Forecast focus fragmentation delivery economics

What it's for

Turns meeting and context-switching signals into the executive question buyers care about—how long current work may still take and what the delay tail may cost—while refusing an unvalidated company-local model.

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.

{
  "current_tasks": [
    {
      "age_days": 2,
      "context_id": "api",
      "context_switches_per_week": 4,
      "evidence_verified": true,
      "fragmentation_index": 0.3,
      "id": "current-api",
      "meeting_fraction": 0.2,
      "size_points": 5,
      "value_per_day": 1000
    }
  ],
  "historical_tasks": [
    {
      "context_id": "api",
      "context_switches_per_week": 0,
      "duration_days": 2.4715868294107564,
      "evidence_verified": true,
      "fragmentation_index": 0.05,
      "id": "task-0-0",
      "meeting_fraction": 0.05,
      "outcome_mature": true,
      "period_index": 0,
      "size_points": 1
    },
    {
      "context_id": "product",
      "context_switches_per_week": 1,
      "duration_days": 6.528949200899003,
      "evidence_verified": true,
      "fragmentation_index": 0.2,
      "id": "task-0-1",
      "meeting_fraction": 0.09,
      "outcome_mature": true,
      "period_index": 0,
      "size_points": 2
    },
    {
      "context_id": "api",
      "context_switches_per_week": 2,
      "duration_days": 6.4712794686704385,
      "evidence_verified": true,

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.

{
  "association_coefficients": {
    "fragmentation_index": 1.3166,
    "log_context_switches_per_week": 0.2129,
    "log_size_points": 0.6868,
    "meeting_fraction": 1.4954
  },
  "configuration": {
    "ridge_penalty": 5,
    "scenario_count": 2,
    "seed": 7,
    "simulations": 2000,
    "tail_probability": 0.95
  },
  "decision": "forecast_available",
  "detail_truncated": false,
  "evidence_diagnostics": {
    "immature_historical_tasks": 0,
    "unsupported_current_contexts": [],
    "unverified_current_tasks": 0,
    "unverified_historical_tasks": 0
  },
  "interpretation": "Attention variables are company-local predictive covariates, not causal effects or individual productivity scores. Changing meetings requires separate prospective intervention evidence.",
  "method": "company_local_ridge_lognormal_aft_latest_period_validation_conditional_survival_monte_carlo",
  "model_validation": {
    "candidate_beats_baseline": true,
    "candidate_duration_mae_days": 0.201,
    "candidate_log_score": 1.0816,
    "duration_mae_improvement": 2.0669,
    "eighty_percent_interval_coverage": 1,
    "global_duration_mae_days": 2.2679,
    "global_log_score": 2.4622,
    "holdout_period": 4,
    "holdout_tasks": 12,
    "log_score_improvement": 1.3806,
    "supported_contexts": 2,
    "training_periods": 4,
    "training_tasks": 48
  },
  "summary": {
    "delay_cost_cvar": 28836.7878,
    "delay_cost_var": 26290.3902,
    "excluded_tasks": 0,
    "expected_delay_cost": 14739.5393,

How it works

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

1. 1 Keep the latest complete period untouched; fit a company-local ridge log-normal AFT model on task size, meeting fraction, fragmentation, context-switch rate and shrunk context effects, alongside a global log-normal baseline.
2. 2 Require later-period improvement in duration MAE and log score plus minimum interval coverage and local-context support; hold or omit unsupported contexts rather than transferring another company's coefficients.
3. 3 Condition each current task's duration distribution on survival beyond its observed age, replay coherent duration/cost scenarios with a common draw, and return expected remaining time plus portfolio delay-cost VaR/CVaR.

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 currenttask completion and delaycost tails" }
  → finds "forecast_focus_fragmentation_delivery_economics"

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

gitrevio_capability_run
  { "capability_id": "forecast_focus_fragmentation_delivery_economics", "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 People, retention & knowledge →