What to log
Record the actor, target account, action taken, method involved, approver if any, and time. For recovery events, also note the reason and any ticket or case number tied to the reset.
How to review
Review the log on a schedule, not only after an incident. Look for repeated resets, unusual geography, policy changes without approvals, and accounts that keep falling back to weaker methods.
How long to keep it
Retention should match your security and compliance needs. Keep enough history to investigate a recovery abuse case and enough structure to export the data to your SIEM or case management system.
Implementation checklist
- Define the minimum event set before rollout.
- Send logs to a system that support and security can both review.
- Link each emergency recovery to a ticket or case record.
Policy cautions
- A log that nobody reads is just storage cost.
- Do not log secrets, backup codes, or QR contents in the audit trail.