NEWSLETTER · ISSUE 30
Going live with clear ownership and a rollback plan
Going live is a decision about who will act, what they will watch, and how they will respond if the change does not behave as intended. A clear launch note connects the tested implementation with the people responsible for operating it.
The aim is a usable decision record. Someone should be able to find the authorized release, the accepted owners, the remaining conditions, and the route back to a secure operating position without reconstructing the whole project.
Confirm authority and ownership
Our fictional subscription startup has corrected the repeated-notification mismatch from its earlier test and retested the affected behavior through the proper process. That result now joins the other readiness evidence. The people with appropriate authority still need to confirm the release conditions and decide whether the proposed launch may proceed.
The developer, business operator, and provider contact walk through their parts together. They confirm who can authorize the change, execute it, observe the results, coordinate a problem, and make a recovery decision. A named person should have accepted the responsibility and be available when needed.
Leave a launch note people can use
Use five prompts to connect the release decision with the work around it:
Release authority: Identify the proposed change and the person or role authorized to decide whether it proceeds, pauses, or requires further review. Record the conditions for that decision and the approved route for changing it. A successful functional test, a calendar invitation, or a copied recipient does not supply release permission or establish PCI DSS compliance.
Accepted owners: Confirm who will execute, monitor, communicate, and coordinate provider support. Identify the accepted backup or escalation route if someone becomes unavailable. Explain the limits of each role's authority. One coordinator can keep the conversation moving while the business and provider retain their respective responsibilities. PCI SSC's guidance makes customer oversight of provider relationships an ongoing responsibility.
Readiness checks: Bring together the relevant test results, corrections, security work, configuration review, access approvals, operating instructions, and unresolved issues for the actual arrangement. Keep each condition's position and owner visible. Confirm what the authorized decision-maker needs to establish before release. A collection of green task labels should not conceal an unobserved result or an unmet requirement.
Rollback and recovery: Describe the conditions that call for pausing or recovering, who can decide and execute, and the approved steps for reaching a secure state. Choose triggers relevant to the service and its risks, with appropriate review. Confirm that any rollback is compatible with the current integration and records. A previous version is not automatically a safe destination. Where rollback is unsuitable, use the properly authorized recovery or containment route.
Observation and handoff: Agree what will be observed, who receives concerns, how provider and business contacts connect, and what establishes a stable operating position. Record what the next owner needs to know and confirm acceptance of the handoff. Keep monitoring, response, and required controls operating through the transition; the launch note does not suspend them.
Prepare a secure recovery route
Reverting code or configuration does not undo charges or refunds, repair order states, or settle transactions already in progress. Preserve the relevant payment and order records. Coordinate reconciliation of pending, repeated, or delayed events through the provider's documented process. Avoid a blind restore that loses current records or an improvised retry that creates duplicate financial activity.
In the fictional rehearsal, the team considers an order update that disagrees with the documented payment state. They explain who can pause the affected automated step through the approved route, who examines the records, and who contacts the provider. They practice the recovery decision using invented records and synthetic messages in an authorized test setting. No live payment activity is needed.
PCI DSS v4.0.1 addresses documented production change procedures, authorized approval, security-impact testing, and handling failures while returning to a secure state. It also addresses confirming applicable requirements on new or changed systems and networks after a significant change, with documentation updated as applicable. This launch note supports that work; it does not replace the applicable requirements or testing.
Observe and hand over
Walk the note with a partner. Leave one readiness condition open and one primary contact unavailable. Can they identify what must wait, the accepted backup, the recovery decision-maker, and the records that must remain intact?
Required incident-response processes continue if there is a suspected security incident. A release or recovery decision does not determine the business's PCI DSS validation and reporting method; confirm that separately with the appropriate compliance-accepting entity. A calmer launch leaves people able to act within their authority and recognize when the next decision belongs elsewhere.
Explore the free Provider Responsibilities guide (opens in a new tab)