An audit trail is only useful if it can answer a short set of questions: who did what, when, from where, to which record, with what result, and whether anyone reviewed it. Logging everything without a review process produces storage bills and false confidence. Logging too little produces an investigation that ends in “we cannot tell”.
This checklist is designed to be used as a working document. Each control has an owner, the evidence it should produce, the review frequency and the sign-off. Adapt the names and systems, but keep the structure.
How to put this checklist to work
Start with the systems that hold money, personal data or administrative power. Do not attempt to log every event in every application on the first pass. For each system, record the data class, the legal or contractual retention requirement, the log destination and the person accountable. The completed register becomes the scope for the rest of the checklist.
Scope the log sources
- System inventory: list applications, databases, cloud accounts, SaaS admin consoles, network devices and endpoints that hold in-scope data. Owner: security lead. Evidence: log source register with data class and retention. Frequency: annual and on major change. Sign-off: operations head.
- Identity population: include employees, contractors, service accounts, integrations and vendor accounts. Owner: identity administrator. Evidence: identity inventory with account type and owner. Frequency: quarterly. Sign-off: security lead.
- Event catalogue: define the events that must be logged for each system, with the fields required. Owner: application or platform owner. Evidence: event catalogue and sample event. Frequency: per release and annually. Sign-off: security lead.
- Legal and contractual retention: confirm minimum and maximum retention for each data class, including any sector rules. Owner: legal or compliance. Evidence: retention schedule with citations. Frequency: annual. Sign-off: business owner.
Identity and access events
Identity is where most investigations start. Capture both successful and failed events, and include the actor, the target, the source address, the user agent, the authentication method and a request identifier.
- Sign-in success and failure, including federated and API authentication.
- Multi-factor challenge, enrolment, reset and bypass.
- Password reset and account recovery.
- Account creation, disablement, deletion and reactivation.
- Role, group and permission changes, including temporary elevation.
- API key, token and service-account creation or rotation.
- Break-glass and emergency access use.
Owner: identity administrator. Evidence: sample events for each type and a parser test. Frequency: continuous collection, monthly test. Sign-off: security lead.
Application and data events
Application logs need to reflect business actions, not only errors. Record the record identifier and the action, and avoid writing secrets or full sensitive payloads into the log.
- View or access of sensitive records, especially by staff outside the owning team.
- Create, update, delete, approve, reject, cancel and reverse actions.
- Export, download, print, bulk update and bulk delete.
- Permission and configuration changes inside the application.
- Failed authorisation attempts and access to denied resources.
- Integration calls that change state, with correlation IDs across services.
Owner: application owner. Evidence: event catalogue, sample trace across two services and a redaction test. Frequency: per release. Sign-off: security lead.
Cloud and infrastructure events
Cloud control-plane logs are frequently missed because the application is logging while the platform is not. Enable the provider's audit service and protect it separately from the workload.
- Console and command-line sign-in, including failed attempts.
- Identity and policy changes, key creation and deletion.
- Network rule, firewall and storage policy changes.
- Database administrative actions, snapshots and restores.
- Backup, replication and deletion events.
- Malware, endpoint and intrusion alerts with their disposition.
Owner: platform lead. Evidence: configuration export, a test alert and a retention check. Frequency: continuous, quarterly review. Sign-off: security lead.
Protect integrity and set retention
- Append-only storage: use write-once or immutable storage for the audit copy, with a separate account or subscription. Owner: platform lead. Evidence: configuration and deletion-attempt test. Frequency: annual. Sign-off: security lead.
- Time synchronisation: all systems use a consistent time source with timezone recorded. Owner: platform lead. Evidence: time-source configuration and sample timestamps. Frequency: quarterly. Sign-off: operations head.
- Access to logs: restrict read access, log access to the log store and review the reader list. Owner: security lead. Evidence: access list and review record. Frequency: quarterly. Sign-off: operations head.
- Retention and deletion: apply the schedule, document legal holds and prove deletion at the end of the period. Owner: compliance. Evidence: retention report and deletion certificate. Frequency: annual. Sign-off: business owner.
- Backup and restore: the audit store must survive the loss of the primary environment. Owner: platform lead. Evidence: restore test with timestamps. Frequency: annual. Sign-off: security lead.
Review, investigate and sign off
- Daily: automated alerts for high-risk events such as privilege escalation, bulk export or repeated failed sign-ins. Owner: security operations. Evidence: alert log with disposition. Sign-off: weekly.
- Weekly: review privileged actions and unresolved high-risk alerts. Owner: security lead. Evidence: review notes and assigned actions. Sign-off: operations head.
- Monthly: manager access review for in-scope systems, with removal of unnecessary access. Owner: line managers. Evidence: signed access review. Sign-off: security lead.
- Quarterly: tabletop investigation using a simulated event; confirm that logs, timestamps and ownership are sufficient. Owner: incident lead. Evidence: exercise report and corrective actions. Sign-off: business owner.
Privacy and employee monitoring
Logging staff activity is itself a privacy decision. Apply purpose limitation, collect the minimum needed, tell people what is recorded and restrict who can read it. Do not log the content of communications unless there is a documented, proportionate reason. Where personal data is involved, align retention and access with the Bangladesh framework that applies to your organisation and take legal advice on sector-specific rules.
Owner: privacy lead. Evidence: employee notice, purpose statement and a data protection assessment for high-risk monitoring. Frequency: annual and before a new monitoring tool. Sign-off: business owner.
The quarterly evidence pack
Collect the log source register, event catalogue, retention schedule, sample investigation, access review, restore test and open gaps with owners and dates. The pack should let an auditor or insurer see that logging is operated, not merely purchased. Sign it at leadership level and keep the previous four packs.
If you need help designing the logging scope or testing whether the trail would survive an incident, start with Managed Technology & Cybersecurity. Our pricing page explains how assessment and retained security work are scoped. For the supplier side of access risk, read how third-party security risk changes as your vendor list grows.
Sources
- NIST, SP 800-92: Guide to Computer Security Log Management, accessed 17 November 2026.
- OWASP, Logging Cheat Sheet, accessed 17 November 2026.
- NIST, SP 800-53 Rev. 5: Security and Privacy Controls, accessed 17 November 2026.
- IETF, RFC 5424: The Syslog Protocol, accessed 17 November 2026.
Responses (0)
No comments yet. Add the first one below.
Leave a response
Your email address will not be published. Required fields are marked *.