Overview
This Data Backup and Recovery Policy ("Policy") describes how Nouvelle Education, Inc. ("Nouvelle") protects platform data, what recovery commitments Nouvelle makes to customers, and what responsibilities customers retain under the Master Services Agreement ("MSA").
This Policy applies to the production Nouvelle platform environment and supplements the obligations set out in Schedule A of the MSA.
Backup Cadence
Nouvelle maintains the following automated backup schedule for the production database:
- Daily automated snapshots — taken at 03:00 UTC each day
- Snapshot retention — 30 daily snapshots retained on a rolling basis
- Point-in-Time Recovery (PITR) — transaction logs retained for 7 days, enabling recovery to any second within that window
- Nightly archive — a full export of the production database taken at 02:00 UTC each night; each night's copy is retained for 365 days and then automatically deleted
- Backup storage — all backups are stored in Google Cloud Storage, separate from the primary database instance
The nightly archive is stored in a private, encrypted Google Cloud Storage bucket, accessible only to authorized Nouvelle personnel, and is used only to recover from data loss or corruption.
Backups are verified automatically by the cloud provider on creation. Nouvelle performs restore verification tests at least annually.
Recovery Objectives
Nouvelle targets the following recovery metrics for production platform failures:
- Recovery Time Objective (RTO): 8 hours — the maximum acceptable time to restore the platform to operational status following a failure
- Recovery Point Objective (RPO): 24 hours — the maximum acceptable data loss window; in practice, PITR delivers near-zero data loss for events within the 7-day log retention window
These objectives apply to infrastructure-level failures. Recovery times for application-level errors (bad deployments, data corruption) may differ based on the nature of the incident.
Customer Responsibilities
As set out in Schedule A of the MSA, customers are responsible for:
- Maintaining their own backup copies of any data they export from the platform
- Ensuring Authorized Users do not store critical records exclusively within the Nouvelle platform without independent backup
- Notifying Nouvelle promptly if data loss or corruption is suspected, to preserve the recovery window
Nouvelle's backup infrastructure protects against infrastructure failures and data corruption. It does not protect against user-initiated deletion of records within the platform unless the deletion is reported within the PITR window. Records deleted earlier may be recoverable from the nightly archive for up to 365 days; such recovery is performed on a best-effort basis and is not covered by the recovery objectives in Section 3.
Disaster Recovery Process
In the event of a platform-impacting failure, Nouvelle follows this recovery sequence:
- Detection — automated uptime monitoring alerts the on-call engineer within 60 seconds of a health check failure
- Assessment — engineer classifies the incident (infrastructure, application, or data) and determines recovery path
- Restore — for database failures, engineer initiates PITR restore to the most recent clean point; for application failures, engineer rolls back to the last known-good deployment revision
- Verification — smoke tests confirm platform functionality before traffic is restored
- Customer notification — affected customers are notified per the Incident Response Policy
Scheduled Maintenance Exclusions
Downtime that occurs during pre-announced scheduled maintenance windows does not count against RTO/RPO commitments or SLA uptime calculations. Nouvelle provides at least 24 hours' advance notice for scheduled maintenance and at least 4 hours' advance notice for emergency maintenance. See the Scheduled and Emergency Maintenance Policy for full details.
Policy Review
This Policy is reviewed annually and updated following any significant infrastructure change, recovery incident, or audit finding. Questions may be directed to security@nouvelleed.com. Related policies: Incident Response & Service Suspension Policy, Scheduled & Emergency Maintenance Policy, Data Retention Policy.