PolicyWorkspacePlanning lab

How PolicyWorkspace handles planning data

PolicyWorkspace supports calculators, estimators, guides, and workflows. This notice explains the product-agnostic data model: browser drafts, backend records for saved work, account flows, exports, events, and product-specific reproducibility metadata.

Product inputs can include names you enter for records, category selections, locations, amounts, time horizons, assumptions, notes, and product-specific details needed to produce a planning record.

Draft records, saved comparison state, local access state, and usage counters can live in browser storage so you can continue a workflow without creating an account.

When you sign in, the backend can store saved planning records, exports, team invitations, member roles, API key metadata, account confirmation state, and one-time authentication-token hashes.

Request metadata, rate-limit state, structured events, deployment logs, and abuse-prevention signals support reliability, security, and troubleshooting.

  • Run product workflows and return reproducible planning records.
  • Restore local drafts and sync records you explicitly save to a workspace.
  • Issue account confirmation, magic-link sign-in, and password reset emails.
  • Create exports, maintain access state, and enforce abuse controls.
  • Support team collaboration, workspace history, and read-only API access.
  • Measure product usage and detect failures without turning product content into ad-like controls.
  • Browser-local data stays in the browser until you clear it, reset the app, or the browser removes it.
  • Workspace records live under the configured backend app-data root or the configured deployed persistence backend.
  • Temporary export artifacts can expire after generation; saved workspace records are separate from temporary download storage.
  • Structured event logs follow the retention policy of the deployed logging environment.
  • The currently live business product keeps reproducibility metadata for generated records, including input snapshots, output records, engine version, dataset version/hash, source records, and generated timestamps.
  • Future calculator products should store only the inputs and output records needed for saved scenarios, exports, and auditability.
  • Guide and workflow products should avoid collecting sensitive identifiers unless a later product spec explicitly justifies them.