An operations handover should let the next responsible person run the product and recover a representative failure. A folder of architecture screenshots is incomplete if nobody knows who receives alerts or how to find a stuck paid workflow.
Organise the handover around tasks rather than the order in which the founder built the system. Keep the information short enough to use during an incident.
Document ordinary operating work first
List customer support, failed-job review, billing reconciliation, access changes, dependency updates, backups, and release checks that actually exist. Name the owner, trigger or frequency, required permissions, and evidence of completion.
Include an account inventory for domain, deployment, data stores, authentication, billing, email, and monitoring. Document recovery contacts and access procedures without embedding actual secret values in the guide.
Rehearse a fictional missing-entitlement case
Suppose a customer purchased study access but the application remains locked. The new operator should locate the account and purchase mapping, inspect event processing, determine the current entitlement, and follow a scoped reconciliation path.
They should also know what not to infer: a checkout screenshot is not the same as trusted provider confirmation, and a broad paid toggle can affect unrelated resources. Record any repair with actor, reason, and before-and-after state.
Next rehearse a failed background job and a safe deployment. The operator should identify the current version, find the failed step, understand retry consequences, and verify the final customer outcome. Improve the documentation wherever they need undocumented founder knowledge.
Keep emergency procedures specific and recoverable
Provide instructions for disabling an affected feature, communicating an incident, restoring data through the tested method, and returning to a known configuration where appropriate. State dependencies and limits, including actions that require another specialist or provider.
Separate technical control from contractual ownership and customer policy. The operator needs to know who can approve refunds, data deletion, exceptional access, or a change to retention. Those decisions should not be hidden in an engineer's script.
The tradeoff is completeness versus usability. A long manual can be difficult to navigate under pressure. Use short task pages with links to deeper references, stable identifiers, and a clear owner for updates. Record the last rehearsal date and material architecture changes that require another check.
The StackFlight product account demonstrates the range of a founder-built SaaS workflow. This handover exercise is proposed guidance and does not claim a particular internal playbook or support process there.
Questions about SaaS operations handover
Is source-code access enough?
No. Operating the product also requires external accounts, configuration, diagnostics, customer policies, and recoverable access.
Should every procedure be automated?
Not necessarily. A clear manual step can suit a pilot if its owner, limits, and evidence are explicit. Hidden manual work is the larger problem.
What proves the handover is usable?
Another authorised person completes ordinary tasks and a controlled recovery without private founder knowledge. Bring remaining dependencies to an operations handover review.