note": "Converts revenue to contribution; gates LTV."
    },
    {
      "symbol": "S",
      "name": "Sales+marketing spend",
      "tier": "economics",
      "type": "control_input",
      "vocab": [
        "CAC",
        "Magic Number"
      ],
      "note": "The primary control input. CAC = S/a."
    },
    {
      "symbol": "x",
      "name": "Market penetration / saturation",
      "tier": "strategy",
      "type": "state",
      "vocab": [
        "TAM / SAM / SOM",
        "Market Opportunity"
      ],
      "note": "Cumulative share of TAM captured. Drives CAC convexity."
    },
    {
      "symbol": "Phi",
      "name": "Product-market fit (latent)",
      "tier": "strategy",
      "type": "latent_multiplier",
      "vocab": [
        "Product-Market Fit",
        "North Star Metric",
        "Retention",
        "NPS"
      ],
      "note": "Unobservable. Inferred from joint pattern of low churn + high win-rate + low CAC + high expansion moving TOGETHER. Enters MULTIPLICATIVELY."
    },
    {
      "symbol": "k(tau)",
      "name": "Lag/memory kernel",
      "tier": "execution",
      "type": "function",
      "vocab": [
        "Dark Funnel",
        "Attribution",
        "Demand Creation",
        "Thought Leadership"
      ],
      "note": "Distributed-lag kernel. Why last-touch attribution structurally misattributes brand/demand-creation."
    },
    {
      "symbol": "Psi",
      "name": "Macro regime scalar",
      "tier": "regime",
      "type": "exogenous_index",
      "vocab": [
        "Rule of 40",
        "Burn Multiple"
      ],
      "note": "Capital-tightness index (10yr Treasury, Cloud multiples, VC pace) re-weighting the growth/FCF objective. Sits above Tier 2."
    },
    {
      "symbol": "B",
      "name": "Buyer-state vector",
      "tier": "prefunnel",
      "type": "latent_state",
      "vocab": [
        "Buying Group",
        "Day-1 Shortlist",
        "In-Market",
        "Intent Data"
      ],
      "note": "B=[shortlist S_p, in-market A_m, journey-completion J_c, group size G]. Pre-funnel; confounds every Tier-1 conversion coefficient. Latent, observed via intent proxies."
    },
    {
      "symbol": "B_r",
      "name": "Brand stock",
      "tier": "prefunnel",
      "type": "stock",
      "vocab": [
        "Brand Awareness",
        "Mental Availability",
        "Category Entry Points",
        "Share of Voice"
      ],
      "note": "Nerlove-Arrow stock of mental availability. dB_r/dt = alpha*I - delta*B_r. Drives shortlist probability. B2B decay ~10-20%/yr."
    },
    {
      "symbol": "S",
      "name": "Switching-cost moat",
      "tier": "economics",
      "type": "latent_modifier",
      "vocab": [
        "Switching Costs",
        "Vendor Lock-in"
      ],
      "note": "Hazard-rate modifier h_eff = h0*exp(-S). High in enterprise/regulated/integrated; ~0 in PLG-SMB. Makes T7 additive when high. Unmeasured publicly; proxied via GRR + low-satisfaction confounder."
    },
    {
      "symbol": "k",
      "name": "PLG viral coefficient + loop state",
      "tier": "prefunnel",
      "type": "loop_state",
      "vocab": [
        "Viral Coefficient",
        "Product-Led Growth",
        "Free-to-Paid Conversion"
      ],
      "note": "u_{t+1}=k*u_t; k_eff=k_raw*retention. B2B k<1 always (~0.2 excellent); divergent k>=1 case never binds."
    },
    {
      "symbol": "alpha",
      "name": "Autonomy level of an agentic execution chain",
      "tier": "execution",
      "type": "control_input",
      "vocab": [
        "Agentic GTM",
        "Human-in-the-Loop"
      ],
      "note": "alpha in [0,1]; 0 = every step human-reviewed, 1 = fully autonomous. The lever the operator actually sets. New in C6+ agentic extension."
    },
    {
      "symbol": "G",
      "name": "Governance / HITL maturity index",
      "tier": "substrate",
      "type": "exogenous_index",
      "vocab": [
        "Human-in-the-Loop",
        "Governance"
      ],
      "note": "G in [0,1] or staged (ad hoc / defined / disciplined). Operationalizes T10. Enters the model as TWO parameters — gamma_max(G) (catch-rate ceiling) and rho(G) (fatigue-rot) — because a single scalar cannot explain the T29 sign-flip in optimal autonomy."
    },
    {
      "symbol": "gamma",
      "name": "Effective error catch-rate of human review",
      "tier": "substrate",
      "type": "function",
      "vocab": [
        "Human-in-the-Loop"
      ],
      "note": "gamma(alpha,G) in [0,1]. Fraction of agent errors human review actually catches at autonomy alpha under governance G. Degrades as review LOAD (1-alpha) rises past what triage discipline sustains — the formal 'HITL rubber-stamp rot' failure mode. See E-A2."
    },
    {
      "symbol": "p_step",
      "name": "Effective per-step error rate of an agent-executed chain",
      "tier": "execution",
      "type": "function",
      "vocab": [
        "Agentic GTM"
      ],
      "note": "p(alpha,G) in [0,1]. Base atomic error p0 net of catches: p = p0*(1-gamma). See E-A1."
    },
    {
      "symbol": "R_chain",
      "name": "Chain reliability across n sequential steps",
      "tier": "execution",
      "type": "function",
      "vocab": [
        "Agentic GTM"
      ],
      "note": "R(alpha,G,n)=[1-p]^n. Bottleneck-theorem analogue to the funnel identity: chain depth n compounds a fixed per-step error into unreliability. Formal reason T11's 'compounding failures' is structural, not merely empirical. See E-A3."
    },
    {
      "symbol": "C_a",
      "name": "Agent execution spend",
      "tier": "economics",
      "type": "flow_out",
      "vocab": [
        "CAC",
        "Agentic GTM"
      ],
      "note": "Inference, orchestration, tooling. USAGE-scaling, distinct from headcount-scaling S. Operationalizes T11 (FinOps->CAC edge): an agentic motion can look cheaper per rep while CAC rises if C_a scales faster than S falls. See E-A5."
    },
    {
      "symbol": "lambda_hitl",
      "name": "HITL-as-data-collection rate",
      "tier": "execution",
      "type": "coefficient",
      "vocab": [
        "Human-in-the-Loop"
      ],
      "note": "lambda in [0,1]. Fraction of reviewed agent actions that convert to labeled training signal for the stage coefficients r_i. Operationalizes T9: HITL as a data-collection mechanism, separate from and additive to its safety-gate role. See E-A6."
    },
    {
      "symbol": "theta",
      "name": "Category-wide noise floor",
      "tier": "prefunnel",
      "type": "loop_state",
      "vocab": [
        "Brand",
        "Demand Generation"
      ],
      "note": "theta(kappa), unbounded above. The signal floor an outbound/inbound touch must clear to register with a buyer. A Nerlove-Arrow adstock analogue to brand-stock B_r, but accumulating with aggregate category agent-adoption kappa. Formalizes T23 'agents are leverage then pollution'. See E-A7."
    },
    {
      "symbol": "kappa",
      "name": "Aggregate agent-adoption share across the competitive category",
      "tier": "regime",
      "type": "exogenous_index",
      "vocab": [
        "Agentic GTM"
      ],
      "note": "kappa in [0,1]. NOT the firm's own alpha — the category-level driver of theta. A shared-resource variable no single firm's governance G can internalize. Drives the T30 category externality."
    },
    {
      "symbol": "B_agent",
      "name": "Agent-evaluator presence within a buying committee",
      "tier": "prefunnel",
      "type": "state",
      "vocab": [
        "Buying Committee",
        "Buyer Persona"
      ],
      "note": "B_agent in [0,1]. Share of agent evaluators inside the buying committee. Extension of the existing buyer-state vector B. Flagged correlational until measured (Gap 7). See E-A9."
    }
  ],
  "equations": [
    {
      "id": "conservation_law",
      "name": "The conservation law (the spine)",
      "epistemic_type": "identity",
      "expression": "dC/dt = a - (delta - g) * C",
      "steady_state": "C* = a / (delta - g)",
      "reads_as": "Customers are a stock with inflow (acquisition) and net outflow (churn minus expansion). The equilibrium size is acquisition over net-churn.",
      "implication": "Acquisition sets the numerator (linear lever); retention sets the denominator (nonlinear lever). As net retention -> 100% (g -> delta), C* diverges with ZERO new acquisition. This is why retention dominates acquisition structurally, not as opinion.",
      "vocab": [
        "ARR",
        "Churn",
        "NRR",
        "GRR",
        "Expansion Revenue",
        "Retention"
      ],
      "agent_relevance": "An execution agent that improves retention/expansion signals (LangChain's account-intelligence flagging churn risk and expansion) is pushing on the DENOMINATOR — higher leverage than its outbound drafting, which pushes on 'a'.",
      "latex": "\\frac{dC}{dt} = a - (\\delta - g)\\,C",
      "steady_state_latex": "C^{*} = \\frac{a}{\\delta - g}"
    },
    {
      "id": "funnel_operator",
      "name": "The funnel as a log-linear operator",
      "epistemic_type": "identity",
      "expression": "R = V * (product of r_i) * p_bar   <=>   ln R = ln V + sum(ln r_i) + ln p_bar",
      "reads_as": "Revenue is volume times the product of stage conversions times price. In log space, contributions are additive.",
      "implication": "BOTTLENECK THEOREM: because stages multiply, the marginal return to fixing the WORST-converting stage dominates improving an already-good one. Doubling a 5% stage doubles throughput; +5pts on a 50% stage adds 10%.",
      "vocab": [
        "Funnel",
        "MQL",
        "SQL",
        "Opportunity",
        "Win Rate",
        "TOFU / MOFU / BOFU"
      ],
      "agent_relevance": "LangChain targeted the lead->qualified-opportunity stage specifically and got 250% — they fixed a constriction, not a comfortable stage. The agent is a bottleneck-stage intervention.",
      "latex": "R = V \\prod_i r_i \\;\\Longleftrightarrow\\; \\ln R = \\ln V + \\sum_i \\ln r_i + \\ln \\bar{p}"
    },
    {
      "id": "unit_economics",
      "name": "The unit-economics master inequality",
      "epistemic_type": "estimator",
      "expression": "LTV > CAC   <=>   (m * p_bar) / delta  >  S / a   <=>   (m * p_bar * a) / (delta * S) > 1",
      "reads_as": "Lifetime contribution must exceed acquisition cost. Expanding LTV exposes churn in the denominator-of-the-denominator, so retention enters quadratically vs how acquisition enters.",
      "implication": "GTM is CONSTRAINED optimization, not maximization. And the BINDING constraint at venture stage is CAC PAYBACK (cash timing), not the LTV:CAC ratio (equilibrium). A 5:1 ratio can still kill you if LTV arrives over 5 years and CAC is due now.",
      "vocab": [
        "LTV",
        "CAC",
        "LTV:CAC",
        "CAC Payback",
        "Magic Number"
      ],
      "benchmark": {
        "LTV:CAC": "3:1 good, 5:1 great, <1:1 fatal",
        "CAC_payback": "<12mo good, <6mo great",
        "Magic_Number": ">1 invest, <0.75 fix engine"
      },
      "v2_correction": "Demoted from identity to estimator. LTV>CAC is an inequality between two ESTIMATORS, not an identity. The LTV side assumes a churn model (falsified as exponential by Fader-Hardie). Only the MRR walk underneath is exact.",
      "latex": "LTV > CAC \\;\\Longleftrightarrow\\; \\frac{m\\,\\bar{p}}{\\delta} > \\frac{S}{a} \\;\\Longleftrightarrow\\; \\frac{m\\,\\bar{p}\\,a}{\\delta\\,S} > 1"
    },
    {
      "id": "saturation",
      "name": "Channel saturation (CAC convexity)",
      "epistemic_type": "causal_regime",
      "expression": "CAC(x) = CAC_0 / (1 - (x / TAM)^beta)",
      "reads_as": "As you exhaust a finite market, cheap demand goes first, so CAC rises convexly with cumulative penetration.",
      "implication": "Every channel has a carrying capacity. Optimal strategy is a PORTFOLIO of channels each run only to its rising-marginal-cost knee. 'Be everywhere' pays fixed costs below scale; 'master one channel' rides it up a convex cost curve. Mix must SHIFT over time as channels saturate — which is why GTM strategies have half-lives.",
      "vocab": [
        "TAM / SAM / SOM",
        "CAC",
        "Channel Strategy",
        "Paid-Led",
        "Content-Led"
      ],
      "caveat": "beta and CAC_0 are estimated per-channel and DECAY. This is a regime edge, not an identity.",
      "latex": "CAC(x) = \\frac{CAC_0}{1 - \\left(\\dfrac{x}{TAM}\\right)^{\\beta}}"
    },
    {
      "id": "motion_threshold",
      "name": "Motion selection as a forced move (step function)",
      "epistemic_type": "causal_regime",
      "expression": "human_sales_viable  <=>  ACV * m * LTV_multiple  >  cost_of_one_human_sales_cycle",
      "reads_as": "Cost-to-serve per human touch is ~fixed; revenue scales with ACV. Below a threshold ACV, a salesperson is unprofitable BY ARITHMETIC.",
      "implication": "Motion is not a preference — it is forced by where ACV sits vs cost-to-serve. The 'PLG <5K / hybrid 5K-50K / sales-led >50K' bands are the footprint of this inequality crossing zero. The hybrid zone is where it's near-equality and the answer is 'let the deal self-select' = product-led sales.",
      "vocab": [
        "GTM Motion",
        "Product-Led Growth",
        "Sales-Led Growth",
        "Hybrid Motion",
        "ACV",
        "PQL",
        "PLG CRM"
      ],
      "agent_relevance": "LangChain (developer-led, technical buyer, expansion-heavy) runs a hybrid motion: self-serve adoption + human outbound for higher-intent leads. Their agent serves exactly the hybrid seam — qualifying which leads warrant a human-approved touch.",
      "latex": "\\text{human sales viable} \\;\\Longleftrightarrow\\; ACV \\cdot m \\cdot \\text{LTV}_{\\text{mult}} > \\text{cost of one human sales cycle}"
    },
    {
      "id": "attribution_convolution",
      "name": "Distributed lags and the attribution problem",
      "epistemic_type": "causal_regime",
      "expression": "r(t) = integral over tau of [ k(tau) * A(t - tau) ] dtau",
      "reads_as": "Observed conversion today is a convolution of PAST activity with a decaying memory kernel.",
      "implication": "Naive last-touch attribution is STRUCTURALLY guaranteed to misattribute: it assigns 100% to the impulse, 0% to the kernel. Brand/demand-creation will always look underperforming to click-attribution, and over-cutting it is a predictable recurring error. Self-reported attribution exists to recover the kernel.",
      "vocab": [
        "Attribution",
        "Self-Reported Attribution",
        "Dark Funnel",
        "Demand Creation",
        "Thought Leadership"
      ],
      "latex": "r(t) = \\int k(\\tau)\\,A(t - \\tau)\\,d\\tau"
    },
    {
      "id": "pmf_multiplier",
      "name": "The product-market-fit multiplier (the master confounder)",
      "epistemic_type": "latent_multiplier",
      "expression": "R ~= Phi * f(GTM_levers)     [multiplicative]     NOT     R ~= Phi + f(GTM_levers)  [additive]",
      "reads_as": "Product-market fit multiplies the effectiveness of every GTM lever simultaneously; it does not add to them.",
      "implication": "Most cross-metric GTM correlations (high NRR <-> efficient CAC, etc.) are CONFOUNDED by Phi driving both. GTM tactics have near-zero causal effect below a Phi threshold and large effect above it. Anything times near-zero Phi is near-zero — the graveyard of well-funded companies with great GTM machinery and no fit. GTM Fit pays off only once PMF clears threshold.",
      "vocab": [
        "Product-Market Fit",
        "Go-to-Market Fit",
        "NRR",
        "Win Rate",
        "CAC",
        "North Star Metric"
      ],
      "contested": true,
      "v2_correction": "Now PIECEWISE (see eq 'pmf_piecewise'). Multiplicative form holds only in low-switching-cost segments. Also: Phi was circular in v1.1 (operationalized via the same NRR/retention it predicts); see eq 'phi_ex_ante' for the de-circularized form.",
      "latex": "R \\approx \\Phi \\cdot f(\\text{GTM levers}) \\quad\\text{[multiplicative]}, \\qquad \\text{not}\\quad R \\approx \\Phi + f(\\text{GTM levers})"
    },
    {
      "id": "sales_velocity",
      "name": "Sales velocity identity",
      "epistemic_type": "identity",
      "expression": "Sales_Velocity = (num_opportunities * win_rate * average_deal_size) / sales_cycle_length",
      "reads_as": "Revenue throughput is opportunities times win-rate times deal-size, divided by how long deals take.",
      "implication": "Four independent levers, and CYCLE LENGTH is in the denominator — compressing the cycle (multi-threading, mutual action plans, faster speed-to-lead) raises velocity as much as winning more. Agents that compress research/drafting time compress cycle length directly.",
      "vocab": [
        "Sales Velocity",
        "Win Rate",
        "Average Deal Size",
        "Sales Cycle",
        "Multi-Threading",
        "Mutual Action Plan",
        "Speed-to-Lead"
      ],
      "latex": "\\text{Sales Velocity} = \\frac{n_{\\text{opps}} \\cdot \\text{win rate} \\cdot \\text{avg deal size}}{\\text{sales cycle length}}"
    },
    {
      "id": "mrr_walk",
      "name": "The MRR walk (the ONE true identity)",
      "epistemic_type": "identity",
      "expression": "MRR_t = MRR_{t-1} + New_t + Expansion_t - Contraction_t - Churn_t",
      "reads_as": "Ending recurring revenue equals starting plus new plus expansion minus contraction minus churn. Exact on a consistent recognition basis.",
      "implication": "This is the only Tier-2 identity. Every other 'metric' is an estimator summarizing trajectories of this walk under assumptions.",
      "leaks_when": "Revenue recognition desyncs from cash (ASC 606, multi-element, usage true-ups). Falsifier: monthly walk vs GAAP delta on 10-Ks; proposed/unrun.",
      "vocab": [
        "MRR",
        "ARR",
        "Expansion Revenue",
        "Churn",
        "Contraction"
      ],
      "provenance": "Construction A (all four converge).",
      "latex": "MRR_t = MRR_{t-1} + \\text{New}_t + \\text{Exp}_t - \\text{Contr}_t - \\text{Churn}_t"
    },
    {
      "id": "ltv_corrected",
      "name": "Corrected LTV under heterogeneous churn (Fader-Hardie sBG)",
      "epistemic_type": "estimator",
      "expression": "LTV = sum_{t>=1} ARPA*margin*S(t)*(1+d)^-t ;  sBG: E[T] = (alpha+beta-1)/(alpha-1) for alpha>1, p_i ~ Beta(alpha,beta)",
      "reads_as": "Lifetime value integrates discounted margin over a survival curve fit to cohort retention, NOT ARPA*margin/churn. Aggregate retention rises over time because high-churn accounts sort out early (heterogeneity), not because individual churn falls.",
      "implication": "Naive exponential LTV overstates 2-3x on a ~10%-churn cohort. Worked: alpha=2.5, beta=6 -> E[T]=5mo -> LTV $800 vs naive $1600 (2x). Use sBG for contractual/time-invariant churn, Beta-discrete-Weibull (BdW) for non-monotone.",
      "vocab": [
        "LTV",
        "Churn",
        "Retention",
        "Cohort Analysis"
      ],
      "provenance": "Fader-Hardie 2007 (sBG) / 2018 (BdW), established; worked numbers from construction A.",
      "status": "established functional form; fit on real cohort panels",
      "latex": "LTV = \\sum_{t \\ge 1} ARPA \\cdot m \\cdot S(t)\\,(1+d)^{-t}, \\qquad E[T] = \\frac{\\alpha + \\beta - 1}{\\alpha - 1}\\;\\;(\\alpha > 1)"
    },
    {
      "id": "pmf_piecewise",
      "name": "Piecewise revenue: multiplicative PMF (low-S) vs additive switching-cost moat (high-S)",
      "epistemic_type": "causal_regime",
      "expression": "R = Phi*f(GTM)                         if S < S*  (low-switching)\nR = Phi*f1(GTM) + S*f2(installed_base)  if S >= S* (high-switching)\nwith effective churn hazard h_eff = h0 * exp(-S)",
      "reads_as": "Where switching costs are low (consumer, PLG-SMB, transactional B2B), revenue is fit AMPLIFIED BY go-to-market: multiplicative. Where switching costs are high (enterprise multi-year, regulated, deeply integrated), a moat term S adds revenue largely INDEPENDENT of product love.",
      "implication": "T7's strong multiplicative form is wrong in ~the most valuable third of enterprise software. Oracle/SAP-class: NPS near/below 0, 30-50% unused seats, yet GRR >95% -- revenue tracks S (lock-in), not Phi. Classical IO (Klemperer; Farrell-Klemperer) is explicitly ADDITIVE: switching-cost rents are a premium ON TOP OF current quality.",
      "S_scale": "Worked: SMB PLG ~20%/yr churn -> S~0; SAP-class <5%/yr -> S~1.4 (from 0.05=0.20*exp(-S)). Regime boundary operationalized via observed GRR>=~95% + low-satisfaction confounder check (since S itself is unmeasurable from public data).",
      "vocab": [
        "Switching Costs",
        "Vendor Lock-in",
        "Retention",
        "GRR",
        "NRR"
      ],
      "provenance": "Construction B (all four converge); Klemperer 1987 / Farrell-Klemperer 2007 (additive switching-cost theory, established); adversarial audit T7 section.",
      "evidence_caveat": "Piecewise STRUCTURE is well-grounded. S* threshold, the exp(-S) hazard form, and S magnitudes are PROPOSED, not fitted on a public panel.",
      "latex": "R = \\begin{cases} \\Phi\\, f(\\text{GTM}) & S < S^{*} \\\\[4pt] \\Phi\\, f_1(\\text{GTM}) + S\\, f_2(\\text{installed base}) & S \\ge S^{*} \\end{cases} \\qquad h_{\\text{eff}} = h_0\\, e^{-S}"
    },
    {
      "id": "psi_objective",
      "name": "Macro regime objective re-weighting (Psi / Rule-of-X)",
      "epistemic_type": "causal_regime",
      "expression": "U = w_g(Psi)*growth + w_f(Psi)*FCF ;  Rule-of-X: X = M(Psi)*growth + FCF_margin ;  M ~ 2-3 by capital tightness",
      "reads_as": "The market's objective function is regime-dependent. Psi (a function of 10yr Treasury, Cloud multiples, VC pace) sets the growth weight M. In ZIRP, growth is weighted ~3x and long payback is fine; in high-rate, M falls and payback must shorten.",
      "implication": "Resolves T3: 'payback over ratio' is true ONLY when capital is expensive (M low) or runway < ~2x payback. In ZIRP it was false; for capital-abundant incumbents it is false. Payback is regime-conditional, not a fixed king.",
      "measured": "10yr: 1.45%(2021)->3.96%(2023)->4.29%(2025); Cloud-100 multiple 26x(2023)->20x(2025); M ~3.0-3.5->2.0-2.5.",
      "vocab": [
        "Rule of 40",
        "CAC Payback",
        "Burn Multiple",
        "Magic Number"
      ],
      "provenance": "Construction C (all four converge); Bessemer Rule-of-X 2024; Macrotrends; adversarial audit blind-spot #1.",
      "evidence_caveat": "Rates/multiples measured. w_g/w_f mapping and M-by-regime PROPOSED.",
      "latex": "U = w_g(\\Psi)\\,\\text{growth} + w_f(\\Psi)\\,\\text{FCF}, \\qquad X = M(\\Psi)\\,\\text{growth} + \\text{FCF margin}, \\qquad M \\approx 2\\text{–}3"
    },
    {
      "id": "buyer_state_coupling",
      "name": "Buyer-state confounds the seller funnel (the de-confounded estimand)",
      "epistemic_type": "conditional_coefficient",
      "expression": "Naive (confounded): Win = Leads * Conversion\nTrue estimand: P(win | seller_touch, S_p, A_m, J_c, Phi, Psi, S)\nwhere observed Opp->Close = P(win | Opp, B-mix), NOT P(win | Opp, seller skill)",
      "reads_as": "The seller funnel is DOWNSTREAM of an invisible buyer-state B = [shortlist position S_p, in-market A_m, journey-completion J_c, buying-group size G]. Funnel conversion coefficients are mixtures over B, not measures of execution.",
      "implication": "THE behavior-changing edge. In v1.1, doubling SDRs doubled pipeline. Here, doubling SDRs into a fixed in-market pool (A_m=1 is only ~5% of accounts) causes GEOMETRIC conversion decay. Stops the model recommending infinite GTM scale-up; shifts capital toward brand (B_r) and PLG. Cold/non-shortlist outbound ~3-5% win vs shortlist/inbound ~30-50%+.",
      "vocab": [
        "Buying Group",
        "Day-1 Shortlist",
        "Intent Data",
        "95-5 Rule",
        "Win Rate",
        "Dark Funnel"
      ],
      "provenance": "Construction D (all four name it the #1 behavior-changing component); 6sense BER 2024/2025; Forrester; Cunningham; prospeo win-rate benchmarks.",
      "evidence_caveat": "Direction strongly supported. Magnitudes (85-95% shortlist, 95-5, ~70% journey pre-contact) are 6sense/LinkedIn-sourced, commercially interested. B is a LATENT state; observed only via noisy intent proxies.",
      "latex": "\\begin{aligned} \\text{Naive (confounded):}\\quad & \\text{Win} = \\text{Leads} \\times \\text{Conversion} \\\\ \\text{True estimand:}\\quad & P(\\text{win} \\mid \\text{seller touch},\\, S_p,\\, A_m,\\, J_c,\\, \\Phi,\\, \\Psi,\\, S) \\end{aligned}"
    },
    {
      "id": "brand_stock",
      "name": "Brand as a stock (Nerlove-Arrow adstock), not a lagged flow",
      "epistemic_type": "causal_regime",
      "expression": "dB_r/dt = alpha*I(t) - delta*B_r(t) ;  P(shortlist S_p=1 | B_r) = sigmoid(lambda0 + lambda1*B_r)",
      "reads_as": "Brand is a STOCK that accumulates mental availability across category entry points and decays slowly. It is HOW you get on the Day-1 shortlist. Past investment yields long-term baseline demand; you cannot switch it on quickly.",
      "implication": "Differs from v1.1 T6, which treated demand-creation as a lagged FLOW inside the attribution convolution. Here brand is the stock that GENERATES the flow. With ~95% of buyers out-of-market, a funnel-only model is blind to where mental availability accumulates.",
      "parameters": "B2B brand decay delta ~10-20%/yr (half-life 1-3yr, Ehrenberg-Bass 'stop advertising' cases). alpha (build rate) not cleanly quantified publicly. Measured via share-of-search, prompted/unprompted awareness, CEP breadth, SOV/SOM.",
      "vocab": [
        "Brand Awareness",
        "Mental Availability",
        "Category Entry Points",
        "Share of Voice",
        "Demand Creation"
      ],
      "provenance": "Construction E (all four converge); Nerlove-Arrow 1962 (established); Ehrenberg-Bass / Binet-Field; LinkedIn 95-5.",
      "evidence_caveat": "Nerlove-Arrow form established; the 'adstock illusion' means a long-lag flow can mimic a stock -- distinguishing them needs a media-pause natural experiment. alpha/delta for SaaS specifically are UNMEASURED. Flagged retrospective-only / liability in 2 of 4 reports.",
      "latex": "\\frac{dB_r}{dt} = \\alpha\\, I(t) - \\delta\\, B_r(t), \\qquad P(S_p = 1 \\mid B_r) = \\sigma(\\lambda_0 + \\lambda_1 B_r)"
    },
    {
      "id": "plg_loop",
      "name": "PLG loop (recursive), distinct from the funnel (multiplicative)",
      "epistemic_type": "causal_regime",
      "expression": "u_{t+1} = u_t*(1 + k - c) ;  pure-viral u_{t+1} = k*u_t, closed form u_0/(1-k), divergent at k>=1 ;  k_eff = k_raw * R (retention)",
      "reads_as": "Product-led growth is a recursive loop, not a chain of funnel conversions. Closed-form divergence requires viral coefficient k>=1. Retention is the UPSTREAM CAUSE of virality (dead users send no invites), reversing the usual arrow.",
      "implication": "Two hard couplings: (1) expansion-as-acquisition under usage pricing BREAKS NRR's intra-account assumption (cross-team adoption is acquisition wearing a retention label); (2) k_eff = k_raw*R, so Phi (driving retention) is upstream of the loop. Hands off to sales-assist at a usage/MRR threshold, where buyer-state B activates.",
      "k_reality": "Sustained k>=1 in B2B is essentially MYTHICAL -- excellent B2B k~0.2, real 'viral' B2B is k<1 + paid amplification. So the divergent closed form NEVER BINDS; PLG is a highly efficient channel, not perpetual motion. Free-to-paid: ~8% median, 15-25% great.",
      "vocab": [
        "Product-Led Growth",
        "Viral Coefficient",
        "Free-to-Paid Conversion",
        "PQL",
        "Time to Value",
        "Expansion Revenue"
      ],
      "provenance": "Construction F (all four converge); Andrew Chen 'Retention is King'; OpenView/Poyer/ProductLed benchmarks.",
      "evidence_caveat": "Loop form established; the k>=1 divergent case is a non-binding limiting case. No public B2B example of sustained k>=1 organically.",
      "latex": "u_{t+1} = u_t\\,(1 + k - c), \\qquad \\text{pure viral: } u_{t+1} = k\\,u_t \\;\\to\\; \\frac{u_0}{1 - k}, \\qquad k_{\\text{eff}} = k_{\\text{raw}} \\cdot R"
    },
    {
      "id": "phi_ex_ante",
      "name": "Ex-ante PMF -- the testable (but not-yet-runnable) form of T7",
      "epistemic_type": "latent_multiplier",
      "expression": "log(growth_it) = b0 + b1*log(Phi_{i,t-1}) + b2*log(GTM_it) + b3*log(Phi_{i,t-1})*log(GTM_it) + e\nb3 > 0 => multiplicative (T7 strong) ;  b3 ~ 0 => additive",
      "reads_as": "To de-circularize T7, fix Phi BEFORE the growth window using a measure that cannot be re-inferred from revenue: cohort-retention floor at week 8-12 (primary), or independent G2/Gartner satisfaction at t-1 (secondary). Then the interaction term b3 settles multiplicative vs additive.",
      "implication": "This is the CORRECT falsifiable core of T7. But the cross-company panel (independent t-1 Phi + subsequent growth + comparable GTM spend) DOES NOT EXIST publicly. Until a private aggregator (Bessemer, SaaS Capital) runs it, T7 remains a LOCALIZED HEURISTIC, not a tested claim.",
      "vocab": [
        "Product-Market Fit",
        "Retention",
        "NPS",
        "North Star Metric"
      ],
      "provenance": "Construction G (all four converge, incl. the 'data does not exist, label it metaphysical' conclusion); Sean Ellis 40% (flagged uncalibrated).",
      "evidence_caveat": "Specified and falsifiable IN PRINCIPLE; unrun in practice. The honest terminal status of the model's most contested thesis.",
      "latex": "\\log(\\text{growth}_{it}) = b_0 + b_1 \\log \\Phi_{i,t-1} + b_2 \\log \\text{GTM}_{it} + b_3 \\log \\Phi_{i,t-1}\\,\\log \\text{GTM}_{it} + \\varepsilon"
    },
    {
      "id": "noise_floor",
      "name": "Irreducible-noise floor (prediction-refusal guardrail)",
      "epistemic_type": "correlational",
      "expression": "theta = Var(noise)/Var(growth) ~ 0.25-0.40 ;  REFUSE forecast if P(unobserved shock in horizon) > tau OR posterior interval width > omega",
      "reads_as": "~25-40% of mid-stage B2B SaaS growth-rate variance is unexplainable by any GTM lever (competitor surprise, macro shock, viral inflection, champion exit). The model must output confidence bands and 'prediction refused' states rather than false point estimates.",
      "implication": "Decision rule: refuse LTV before the cohort-retention floor flattens (month 6-12); refuse forecasts spanning a competitor surprise; widen intervals through Psi regime transitions. This marks the BOUNDARY of the model's own predictive power.",
      "is_it_reducible": "Falsification/honesty: the floor is PROVISIONAL, not physics. If richer data (telemetry, intent, founder history) pushes out-of-sample R^2 from ~0.6 to ~0.85, theta falls to ~0.15. So part of 'irreducible' is really data-availability. Label as a contingent construct.",
      "vocab": [
        "Forecast Accuracy",
        "Sales Forecasting",
        "Leading vs Lagging Indicators"
      ],
      "provenance": "Construction H (all four converge); Bill Gross TED (42% timing, hindsight-biased); Taleb heavy tails; Roberge 'sales is predictable' counter.",
      "evidence_caveat": "25-40% is a REASONED ASSUMPTION synthesizing Gross/Taleb, NOT an empirically pinned constant. No public benchmark has fit theta = 1 - out-of-sample R^2 for B2B SaaS growth.",
      "latex": "\\theta = \\frac{\\operatorname{Var}(\\text{noise})}{\\operatorname{Var}(\\text{growth})} \\approx 0.25\\text{–}0.40"
    },
    {
      "id": "agentic_error_step",
      "name": "E-A1 · Effective per-step error rate under partial autonomy",
      "epistemic_type": "causal_regime",
      "expression": "p(alpha, G) = p0 * [1 - gamma(alpha, G)]",
      "latex": "p(\\alpha, G) = p_0\\,\\bigl[1 - \\gamma(\\alpha, G)\\bigr]",
      "reads_as": "The per-step error an agent chain actually ships is its zero-oversight base rate p0 knocked down by whatever fraction human review catches.",
      "implication": "Autonomy only lowers cost through reliability if gamma is real. Raising alpha raises p unless governance holds gamma up.",
      "caveat": "Holds only while chain steps are conditionally independent given alpha,G. Breaks under correlated/systemic failure (one upstream data-source outage corrupting every step at once) — not covered.",
      "agent_relevance": "This is the error rate an executing agent (e.g. LangChain's SDR chain) actually inherits per action; the operator tunes it via alpha and G, not by prompt tweaks alone.",
      "vocab": [
        "Agentic GTM",
        "Human-in-the-Loop"
      ]
    },
    {
      "id": "reviewer_fatigue_catchrate",
      "name": "E-A2 · Reviewer-fatigue-adjusted catch-rate",
      "epistemic_type": "causal_regime",
      "expression": "gamma(alpha, G) = clip[ gamma_max(G)*(1-alpha) - rho(G)*(1-alpha)^2, 0, gamma_max(G) ]",
      "latex": "\\gamma(\\alpha, G) = \\operatorname{clip}\\!\\bigl[\\gamma_{\\max}(G)(1-\\alpha) - \\rho(G)(1-\\alpha)^2,\\; 0,\\; \\gamma_{\\max}(G)\\bigr]",
      "reads_as": "Review catches errors up to a governance ceiling gamma_max(G), but catch efficacy DEGRADES as review load (1-alpha) rises past what triage discipline sustains.",
      "implication": "Governance enters as two parameters, not one: gamma_max (ceiling) AND rho (fatigue-rot). Mature governance is defined partly by a LOWER rho — discipline that resists alert fatigue as volume scales. This two-parameter split is what drives the T29 sign flip.",
      "caveat": "gamma_max and rho both require operator-specific calibration; no public source publishes reviewer catch-rate by review load (see measurement-gap register).",
      "agent_relevance": "Formal statement of the 'HITL rubber-stamp rot' failure mode: past a load threshold, adding human review to an agent chain catches FEWER real errors, not more.",
      "vocab": [
        "Human-in-the-Loop",
        "Governance"
      ]
    },
    {
      "id": "agentic_chain_reliability",
      "name": "E-A3 · Chain reliability",
      "epistemic_type": "identity",
      "expression": "R(alpha, G, n) = [1 - p(alpha, G)]^n",
      "latex": "R(\\alpha, G, n) = \\bigl[1 - p(\\alpha, G)\\bigr]^{n}",
      "reads_as": "Reliability of an n-step agent chain is the per-step success rate compounded n times.",
      "implication": "Bottleneck-theorem analogue to the funnel identity R=V*prod(r_i)*p_bar: chain DEPTH compounds a fixed per-step error. Autonomy safe on a 3-step task can be unsafe on an 8-step one. This is why T11's compounding-failure claim is a structural property of chain depth.",
      "caveat": "Identity ONLY given independent per-step failure (E-A1's assumption). Under correlated failure the true reliability is lower than this product.",
      "agent_relevance": "Tells an operator the max chain length an agent can run at a target reliability given its calibrated p — a hard design constraint on how many steps to automate before a human gate.",
      "vocab": [
        "Agentic GTM"
      ]
    },
    {
      "id": "agentic_effective_cost",
      "name": "E-A4 · Effective cost per completed task",
      "epistemic_type": "estimator",
      "expression": "C_eff(alpha, G) = n*[c_infer + c_human*(1-alpha)] + [1 - R(alpha, G, n)]*C_fail*(1 - kappa_gov(G))",
      "latex": "C_{\\text{eff}}(\\alpha, G) = n\\,[c_{\\text{infer}} + c_{\\text{human}}(1-\\alpha)] + [1 - R(\\alpha, G, n)]\\,C_{\\text{fail}}\\,(1 - \\kappa_{\\text{gov}}(G))",
      "reads_as": "Cost per finished task = execution cost (fixed inference plus human review scaling with load 1-alpha) plus expected failure cost, discounted by the share of failure cost governance actually absorbs.",
      "implication": "kappa_gov(G) is the fraction of failure COST governance absorbs (rollback, incident response) — distinct from gamma, the fraction of raw errors it PREVENTS. The two enter separately. Minimizing C_eff over alpha is the operator's real decision; its argmin flips across the G threshold (see worked example / T29).",
      "caveat": "Estimator, not identity: c_infer, c_human, C_fail, kappa_gov all need operator-specific calibration and are not structurally guaranteed across firms or task types.",
      "agent_relevance": "The objective an agentic-GTM operator should actually minimize when choosing autonomy — not activity cost, not per-rep cost.",
      "vocab": [
        "Agentic GTM",
        "CAC",
        "Human-in-the-Loop"
      ]
    },
    {
      "id": "agentic_cac_split",
      "name": "E-A5 · Agent execution spend inside CAC",
      "epistemic_type": "estimator",
      "expression": "CAC = (S + C_a) / a,   C_a = sum_tasks n*c_infer*(volume)",
      "latex": "\\mathrm{CAC} = \\frac{S + C_a}{a},\\qquad C_a = \\sum_{\\text{tasks}} n\\,c_{\\text{infer}}\\,(\\text{volume})",
      "reads_as": "Split the CAC numerator: headcount-scaling spend S plus usage-scaling agent spend C_a, over acquisitions a.",
      "implication": "v3.1's CAC estimator implicitly buried agent spend inside S. Separating C_a makes T11's FinOps->CAC edge explicit: an agentic motion can cut S per rep while CAC still RISES if C_a scales faster than S falls.",
      "caveat": "Refines the existing CAC estimator; inherits the same estimator status (segmentation, horizon, accounting basis all contestable).",
      "agent_relevance": "The line item a FinOps-aware GTM agent must expose so an operator sees inference spend as a first-class acquisition cost, not hidden opex.",
      "vocab": [
        "CAC",
        "Agentic GTM"
      ]
    },
    {
      "id": "hitl_learning_rate",
      "name": "E-A6 · HITL as a learning-rate term on coefficient drift",
      "epistemic_type": "causal_regime",
      "expression": "dr_i/dt = f_existing(r_i, ...) + lambda*gamma(alpha, G)*(target_i - r_i)",
      "latex": "\\frac{dr_i}{dt} = f_{\\text{existing}}(r_i, \\dots) + \\lambda\\,\\gamma(\\alpha, G)\\,(\\text{target}_i - r_i)",
      "reads_as": "Stage coefficients r_i drift toward a target at a rate set by lambda (share of reviews that become training signal) times the review volume gamma actually producing catches.",
      "implication": "Makes the 'Coefficient Compression' drift rate ENDOGENOUS. Operationalizes T9: HITL is a data-collection mechanism (this term), separate from and additive to its safety-gate role (E-A2/E-A4).",
      "caveat": "Directionally motivated by T9; lambda is unmeasured — no surveyed vendor tracks review-as-training-signal vs review-as-gate separately.",
      "agent_relevance": "The mechanism by which a well-instrumented agent stack compounds: every reviewed action is a labeled correction to a behavioral coefficient, not just a caught error.",
      "vocab": [
        "Human-in-the-Loop",
        "Agentic GTM"
      ]
    },
    {
      "id": "category_noise_floor",
      "name": "E-A7 · Category-wide noise floor from aggregate agent adoption",
      "epistemic_type": "causal_regime",
      "expression": "dtheta/dt = eta*kappa - delta_theta*theta,   theta_steady = eta*kappa / delta_theta",
      "latex": "\\frac{d\\theta}{dt} = \\eta\\,\\kappa - \\delta_\\theta\\,\\theta,\\qquad \\theta^{*} = \\frac{\\eta\\,\\kappa}{\\delta_\\theta}",
      "steady_state": "theta_steady = eta*kappa / delta_theta",
      "steady_state_latex": "\\theta^{*} = \\dfrac{\\eta\\,\\kappa}{\\delta_\\theta}",
      "reads_as": "The buyer's noise floor accumulates with aggregate category agent-adoption kappa and decays slowly — an adstock for outbound saturation.",
      "implication": "Formalizes T23/T30: as kappa rises across the category, theta rises and the marginal GTM alpha any single firm's agent can extract falls. A category-level tragedy of the commons no single vendor's governance G can undo.",
      "caveat": "Structural analogue to the brand-stock equation; eta and delta_theta are unmeasured (no source tracks a category noise-floor series).",
      "agent_relevance": "Tells an agent operator that first-mover outbound alpha is time-limited by construction — the model's own reason not to over-index on an outbound-agent moat.",
      "vocab": [
        "Brand",
        "Demand Generation",
        "Agentic GTM"
      ]
    },
    {
      "id": "agentic_cac_holdout",
      "name": "E-A8 · Controlled agentic-CAC holdout estimator",
      "epistemic_type": "estimator",
      "expression": "dCAC_agentic = CAC_agentic - CAC_manual  (matched cohort, net of deliverability decay + lambda_maint)",
      "latex": "\\Delta\\mathrm{CAC}_{\\text{agentic}} = \\mathrm{CAC}_{\\text{agentic}} - \\mathrm{CAC}_{\\text{manual}}",
      "reads_as": "The matched-cohort cost-per-qualified-opportunity difference an agentic motion would have to prove — net of deliverability decay and operator maintenance.",
      "implication": "Written down specifically so its inputs can be named as UNMEASURED. This is the formal shape of the comparison T15 says has no public holdout. Included so a vendor sees exactly what a validating experiment must hold constant.",
      "caveat": "Explicitly flagged LOW-CONFIDENCE: no source in the literature synthesis provides inputs to compute this today. Highest-priority entry in the measurement-gap register.",
      "contested": true,
      "agent_relevance": "The single experiment that would move agentic-GTM ROI messaging from vendor-claimed to audited; the model refuses to assert the sign until it is run.",
      "vocab": [
        "CAC",
        "Agentic GTM"
      ]
    },
    {
      "id": "buyer_state_agent_ext",
      "name": "E-A9 · Buyer-state extension for AI-mediated buying committees",
      "epistemic_type": "correlational",
      "expression": "B_ext = B (+) [B_agent]   (B_agent: agent-evaluator share in the committee)",
      "latex": "B_{\\text{ext}} = B \\oplus [B_{\\text{agent}}]",
      "reads_as": "Append an agent-evaluator share term to the human buyer-state vector B.",
      "implication": "Restates Gap 7 as a variable: no source tests whether T12's ~95% pre-contact primacy strengthens, weakens, or holds when the buying committee itself includes AI evaluators.",
      "caveat": "Correlational, not causal, until measured — 2025-26 agent-buying preprints study individual shopping agents, not B2B buying-committee dynamics.",
      "agent_relevance": "Flags that a seller-side agent may increasingly be selling TO a buyer-side agent — a regime the pre-funnel tier does not yet model causally.",
      "vocab": [
        "Buying Committee",
        "Buyer Persona"
      ]
    }
  ],
  "theses": [
    {
      "id": "T0",
      "claim": "GTM is a causal identity wrapped in a correlational membrane whose coefficients are functions of regime and of a latent PMF multiplier.",
      "type": "meta",
      "load_bearing": true
    },
    {
      "id": "T1",
      "claim": "Acquisition is a linear lever; retention is a nonlinear one (it shrinks the steady-state denominator). A field obsessed with top-of-funnel is optimizing the lower-exponent term.",
      "from_equation": "conservation_law"
    },
    {
      "id": "T2",
      "claim": "Bottleneck theorem: because funnel stages multiply, fixing the worst-converting stage dominates improving a good one.",
      "from_equation": "funnel_operator"
    },
    {
      "id": "T3",
      "claim": "[v2.0 REGIME-CONDITIONAL] CAC payback binds harder than the LTV:CAC ratio ONLY when capital is expensive (Psi tight, M low) or runway < ~2x payback. In ZIRP it was false; for capital-abundant incumbents it is false. Payback is not a fixed king -- it is a function of Psi.",
      "from_equation": "psi_objective",
      "v2_status": "was stated as a fixed law; now regime-conditional via Psi"
    },
    {
      "id": "T4",
      "claim": "Every channel has a carrying capacity and convex CAC; the optimum is a time-shifting portfolio, which is why GTM strategies have half-lives.",
      "from_equation": "saturation"
    },
    {
      "id": "T5",
      "claim": "Motion is a forced move set by ACV vs cost-to-serve, not a preference; the PLG/hybrid/sales-led bands are that inequality's footprint.",
      "from_equation": "motion_threshold"
    },
    {
      "id": "T6",
      "claim": "GTM has distributed lags, so last-touch attribution structurally misattributes and over-cutting brand is a recurring institutional error.",
      "from_equation": "attribution_convolution"
    },
    {
      "id": "T7",
      "claim": "[v2.0 PIECEWISE] PMF is a multiplicative amplifier of GTM ONLY in low-switching-cost segments (R=Phi*f). In high-switching-cost segments (enterprise, regulated, integrated) revenue is ADDITIVE: R=Phi*f1 + S*f2, where a switching-cost moat S generates revenue largely independent of product love (Oracle/SAP existence proof). The strong multiplicative form was circular (Phi proxied by the NRR it predicts) and is wrong in ~the most valuable third of enterprise software.",
      "from_equation": "pmf_piecewise",
      "load_bearing": true,
      "contested": true,
      "v2_status": "was strong-multiplicative + circular; now piecewise + de-circularized via phi_ex_ante (but panel unrun)"
    },
    {
      "id": "T8",
      "claim": "An agentic GTM system executes the behavioral tier and is uniquely good at moving conversion coefficients and compressing cycle time — but only within the configuration and constraints set by the human-owned strategy and economics tiers.",
      "type": "architecture",
      "evidence": "LangChain GTM agent"
    },
    {
      "id": "T9",
      "claim": "Human-in-the-loop is not just a safety mechanism; it is the data-collection mechanism by which the behavioral coefficients are learned. A GTM model that can't ingest its own correction signal is dead on arrival.",
      "type": "architecture",
      "evidence": "LangChain memory loop"
    },
    {
      "id": "T10",
      "claim": "Every serious agentic-GTM vendor independently converges on the same micro-architecture (Sense-Reason-Act-Learn under human-in-the-loop, governed, on a unified platform) — which validates the three-tier model from the outside: they are all building the SUBSTRATE that lets agents execute Tier 1 while reading Tiers 2-3. The competitive edge has shifted from WHETHER you adopt agents to how DISCIPLINED and GOVERNED the implementation is.",
      "type": "architecture",
      "evidence": "Apollo, Poexis, Wyzard convergence",
      "load_bearing": true
    },
    {
      "id": "T11",
      "claim": "Agents invert software's cost/reliability profile (cheap to build, expensive+unreliable to run, compounding failures), which adds a FinOps->CAC edge to the economics tier and makes human-in-the-loop a structural necessity (error-chain interruption) rather than merely a safety nicety.",
      "from": "agent_properties",
      "load_bearing": true
    },
    {
      "id": "T12",
      "claim": "BUYER-STATE PRIMACY: the seller funnel observes only ~5% of value creation; the outcome is largely determined pre-contact by Day-1 shortlist position. Funnel conversion coefficients are CONDITIONAL (mixtures over buyer-state), not causal. Doubling seller activity into a fixed in-market pool yields geometric decay, not proportional pipeline. This is the single biggest behavioral change from v1.1.",
      "from_equation": "buyer_state_coupling",
      "load_bearing": true,
      "contested": false,
      "evidence_caveat": "Direction well-supported; magnitudes 6sense/LinkedIn-sourced (commercially interested)."
    },
    {
      "id": "T13",
      "claim": "BRAND IS A STOCK, NOT A FLOW: mental availability accumulates (Nerlove-Arrow) and is HOW you reach the Day-1 shortlist. Brand stock B_r -> buyer-state B -> outcome is the real 'Tier 0' causal chain. Demand-creation is not a lagged flow inside attribution (T6) but a stock that generates it.",
      "from_equation": "brand_stock",
      "load_bearing": true,
      "contested": true,
      "evidence_caveat": "Adstock illusion: a long-lag flow can mimic the stock; needs a media-pause experiment to distinguish. alpha/delta unmeasured for SaaS."
    },
    {
      "id": "T14",
      "claim": "PLG IS A LOOP, NOT A FUNNEL, AND k<1 ALWAYS BINDS: B2B virality never sustains k>=1, so the divergent closed form is a non-binding limit; PLG is a highly efficient channel, not perpetual motion. Retention is upstream of virality (k_eff=k_raw*R), and expansion-as-acquisition breaks NRR's intra-account assumption.",
      "from_equation": "plg_loop",
      "load_bearing": false,
      "contested": false
    },
    {
      "id": "T15",
      "claim": "[AGENTIC, v2.0 RECALIBRATED] Agentic GTM is technically calibrated but commercially over-claimed. Cost-per-ACTIVITY collapses via model-routing; cost-per-QUALIFIED-OPPORTUNITY net of deliverability decay and 15-20hr/wk operator maintenance is UNPROVEN (no public holdout). The autonomous-AI-SDR thesis is falsified in public (11x collapse; ZoomInfo 'worse than our SDRs'); capital flipped to AE-augmentation (Rox $1.2B by REJECTING SDR-replacement). Hybrid beats autonomous; Tier-3 stays human in all cases.",
      "from_equation": null,
      "load_bearing": true,
      "contested": false,
      "evidence_caveat": "Architecture patterns triangulate across ~70 cases; outcome magnitudes mostly self-reported, no holdouts."
    },
    {
      "id": "T16",
      "claim": "AGENTS WORK THE LEAST-LEVERAGED 5%: because ~95% of value creation is pre-contact (buyer-state + brand), agents deployed almost entirely in the seller funnel optimize the visible tail. BEAR: they polish the harvested ~5% while brand/shortlist go unaddressed, degrading coefficients via signal-exhaustion. BULL: agents are the only scalable way to nurture the 95% out-of-market into future shortlist position. Which dominates is unproven; today's tooling makes the bear case empirically closer.",
      "from_equation": "buyer_state_coupling",
      "load_bearing": true,
      "contested": true
    },
    {
      "id": "T17",
      "claim": "MAXIMAL-BUT-TAGGED: a complete map is only safe if its legend is honest. v3.0's coverage is bought with falsifiability; the measurement_status tag on every construct is the sole thing preventing the model from explaining any outcome after the fact.",
      "load_bearing": true,
      "contested": false
    },
    {
      "id": "T18",
      "claim": "REALIZED < DESIGNED: realized growth is min(GTM physics, organizational alignment bandwidth). In most firms the binding constraint is alignment bandwidth, not lead quality — so the org_cognition layer often dominates the equation layer.",
      "from_layer": "org_cognition",
      "load_bearing": true,
      "contested": true
    },
    {
      "id": "T19",
      "claim": "REFLEXIVITY, NOT DRIFT: a tactic's coefficient is highest before imitation and decays toward the market mean as it is copied. The 'half-life of a GTM strategy' is an endogenous reflexive effect, measurable as effect-size decay, not exogenous noise.",
      "from_layer": "competitive_ecology",
      "load_bearing": true,
      "contested": false
    },
    {
      "id": "T20",
      "claim": "THE SUBSTRATE IS NOT NEUTRAL: what the CRM measures, the org optimizes (Goodhart). Observability mutates the system observed; therefore the measurement protocols the model depends on are themselves interventions.",
      "from_layer": "substrate_epistemics",
      "load_bearing": true,
      "contested": false
    },
    {
      "id": "T21",
      "claim": "PHI-VECTOR > PHI-SCALAR, BUT ONLY IF MEASURED: decomposing PMF into sub-factors adds rigor only where each sub-factor is independently observable. Today only retention pull is; the rest remain hypotheses. Decomposition without measurement multiplies the ghost, it does not exorcise it.",
      "from_layer": "phi_decomposition",
      "load_bearing": true,
      "contested": true
    },
    {
      "id": "T22",
      "claim": "TOPOLOGY VS DRIFT: the dangerous regime change is the one that rewrites the graph (AI search reshaping discoverability), not the one that shifts a coefficient. Misclassifying topology change as drift is fatal precisely in the highest-stakes periods.",
      "from_layer": "regime_dynamics",
      "load_bearing": true,
      "contested": false
    },
    {
      "id": "T23",
      "claim": "AGENTS ARE LEVERAGE THEN POLLUTION: agentic GTM is alpha to the first mover and equilibrium degradation at scale — homogenized messaging, inflated noise floor, anti-agent defenses. Marginal GTM alpha from agents decays as adoption rises.",
      "from_layer": "agent_environmental_effects",
      "load_bearing": true,
      "contested": true
    },
    {
      "id": "T24",
      "claim": "POWER LAW BREAKS THE AVERAGE: under revenue/rep/channel concentration, mean-coefficient reasoning and the naive Bottleneck Theorem mislead. Optimization must be tail-aware; a few units carry the outcome.",
      "from_layer": "power_law",
      "load_bearing": true,
      "contested": false
    },
    {
      "id": "T25",
      "claim": "NAME THE OBJECTIVE OR OPTIMIZE INCOHERENTLY: growth, efficiency, survival, positioning, and optionality conflict. Without an explicit regime-conditional weight vector, the system silently optimizes whatever the comp plan rewards.",
      "from_layer": "objective_function",
      "load_bearing": true,
      "contested": false
    },
    {
      "id": "T26",
      "claim": "CREATION HAS DIFFERENT PHYSICS: in category-creation mode the funnel/Phi machinery is largely inapplicable; outcomes come from narrative and reality-construction. The model is strong on playing the game and weak on making it — and says so.",
      "from_layer": "category_formation",
      "load_bearing": false,
      "contested": true
    },
    {
      "id": "T27",
      "claim": "THE WALK IS THE ONLY INSTANCE: the MRR conservation law is the subscription case of a family. Usage, transaction, and marketplace models need different stocks and break the ACV-driven motion math. The economics tier is revenue-model-conditional.",
      "from_layer": "business_model_generalization",
      "load_bearing": true,
      "contested": false
    },
    {
      "id": "T28",
      "claim": "POROUS FOR HUMANS, GATED FOR AGENTS: the tier boundary is normative, not descriptive — execution constantly forces strategy revision, and that feedback loop is the real engine. The safety property is preserved by letting agents PROPOSE Tier-3 changes that humans ratify, never enact unilaterally.",
      "from_layer": "cross_tier_coupling",
      "load_bearing": true,
      "contested": false
    },
    {
      "id": "T8.1",
      "claim": "[AGENTIC, C6+ EXTENSION] The behavioral tier an agent executes is governed by TWO separable levers, not one: how much of the chain it runs (autonomy alpha) and how mature the governance around it is (G). Raising alpha without raising G does not merely fail to help — it can raise effective cost (E-A4).",
      "type": "architecture",
      "from_equation": "reviewer_fatigue_catchrate",
      "load_bearing": true,
      "contested": false,
      "evidence_caveat": "Refines T8. Mechanistic; depends on E-A2's gamma_max/rho calibration, currently unmeasured."
    },
    {
      "id": "T29",
      "claim": "[AGENTIC, C6+ EXTENSION] Governance maturity is not a safety tax on autonomy; it is a PRECONDITION for autonomy to lower cost at all. Under low G the fatigue term rho makes review a poor investment and cost-minimizing alpha drifts toward full autonomy despite worse reliability; under high G, review pays for itself and cost-minimizing alpha drifts toward heavy oversight at a LOWER absolute cost floor. Which side of the G threshold a vendor sits on decides the strategy, not the nominal autonomy level.",
      "type": "architecture",
      "from_equation": "agentic_effective_cost",
      "load_bearing": true,
      "contested": false,
      "evidence_caveat": "Sign-flip demonstrated in a worked example with illustrative (not vendor-calibrated) parameters; gamma_max(G), rho(G) are the model's highest-priority unmeasured governance parameters."
    },
    {
      "id": "T30",
      "claim": "[AGENTIC, C6+ EXTENSION] Agent adoption carries a category externality (E-A7) that no firm-level governance internalizes. 'Agents are leverage then pollution' (T23) is not just diminishing returns to one firm's adoption — it is degradation of a SHARED resource (buyer attention / signal-to-noise) with AGGREGATE category adoption kappa, so first-mover alpha is structurally time-limited regardless of any single vendor's execution quality.",
      "type": "architecture",
      "from_equation": "category_noise_floor",
      "load_bearing": true,
      "contested": false,
      "evidence_caveat": "Structural analogue to the brand-stock adstock; the category noise-floor series theta is not yet instrumented even directionally."
    }
  ],
  "failure_modes": [
    {
      "name": "Identity/correlation conflation",
      "description": "Treating a regime or correlational edge with the confidence of an identity. The original sin. Mitigation: every edge carries epistemic_type; counterfactuals are gated by type."
    },
    {
      "name": "Optimizing the confounded proxy",
      "description": "Chasing NRR/engagement to 'cause' revenue when both are driven by Phi. Mitigation: never run a counterfactual on a correlational edge."
    },
    {
      "name": "Ratio over payback",
      "description": "Celebrating 5:1 LTV:CAC while running out of cash. Mitigation: payback is the binding constraint at venture stage."
    },
    {
      "name": "Comfortable-stage optimization",
      "description": "Improving the funnel stage you're already good at instead of the bottleneck. Mitigation: bottleneck theorem; always intervene at the min-conversion stage."
    },
    {
      "name": "Brand over-cut",
      "description": "Cutting demand-creation because last-touch shows no ROI. Mitigation: model the convolution kernel; use self-reported attribution."
    },
    {
      "name": "Tactics-before-fit",
      "description": "Scaling GTM machinery below threshold Phi. Mitigation: PMF gate before motion scale-up. The most expensive failure mode."
    },
    {
      "name": "Static-model rot",
      "description": "Trusting fitted coefficients after the regime has moved (channel saturated, competitor copied you, macro shifted). Mitigation: regime-tier coefficients carry half-lives and are re-estimated; treat drift as a bug (LangChain's CI-eval principle)."
    },
    {
      "name": "Agent overreach",
      "description": "Letting an execution agent silently drift Tier-3 configuration (ICP/motion/pricing) or act without human-in-the-loop on relationship-destroying touches. Mitigation: the agent_interface boundary — execute Tier 1, read Tiers 2-3, never rewrite them."
    },
    {
      "name": "Ungoverned agent spend",
      "description": "Agent execution cost (high, usage-based) silently inflates CAC. Mitigation: Agent FinOps tracking per task/account; the FinOps->CAC edge is real."
    },
    {
      "name": "Compounding-error blast radius",
      "description": "A small early agent misjudgment cascades through a multi-step chain into a relationship-damaging send. Mitigation: human-in-the-loop checkpoints + data-quality gates + escalation paths interrupt the chain."
    },
    {
      "name": "POC abandonment",
      "description": "~30% of GenAI projects die after POC on poor data quality / unclear value. Mitigation: phased rollout starting with the lowest-risk highest-signal loop, proven against a manual control group."
    },
    {
      "name": "Treating agents as automation",
      "description": "Buying a 'standalone agent' expecting OS-level breadth/governance, or wiring agents as if-this-then-that. Mitigation: the maturity ladder — match the rung to the need; agents reason, automation triggers."
    },
    {
      "name": "Identity laundering (the v1.1 original sin, named)",
      "description": "Tagging LTV/NRR/CAC-payback as 'identities' when they are assumption-dependent estimators. Inflates certainty and propagates upward into a circular PMF thesis. Mitigation: only the MRR walk is identity; everything else carries its hidden_assumption."
    },
    {
      "name": "Regime-blind benchmarking",
      "description": "Applying a ZIRP-era benchmark (e.g. 24-mo payback is fine) in a high-rate regime. Mitigation: Psi gates which objective weighting and payback target apply."
    },
    {
      "name": "Funnel-causal fallacy",
      "description": "Reading a Tier-1 conversion coefficient as seller skill and scaling headcount to 'double pipeline', when it is a mixture over buyer-state. Causes geometric decay into a fixed in-market pool. Mitigation: condition on B; de-confounded estimand."
    },
    {
      "name": "Brand-as-flow error",
      "description": "Cutting brand spend because attribution shows low last-touch credit, treating a stock as a flow. Mitigation: B_r is a stock with slow decay; judge via share-of-search/CEP breadth, not last-touch."
    },
    {
      "name": "Phi circularity",
      "description": "Proving PMF with the NRR/retention it is supposed to predict. Mitigation: fix Phi ex-ante (week-8-12 retention floor or t-1 third-party satisfaction); run the nested regression -- or admit it is unrun and label T7 a heuristic."
    },
    {
      "name": "Agentic over-claim (cost-per-activity vs cost-per-opportunity)",
      "description": "Citing collapsed cost-per-email as a CAC win. Cost-per-qualified-opportunity net of deliverability decay and operator maintenance is unproven. Mitigation: demand a holdout; treat uncontrolled ROI multiples as marketing math."
    },
    {
      "name": "Optimizing the visible 5%",
      "description": "Pointing agents entirely at the seller funnel while ~95% of the decision is pre-contact. Mitigation: ask whether the agent moves B_r or B, not just closes meetings faster."
    },
    {
      "name": "Maximal-map mistaken for surveyed territory",
      "description": "Reading v3.0's comprehensive coverage as uniform reliability. Mitigation: obey the measurement_status tag — never feed an unmeasurable_hypothesis as a number into a decision."
    },
    {
      "name": "Automated the wrong playbook",
      "description": "Agents executing a stale Tier-3 config flawlessly because the execution->strategy feedback loop was severed. Mitigation: cross_tier_coupling — agents propose, humans ratify, loop stays live."
    },
    {
      "name": "HITL rubber-stamp rot",
      "description": "Reviewer alert fatigue turns human approvals into noise, corrupting the learning signal. Mitigation: down-weight approvals when approve-rate and downstream quality diverge (learning_layer)."
    },
    {
      "name": "Agent equilibrium degradation",
      "description": "Everyone's agents homogenize outreach and inflate the noise floor, decaying marginal alpha. Mitigation: track reply-rate vs market send volume; differentiate or retreat from saturated channels."
    },
    {
      "name": "Ontology drift unnoticed",
      "description": "Stage definitions inflate under quota pressure; the same label now means something weaker. Mitigation: monitor conversion-distribution shift under a fixed label (ontology_instability)."
    },
    {
      "name": "Average-case blindness under power law",
      "description": "Mean LTV / mean r_i reasoning when a few units carry the base. Mitigation: tail-aware metrics; treat top-decile concentration as tail risk, not strength."
    },
    {
      "name": "Objective incoherence",
      "description": "Optimizing conflicting goals at once (short-term CAC vs brand, ACV vs PLG loops) with no stated weights. Mitigation: make the regime-conditional objective vector explicit (objective_function)."
    },
    {
      "name": "Category myopia",
      "description": "Optimizing the funnel inside a game a competitor is busy redefining. Mitigation: watch branded-category-search share; know whether you are in creation or optimization mode."
    }
  ],
  "measurement_gap_register": {
    "purpose": "Marks the boundary between honest-but-unparameterized and operational. These parameters the rebuilt model NEEDS but CANNOT currently measure from public data. All four construction reports converge on this list.",
    "unmeasurable_from_public_data": [
      {
        "param": "Ex-ante Phi panel",
        "consequence": "T7's nested regression (b3 sign = multiplicative vs additive) cannot be run; T7 stays a localized heuristic until a private aggregator (Bessemer, SaaS Capital) collects independent t-1 satisfaction + subsequent growth + GTM spend."
      },
      {
        "param": "Switching-cost S at account level",
        "consequence": "Regime boundary (multiplicative vs additive) can't be set directly; proxied via GRR>=95% + low-satisfaction confounder."
      },
      {
        "param": "Brand alpha (build rate) and delta (decay) for SaaS",
        "consequence": "Brand-stock dynamics are directional only; flagged retrospective-only / forecasting liability."
      },
      {
        "param": "Sustained k>=1 in B2B",
        "consequence": "No public example exists; the divergent PLG closed form is a non-binding limiting case."
      },
      {
        "param": "Noise-floor theta as a stable constant",
        "consequence": "25-40% is a reasoned assumption (Gross/Taleb), not a fitted 1 - out-of-sample-R^2 for B2B SaaS."
      },
      {
        "param": "Controlled CAC under agentic GTM",
        "consequence": "No public holdout exists; whether agents lower cost-per-qualified-opportunity net of decay+maintenance is unproven."
      },
      {
        "param": "dCAC_agentic (E-A8)",
        "consequence": "No public holdout exists; whether an agentic motion lowers cost-per-qualified-opportunity net of deliverability decay + maintenance is unproven. Single most consequential hole for ROI-based GTM messaging.",
        "what_it_would_take": "A matched-cohort controlled experiment: identical ICP/segment split into agentic vs manual arms, tracked to cost-per-qualified-opportunity net of deliverability decay and operator-maintenance hours, over a full sales-cycle horizon.",
        "why_unmeasured_today": "T15's review found the autonomous-AI-SDR case (11x/ZoomInfo) publicly collapsed before any credible holdout ran; every vendor figure is self-reported, not independently audited.",
        "closest_existing_proxy": "McKinsey's 39%-report-EBIT-impact figure (directional, not causal, not agentic-GTM-specific).",
        "priority": "Highest — Gap 2 from prior synthesis."
      },
      {
        "param": "gamma_max(G), rho(G) (E-A2)",
        "consequence": "T29's sign-flip result cannot be calibrated for any real deployment; the optimal-autonomy decision stays illustrative.",
        "what_it_would_take": "Instrumented review logs across governance tiers: reviewer catch-rate as a function of review-queue volume, at firms independently scored on a governance-maturity rubric.",
        "why_unmeasured_today": "No surveyed source publishes reviewer catch-rate by review load or governance tier; McKinsey's survey measures control ADOPTION, not efficacy.",
        "closest_existing_proxy": "McKinsey ~30%-at-maturity-level-3+ (measures governance adoption, not catch-rate efficacy).",
        "priority": "High."
      },
      {
        "param": "theta, delta_theta (E-A7, category noise floor)",
        "consequence": "T23/T30's category-externality claim is not yet even directionally instrumented.",
        "what_it_would_take": "A time series of buyer-side response/engagement rates to outbound signals, segmented by category-wide agent-adoption share kappa, across a full multi-year adoption cycle to fit an adstock decay rate.",
        "why_unmeasured_today": "No source tracks a category-wide noise floor directly; CB Insights' 4,700% YoY referral growth measures adoption VOLUME, not per-unit signal degradation.",
        "closest_existing_proxy": "a16z's qualitative brand/trust-decisiveness framing (directional, not a measured series).",
        "priority": "Medium."
      },
      {
        "param": "lambda (E-A6, HITL learning-rate)",
        "consequence": "Cannot validate whether T9's data-collection framing is operationally real or aspirational.",
        "what_it_would_take": "Tracking of how many human-reviewed agent actions are actually fed back into r_i retraining pipelines vs discarded after one-off review, across multiple vendor implementations.",
        "why_unmeasured_today": "The review-as-safety-gate vs review-as-training-signal distinction is not tracked as a separate metric in any surveyed source.",
        "closest_existing_proxy": "None identified.",
        "priority": "Medium."
      },
      {
        "param": "B_agent (E-A9, buying-committee agent-evaluator share)",
        "consequence": "Cannot test whether T12's ~95% pre-contact primacy holds when the buying committee itself includes AI evaluators (Gap 7).",
        "what_it_would_take": "A study of B2B buying-committee composition for agentic-platform purchases, tracking whether/how AI tools participate in shortlisting, evaluation, negotiation, and whether the ~95% pre-contact figure holds there.",
        "why_unmeasured_today": "2025-26 agent-buying preprints (Allouah et al.; role-coherence paper) study individual shopping agents, not B2B buying-committee dynamics.",
        "closest_existing_proxy": "T12's general (non-agent-specific) buyer-state primacy finding (95-5 Rule, 6sense).",
        "priority": "Medium — Gap 7 from prior synthesis."
      },
      {
        "param": "kappa_gov(G) (E-A4, failure-cost absorption)",
        "consequence": "E-A4's cost-floor estimate lacks its governance-absorption input; secondary to calibrating gamma_max/rho first.",
        "what_it_would_take": "Incident post-mortems across governed vs ungoverned deployments, quantifying what fraction of a failure's realized cost (rework, reputational, churn) governance tooling actually mitigated vs absorbed in full.",
        "why_unmeasured_today": "No surveyed source reports incident-level cost breakdowns for agentic-GTM failures; failure-mode annotations are qualitative, not costed.",
        "closest_existing_proxy": "None identified.",
        "priority": "Low-medium."
      }
    ],
    "honest_terminal_status": "The model is operational inside measured regimes and explicitly refuses to overclaim outside them. T7 in particular is specified and falsifiable IN PRINCIPLE but unrun IN PRACTICE."
  },
  "open_tensions_v3": {
    "purpose": "The maximal version owes an explicit accounting of what completeness cost. These are not resolved.",
    "tensions": [
      "ADD vs SUBTRACT: v3.0 chose completeness; the disciplined critique (doc 10) chose parsimony. By that standard v3.0 is LESS falsifiable than the minimal core. The measurement_status tags are a mitigation, not a resolution.",
      "MAP vs ENGINE: v3.0 is a far more complete MAP, but adding seventeen layers moved it further from a runnable ENGINE, not closer. Only the MRR-walk slice (gtm_engine_slice.py) executes. The ratio of spec-to-machinery got worse, exactly as the harshest critiques predicted.",
      "PHI: decomposed but still mostly unmeasurable. Eight sub-factors, one (retention pull) currently measurable. The ghost-variable risk is reduced in principle and unchanged in practice.",
      "LEARNING: specified, not built. The update rule answers the 'silicon BDR' critique on paper; it still needs experimental traffic most ventures cannot spare, and full causal-graph revision is unimplemented.",
      "COVERAGE vs USABILITY: the operator layer (Q) is the bet that this collapses to a few levers. If it does not, v3.0 is a monograph operators admire and route around — the precise fate the critiques warned of."
    ],
    "the_honest_one_liner": "v3.0 is the most complete thing we have built and the least falsifiable. Whether that is the right trade depends entirely on whether you want a map of the whole territory or a vehicle that drives."
  }
}