Forecast delivery to cash conversion

What it's for

Forecasts when delivered technical value is likely to become cash—and whether acceptance or invoicing delay threatens liquidity—using censoring-aware local evidence rather than a static milestone count.

What you give it

Inputs split into evidence read from your connected systems, calibration your team owns, and numerical controls that affect precision but never the result's meaning.

{
  "completed_stage_episodes": [
    {
      "cohort": "enterprise",
      "duration_periods": 2,
      "evidence_verified": true,
      "id": "delivery_ready-0",
      "outcome": "advance",
      "stage": "delivery_ready",
      "start_age_periods": 0
    },
    {
      "cohort": "enterprise",
      "duration_periods": 2,
      "evidence_verified": true,
      "id": "delivery_ready-1",
      "outcome": "advance",
      "stage": "delivery_ready",
      "start_age_periods": 1
    },
    {
      "cohort": "enterprise",
      "duration_periods": 2,
      "evidence_verified": true,
      "id": "delivery_ready-2",
      "outcome": "exit",
      "stage": "delivery_ready",
      "start_age_periods": 2
    },
    {
      "cohort": "enterprise",
      "duration_periods": 2,
      "evidence_verified": true,
      "id": "delivery_ready-3",
      "outcome": "censored",
      "stage": "delivery_ready",
      "start_age_periods": 0
    },
    {
      "cohort": "enterprise",
      "duration_periods": 2,
      "evidence_verified": true,
      "id": "delivery_ready-4",
      "outcome": "advance",

What you get back

This is the actual output of running the example above — computed by the same function the platform calls, not an illustration.

{
  "configuration": {
    "horizon_periods": 4,
    "prior_advance_probability": 0.2,
    "prior_exit_probability": 0.03,
    "prior_strength": 10,
    "scenario_probability_sum_before_normalization": 1,
    "seed": 7,
    "simulations": 100
  },
  "decision": "delivery_to_cash_forecast_supported",
  "forecast": {
    "cash_collected": {
      "mean": 77.73,
      "p10": 4,
      "p50": 80,
      "p90": 180
    },
    "liquidity_breach_probability": 0,
    "minimum_unrestricted_cash": {
      "mean": 99.5,
      "p10": 97.8,
      "p50": 100,
      "p90": 100
    },
    "remaining_value_by_stage_mean": {
      "accepted": 40,
      "delivery_ready": 8.4,
      "invoiced": 58
    }
  },
  "guardrails": [
    "Stage histories must be complete right-censored episodes; snapshots bias conversion hazards.",
    "Cohorts must be lawful aggregate operating segments, never named-customer credit scores.",
    "Scenario effects require governed evidence and local backtesting; language models must not invent transition effects.",
    "Accounting recognition and customer acceptance remain authoritative-system facts, not inferences from Git activity."
  ],
  "method": "right_censored_empirical_bayes_multistage_delivery_to_cash_simulation",
  "support": {
    "completed_episode_count": 90,
    "current_milestone_count": 3,
    "evidence_complete": true,
    "exposure_periods_by_stage": {
      "accepted": 60,

How it works

Forecasting & survival — Estimate when something completes or fails, with censoring and unresolved work handled honestly rather than dropped.

1. 1 Expand every resolved or censored stage episode into at-risk periods and an optional terminal advance/exit event; never relabel censoring as failure.
2. 2 Estimate cohort-age advance and exit hazards with stage-pooled empirical-Bayes shrinkage, then gate all three stages on exposure support and authoritative evidence.
3. 3 Simulate each current milestone sequentially through delivery-ready, accepted, invoiced and paid states under shared scenarios; credit cash only at payment or governed invoice-exit recovery and preserve period-by-period liquidity paths.

Before you trust it

Every tool in the catalog ships with the conditions under which its answer is meaningful — and the conditions under which it should abstain instead of guessing.

Calibrating it to your org

Calibration workflow

1. 1 Define the management decision, target outcome, aggregate unit, privacy boundary, cadence, and prediction/intervention horizon for this organization.
2. 2 Build a tenant-scoped historical cohort using only information available before each prediction or decision; preserve zero periods, censoring, assignment probabilities, and unresolved outcomes when the method requires them.
3. 3 Estimate statistical parameters on training history, but obtain costs, utilities, risk tolerance, practical-effect thresholds, capacity, and policy constraints from accountable decision owners.
4. 4 Validate on later time windows or held-out aggregate units at the deployment grain, against a simple baseline and the function-specific validation strategy.
5. 5 Deploy only if the returned decision clears evidence, overlap, calibration, robustness, and guardrail checks; warning, unsupported, schema-gap, and fallback decisions are abstentions.
6. 6 Monitor realized outcomes, data drift, coverage, and decision regret; recalibrate at a governed cadence or after a detected regime/definition change, never merely because a stakeholder dislikes the result.

Call it from your AI

You don't wire up 388 tools in your MCP client. The GitRevio MCP server exposes 18 tools, three of which let an agent search the catalog, read a tool's schema, and run it — so the assistant finds this one on its own.

gitrevio_capabilities_search
  { "q": "forecast how deliveryready accepted and invoiced" }
  → finds "forecast_delivery_to_cash_conversion"

gitrevio_capability_describe
  { "capability_id": "forecast_delivery_to_cash_conversion" }
  → returns the input schema and agent guidance shown on this page

gitrevio_capability_run
  { "capability_id": "forecast_delivery_to_cash_conversion", "arguments": { ... } }
  → returns the result shown above

Works in Claude Desktop, Claude Code, Cursor, Cline, Continue.dev, Goose and Aider. See the MCP server.

Related tools

See every tool in Finance & unit economics →