SPERT® Forecaster

Agile Release Forecasting with Monte Carlo Simulation

Forecast how many sprints a backlog will take, as a probability rather than a single date. Statistical PERT® models uncertain outcomes with probability distributions and Monte Carlo simulation, working from recorded sprint history or from subjective estimates when none exists yet.

forecaster.spertsuite.com

What it does

• Monte Carlo simulation, 10,000 trials by default and up to 50,000, across six distributions: truncated normal, lognormal, gamma, triangular, uniform, and bootstrap sampling from your own sprint history
• Percentile results (P50 through P90, plus any custom percentile), with burn-up, cumulative distribution and histogram charts and a deadline probability panel
• Ordered milestones forecast individually, and productivity adjustments that scale velocity over named date ranges such as holidays
• CSV export, a print-friendly report, and JSON project import/export
• Local-first: everything runs in your browser. Optional Google or Microsoft sign-in adds cloud sync and project sharing; a connected AI assistant can be given a read-only snapshot.

Development

npm install
npm run dev          # development server
npm test             # test suite
npm run shipgate     # release gate: version surfaces, lint, typecheck, tests, build

Node 24 (see .nvmrc). Firebase environment variables are optional — without them the app runs in local-only mode.

Legal

Reference copies of the Terms of Service and Privacy Policy are in /legal. The canonical versions used by the app at runtime are hosted at:

• https://spertsuite.com/TOS.pdf
• https://spertsuite.com/PRIVACY.pdf

License

GNU General Public License v3.0, with non-permissive additional terms under Section 7.

SPERT® and Statistical PERT® are registered trademarks with the United States Patent and Trademark Office.