Forecast fundraising close and runway
Forecast whether enough primary capital closes before runway pressure by fitting empirical-Bayes age-state competing-risk hazards to advanced, closed, lost and right-censored stage episodes, then simulating every live opportunity under one common market scenario and an explicit burn-before-close cash convention.
What it's for
Connects the fundraising CRM to runway in one answer: how likely target cash is to arrive, whether the company stays above its buffer while waiting, and what tail rescue capital closes the gap.
What you give it
Every tool is a pure function: it computes from what you pass it and reads nothing on its own. Inputs split into evidence you supply from your connected systems, calibration your team owns, and numerical controls that affect precision but never the result's meaning.
| Field | Type | Role | Required |
|---|---|---|---|
| age_bucket_boundaries | array of integer | Evidence | Optional |
| current_opportunities | array of objects (5 fields) ≥ 1 item | Evidence | Yes |
| historical_stage_episodes | array of objects (5 fields) ≥ 4 items | Evidence | Yes |
| horizon_periods | integer ≥ 1, ≤ 260 | Your calibration | Yes |
| initial_unrestricted_cash | number ≥ 0 | Your calibration | Yes |
| market_scenarios | array of objects (6 fields) ≥ 1 item | Evidence | Yes |
| max_detail_rows | integer ≥ 1, ≤ 500 | Numerical control | Optional |
| minimum_liquidity_probability | number ≥ 0, ≤ 1 | Your calibration | Optional |
| minimum_stage_period_exposure | number ≥ 0 | Your calibration | Optional |
| minimum_target_probability | number ≥ 0, ≤ 1 | Your calibration | Optional |
| minimum_unrestricted_cash | number ≥ 0 | Your calibration | Yes |
| prior_strength | number ≥ 0 | Your calibration | Optional |
| seed | integer ≥ 0, ≤ 2147483647 | Numerical control | Optional |
| simulations | integer ≥ 200, ≤ 50000 | Numerical control | Optional |
| stage_order | array of string ≥ 2 items | Evidence | Yes |
| tail_probability | number ≥ 0.001, ≤ 0.5 | Your calibration | Optional |
| target_primary_proceeds | number ≥ 0 | Your calibration | Yes |
Each market_scenarios
record
| Field | Type | Required |
|---|---|---|
| advance_hazard_multiplier | number (≥ 0, ≤ 10) | Yes |
| amount_realization_fraction | number (≥ 0, ≤ 2) | Yes |
| cash_burn_by_period | array of number (≥ 1 item) | Yes |
| id | string (non-empty) | Yes |
| loss_hazard_multiplier | number (≥ 0, ≤ 10) | Yes |
| probability | number (≥ 0, ≤ 1) | Yes |
{
"current_opportunities": [
{
"age_periods": 1,
"current_stage": "diligence",
"evidence_verified": true,
"id": "fund-alpha",
"potential_primary_proceeds": 600
},
{
"age_periods": 0,
"current_stage": "term_sheet",
"evidence_verified": true,
"id": "fund-beta",
"potential_primary_proceeds": 400
}
],
"historical_stage_episodes": [
{
"duration_periods": 1,
"evidence_verified": true,
"id": "contacted-0",
"outcome": "lost",
"stage": "contacted"
},
{
"duration_periods": 2,
"evidence_verified": true,
"id": "contacted-1",
"outcome": "censored",
"stage": "contacted"
},
{
"duration_periods": 3,
"evidence_verified": true,
"id": "contacted-2",
"outcome": "advanced",
"stage": "contacted"
},
{
"duration_periods": 4,
"evidence_verified": true,
"id": "contacted-3",
"outcome": "advanced",Truncated for display — the full payload is 484 lines.
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": {
"age_bucket_boundaries": [
1,
3,
6
],
"horizon_periods": 6,
"period_cash_timing": "burn_then_financing_close",
"simulations": 500,
"tail_probability": 0.1
},
"decision": "fundraising_or_runway_action_required",
"failed_gates": [
"minimum_liquidity_probability"
],
"guardrails": [
"Stage transitions and losses are competing risks with age-bucket hazards shrunk toward each stage's own base rate; censored episodes contribute exposure without fabricated terminal outcomes.",
"Each simulation draws one common market scenario for every live opportunity, preserving shared fundraising conditions. The stated cash convention subtracts period burn before adding closes.",
"Memory within an age bucket, investor dependence beyond the common scenario, negotiation, legal terms, dilution, bridge availability, and omitted cash obligations remain outside this model. This is a conditional planning forecast, not a financing guarantee or solvency opinion."
],
"method": "empirical_bayes_age_state_competing_risks_fundraising_simulation",
"opportunity_diagnostics": [
{
"age_periods": 0,
"current_stage": "term_sheet",
"opportunity_id": "fund-beta",
"potential_primary_proceeds": 400,
"simulated_close_probability": 0.684
},
{
"age_periods": 1,
"current_stage": "diligence",
"opportunity_id": "fund-alpha",
"potential_primary_proceeds": 600,
"simulated_close_probability": 0.336
}
],
"stage_hazard_diagnostics": [
{
"age_bucket": 0,
"loss_events": 1,
"period_exposure": 20,
"posterior_loss_hazard": 0.06,Truncated for display — the full payload is 165 lines.
How it works
Forecasting & survival — Estimate when something completes or fails, with censoring and unresolved work handled honestly rather than dropped.
- 1 Convert historical fundraising histories into stage episodes and person-period exposure, preserving right censoring; estimate age-bucket advance/close and loss hazards with stage-local empirical-Bayes shrinkage.
- 2 For each simulation draw one coherent market regime shared by every opportunity, adjust the competing hazards, advance or absorb each live opportunity, subtract burn before adding realized primary proceeds, and retain the full cash shortfall path.
- 3 Report opportunity close probabilities, target-capital probability, liquidity survival, their joint probability, expected and tail rescue capital, and abstain when evidence or stage exposure is inadequate.
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.
Assumptions & guardrails
- Training examples precede their outcomes, censoring and unresolved work are represented, and deployment populations remain comparable to validation cohorts.
- Stage episodes include every eligible opportunity and correct censoring; stages and period cadence are stable; common market scenarios jointly specify hazard, amount and burn effects; live amounts are primary proceeds; cash obligations and timing are complete.
- A predictive interval or risk estimate is not a deadline promise, causal explanation, or individual-performance judgment.
- The model is conditional on represented stage, market and cash mechanics; it omits unmodeled negotiation, dependence, legal terms, dilution and financing alternatives. It is not a financing guarantee, solvency opinion, investor ranking or securities advice.
Minimum evidence
- historical_stage_episodes: at least 4 rows/items
- current_opportunities: at least 1 rows/items
- stage_order: at least 2 rows/items
- market_scenarios: at least 1 rows/items
- initial_unrestricted_cash: required and organization-defined
- minimum_unrestricted_cash: required and organization-defined
- target_primary_proceeds: required and organization-defined
- horizon_periods: required and organization-defined
How to validate it
Use chronological train/calibration/test windows, compare proper scores and decision value with a simple baseline, and recalibrate only from outcomes resolved after prediction time.
Calibrating it to your org
Same for everyone
The mathematical kernel, validation rules, method version, and JSON output semantics are organization-independent; no tenant-trained coefficients or company benchmark is embedded in the function.
Specific to you
- version-matched CRM event histories transformed into non-overlapping stage episodes with correct right censoring, joined to treasury burn and cash plus finance-owned common market scenarios
- stage/cadence stability, episode eligibility and censoring, market scenario dependence, amount realization, primary-proceeds perimeter, unrestricted cash, complete burn/obligation timing, minimum liquidity, capital target, empirical-Bayes prior, age buckets, simulation precision, probability/tail gates and activation authority
Calibration workflow
- 1 Define the management decision, target outcome, aggregate unit, privacy boundary, cadence, and prediction/intervention horizon for this organization.
- 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 Estimate statistical parameters on training history, but obtain costs, utilities, risk tolerance, practical-effect thresholds, capacity, and policy constraints from accountable decision owners.
- 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 Deploy only if the returned decision clears evidence, overlap, calibration, robustness, and guardrail checks; warning, unsupported, schema-gap, and fallback decisions are abstentions.
- 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 20 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 whether enough primary capital closes" }
→ finds "forecast_fundraising_close_and_runway"
gitrevio_capability_describe
{ "capability_id": "forecast_fundraising_close_and_runway" }
→ returns the input schema and agent guidance shown on this page
gitrevio_capability_run
{ "capability_id": "forecast_fundraising_close_and_runway", "arguments": { ... } }
→ returns the result shown aboveWorks in Claude Desktop, Claude Code, Cursor, Cline, Continue.dev, Goose and Aider. See the MCP server.