TL;DR — Audit prep should not be a fire drill. The teams that survive audits with their sanity intact treat evidence collection as a continuous discipline, not a quarterly sprint. The artifacts and runbook below replace the “drop everything and hunt for screenshots” pattern.
Audit season catches teams unprepared every year. The audit is not the problem; the lack of an evidence pipeline is. With the right setup, an audit is a review of artifacts that already exist. Without it, it is six weeks of digging through Slack for screenshots.
The artifacts you need ready
- Access reviews — quarterly snapshots of who has access to what, with manager approval recorded
- Change logs — every production change tied to a ticket, a reviewer and a deployment record
- Vulnerability scans — monthly runs of dependency, container and infrastructure scans
- Incident records — postmortems for every meaningful incident
- Vendor reviews — annual review of every subprocessor, with current SOC 2 reports collected
- Training records — security training completed by every employee, dated
- Backup tests — actual PITR tests run quarterly, evidenced
- Encryption records — TLS configs, KMS key rotation logs
The evidence pipeline
- Wire your tooling to produce evidence as a byproduct of normal operation
- GitHub for change logs and reviews. PR templates enforce review evidence.
- Cloud-native logging for access reviews — IAM audit logs streamed to a long-term store
- Dependabot + Trivy for vulnerability scans — automated, archived
- Sentry / OTel for incident detection, your postmortem template for the document
- HR system for training records — integrated with your security training vendor
The audit runbook
- Six months out — gap analysis against the standard, fix gaps
- Three months out — evidence walkthrough with your auditor
- One month out — final controls review, document any compensating controls
- During audit — one person owns auditor communications; the rest of the team is shielded
- Post-audit — review findings, update the runbook for next year
The mistakes to avoid
- Treating audit prep as a discrete project — it has no end, only a cadence
- Documenting things you do not actually do — easier to do than document, easier to audit too
- Relying on one person — when they leave, evidence collection collapses
Frequently asked questions
When should we start preparing for an audit?
For first-time audits, six months out. For renewals, evidence collection should run continuously — audit prep should be a verification step, not a sprint.
Can we automate evidence collection?
Most of it, yes. Access reviews, change logs, deployment records, vulnerability scans — all can come from existing tooling with the right integrations. Manual collection is the failure mode you want to avoid.
Working on something similar?
T-Square architects, builds and operates production systems for learning, AI and custom software products. Talk to a senior engineer for a second opinion.