Before you trust FlowOne with your business, challenge us.

The standby node takes over automatically — no phone call needed. In our regular failover drills the takeover completes in 4–6 minutes with replication lag held under 30 seconds; those figures are observed in drills, not a contractual SLA. The failover replay on this page shows the exact sequence.

That is what disaster recovery exists for, separately from high availability: independent encrypted backups, a separate recovery environment and documented restore procedures. Recovery follows the DR values defined in your agreement — not improvisation.

The honest answer: they are defined per deployment and service level, and we put them in writing before you commit. The numbers we publish on this page are drill observations; the contract is where they become commitments with remedies attached.

Measurable terms: availability, incident response, escalation path, RPO, RTO, maintenance windows, notification obligations — and remedies if we miss. We deliberately separate technical targets from what we sign.

Restore is drilled on a regular schedule, not once before the sales call. We share the drill logs and results under NDA — ask for the most recent one and we will show you.

A stopped integration raises an alert — silence itself is a failure condition. Retries and queues absorb transient errors, scheduled reconciliation verifies the systems still agree, and every sync is logged, so drift is caught before a customer catches it.

Reconciliation, not confidence: counts, records, attachments, permissions and critical fields are compared between the old and new systems, then your team validates real work in user acceptance testing before cutover.

The rollback plan is written and rehearsed before cutover, the old systems stay available during an agreed rollback window, and ownership of the go/no-go decision is agreed beforehand. A failed cutover is a planned scenario, not a crisis.

Yes, within the agreed rollback window. Changes made in the new system after cutover are tracked, and reconciliation covers carrying them back — that case is part of the rollback plan, not an afterthought.

You do — FlowOne gains no ownership of customer business data, full stop. Retention, export rights and deletion procedures are written into the agreement, so ownership is enforceable rather than a slogan.

Yes: structured records in open formats, files and documents, metadata, users, permissions, integration configurations and documentation. The export manifest on this page is the shape of what a real handover produces.

It depends on the deployment. On infrastructure you control, the running system and its data stay in your hands regardless of what happens to us. On FlowOne-managed infrastructure the contract does the work: export procedures, backup access, transition assistance, an agreed termination period — and source-code escrow where commercially appropriate.

That is the design goal of the exit plan: documented export formats, migration documentation and a defined handover mean a competent IT team can execute the move without our goodwill. Leaving may require work. It should never require permission.

The source code is FlowOne intellectual property unless another arrangement — such as escrow — is agreed. Your data, exports, configurations and documentation are not. We state this explicitly because vague answers here are a red flag.

Exit planning defines this up front: what is exportable, the timing, migration support during transition, and the cost of additional transition work — agreed at onboarding, not negotiated when the relationship is already strained.