Backups need monthly drills, daily verification, and on-call response. None of that gets done reliably as a side project.
The subscription that operates the backup infrastructure DR Setup builds. Continuous attention, contracted rather than dependent on internal bandwidth.

Monthly subscription operating the backup infrastructure deployed by Disaster Recovery Setup. The recurring deliverables: daily verification that all scheduled backups completed successfully with alert on any failure within 4 hours; monthly restore drill into a separate test environment with documented RTO and RPO; AES-256-GCM encryption key rotation on the documented annual schedule; retention policy enforcement across the three-tier hot/warm/cold storage (30/90/365 days); alert response with on-call rotation for backup failures, storage anomalies, or restore requests; restore assistance when needed including coordination with operational team, restore procedure execution, post-restore validation, and incident documentation; quarterly backup health report covering verification statistics, drill outcomes, retention status, and any anomalies. The service prerequisite is Disaster Recovery Setup (EUR 1,299 one-time) which builds the backup infrastructure that this subscription operates. EUR 79 per month, billed monthly or annually with 10% annual discount.

Three operational reasons. First: backup operations are continuous work that ages poorly without dedicated attention. Daily verification, monthly drills, retention enforcement, key rotation, alert response all need to happen reliably over years. Operations that start backup work enthusiastically often quietly degrade as the team gets pulled to other priorities; the backup keeps running but nobody verifies the restore would actually succeed. The subscription model produces sustained attention because the operational responsibility is contracted rather than dependent on internal bandwidth. Second: restore drills require expertise that does not get exercised in normal operations. A monthly drill that successfully restores backups into a test environment catches issues (PostgreSQL version mismatches, missing extensions, file permission problems, encryption key access issues) that would only otherwise surface during a real incident. The drill expertise grows over time with each iteration; running drills once a year in-house produces less reliable outcomes than running them monthly with continuous learning. Third: incident response benefits from operational continuity. A backup failure alert at 3am gets handled by an on-call rotation with documented procedures rather than by whoever happens to be available; restore assistance during a real incident benefits from the team that built the backup infrastructure handling the restore versus someone discovering the procedures for the first time under pressure.

Once per calendar month, we restore your backups into a separate test environment (not production), measure actual restore performance, and document outcomes. The drill cadence: scheduled for the third week of each month to allow week-1 backup completion verification before drill begins; runs against the most recent full backup plus WAL replay to a specific test target time (varied each month so we drill different recovery points across the year); restores PostgreSQL, MailWizz state, PowerMTA configuration, and applies suppression log; runs validation queries against the restored database to confirm row counts, schema integrity, and data consistency; measures RTO from drill start to functional system; measures RPO as the data loss between target time and successful restore completion; produces a monthly drill report documenting procedure deviations, time elapsed, validation outcomes, and corrective actions if any. The drill catches the small issues that would compound during a real incident: certificate expiry on the test environment, PostgreSQL extension version drift, network access changes, encryption key access issues. Each catch in the test environment is one less issue during a real recovery.

Documented escalation pattern with response times scaled to severity. Severity 1 (verification failure on hot tier backup, complete backup process failure, encryption key access failure): alert within 1 hour, investigation begins immediately, customer notification within 4 hours, root cause analysis within 24 hours, remediation within 48 hours. Severity 2 (warm tier verification failure, partial backup completion, retention enforcement anomaly): alert within 4 hours, investigation within 8 hours, customer notification within 24 hours, remediation within 5 business days. Severity 3 (cosmetic anomalies, dashboard issues, non-blocking warnings): logged and addressed in the next quarterly review cycle. The escalation pattern matters because most backup failures are not catastrophic but they degrade over time if ignored. A storage allocation slowly filling up takes months to become a problem but the symptoms appear weeks earlier. The verification process surfaces symptoms; the escalation pattern ensures they get addressed before they cascade.

Restore assistance is included in the subscription with response time scaled to incident severity. Severity 1 (production system down, customer-impacting data loss, suspected breach requiring DR action): on-call response within 30 minutes during operating hours, within 2 hours outside operating hours. Severity 2 (operational data issue requiring restore, scheduled DR migration, planned production cutover): response within 4 hours, scheduled within agreed window. Severity 3 (test environment restore, training drill, evidence collection for compliance audit): response within 1 business day, scheduled per customer preference. The restore assistance includes: incident scoping conversation to confirm restore target and scope, restore procedure execution against your specific backup configuration, post-restore validation against documented success criteria, incident timeline documentation suitable for inclusion in your incident response records, debrief covering what worked and what needs improvement in the runbook. Restore assistance is service-bounded to two incidents per calendar year at the standard subscription tier; additional restores or complex multi-component restores can be quoted as one-time engagements.

Annual rotation by default; quarterly rotation available on request for operations with stricter compliance requirements. The mechanism: AES-256-GCM symmetric keys are rotated by generating new keys, re-encrypting recent backups with the new key, retaining old keys for the duration of backup retention so older backups remain decryptable. The rotation is coordinated to avoid simultaneous re-encryption load by phasing across the three retention tiers (hot tier first, warm tier across two weeks, cold tier across one month). Old keys are decommissioned only after all backups encrypted under them have aged out of retention; this means the most recent expired key remains accessible for 365+30 days after rotation to support the longest retention tier. Key storage uses separate infrastructure from backup blob storage with its own access controls and audit logging. Key access events (read, rotate, decommission) are logged and reviewed during the quarterly health report. Quarterly rotation costs EUR 99/month instead of EUR 79/month to cover the additional operational load.

DR Setup is the foundational one-time engagement; BaaS is the ongoing operations layer. The relationship: DR Setup deploys the backup infrastructure with three-tier retention, encryption, dual-region storage, and the initial restore drill that proves the setup works. BaaS subscribes to operating that infrastructure over time. Customers can buy DR Setup alone and operate the backups themselves using the delivered runbooks; the runbooks document every procedure needed. Customers who recognize backup operations as continuous work that benefits from external ownership add BaaS. The typical pattern: DR Setup completes in 10 business days at EUR 1,299 one-time, BaaS begins immediately afterward at EUR 79/month. Many operations underestimate the operational load of running backups in-house, attempt it for 3-6 months, observe degradation, and subscribe to BaaS retrospectively. The retrospective subscription works equally well; the runbook is the same whether the customer operated it themselves first or not.

Yes, several upgrade paths are available. Standard tier EUR 79/month covers monthly drills, annual key rotation, two restore assists per year. Enhanced tier EUR 149/month adds weekly drills, quarterly key rotation, four restore assists per year, plus 30-minute on-call response 24x7. Compliance tier EUR 249/month adds bi-weekly drills, monthly key rotation, unlimited restore assists, plus quarterly compliance evidence packaging suitable for SOC 2 Type II observation window evidence collection. Add-ons available across tiers: additional storage region (EUR 39/month per region beyond default two), extended retention to 7 years for compliance hold (EUR 79/month), restore-to-cloud target (EUR 49/month per target environment), priority on-call routing (EUR 39/month). The tier and add-on combinations let operations match the subscription cost to their actual risk tolerance and compliance load rather than overbuying for unused capacity.