Forecast customer lifetime value uncertainty

What it's for

Turns retention, margins, and acquisition spend into a full uncertainty distribution for prospective cohort value rather than a fragile single CLV number.

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.

{
  "discount_rate_per_period": 0.01,
  "horizon_periods": 36,
  "minimum_probability_positive_value": 0.8,
  "posterior_draws": 5000,
  "seed": 29,
  "segments": [
    {
      "acquisition_cost_per_customer": 250,
      "churned_observations": 150,
      "id": "self-serve",
      "period_contribution_margin_samples": [
        85,
        90,
        100,
        110,
        120,
        95
      ],
      "prospective_customers": 2000,
      "retained_observations": 850,
      "retention_cost_per_customer_period": 10
    }
  ]
}

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.

{
  "assumptions": [
    "Retention opportunities are comparable, outcome-mature, and exchangeable within each segment; retention is stationary over the forecast horizon unless scenarios are modeled separately.",
    "Positive period contribution margins follow a lognormal family; uncertainty covers beta-binomial retention sampling and lognormal parameter estimation, not model-form or causal-design error.",
    "Prospective segments do not overlap, acquisition and retention costs are incremental, and churn, margin, discounts, currency, cadence, and horizon share one finance definition."
  ],
  "configuration": {
    "beta_prior_alpha": 1,
    "beta_prior_beta": 1,
    "confidence_level": 0.9,
    "discount_rate_per_period": 0.01,
    "horizon_periods": 36,
    "minimum_probability_positive_value": 0.8,
    "posterior_draws": 5000,
    "seed": 29,
    "tail_probability": 0.1
  },
  "decision": "prospective_customer_value_supported",
  "method": "beta_retention_lognormal_margin_clv_v1",
  "segment_diagnostics": [
    {
      "clv_per_customer_interval": [
        234.2836,
        423.3078
      ],
      "expected_clv_per_customer": 323.2137,
      "expected_segment_value": 646427.4346,
      "posterior_mean_period_contribution_margin": 101.1591,
      "posterior_mean_retention_probability": 0.849,
      "probability_positive_clv": 1,
      "prospective_customers": 2000,
      "segment_id": "self-serve"
    }
  ],
  "summary": {
    "expected_portfolio_lifetime_value": 646427.4346,
    "portfolio_cvar_lifetime_value_loss": -458003.3325,
    "portfolio_lifetime_value_interval": [
      468567.1702,
      846615.6815
    ],
    "probability_positive_portfolio_lifetime_value": 1,
    "prospective_customer_count": 2000,
    "segment_count": 1

How it works

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

1. 1 Freeze non-overlapping customer segments, mature retention opportunities, positive period contribution-margin samples, prospective cohort size, acquisition/retention costs, cadence, horizon, and discounting.
2. 2 Draw segment retention posteriors and lognormal margin parameters, evaluate the discounted survival series, subtract incremental costs, and aggregate prospective-cohort CLV draws.
3. 3 Use probability-positive, interval, and CVaR outputs for the commercial decision, while separately stress nonstationary retention, margin regimes, competitive effects, and model-form error.

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 prospective customer lifetime value by" }
  → finds "forecast_customer_lifetime_value_uncertainty"

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

gitrevio_capability_run
  { "capability_id": "forecast_customer_lifetime_value_uncertainty", "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 Customer, revenue & pricing →