Home/Refreshers/Working an insider case: refresher
Working an insider case
Triage, authority, evidence, and hand-over for a case about someone who is allowed in.
Remember
- Most signals about a person have an ordinary cause. Triage quietly before anything else.
- The acceptable use policy and the monitoring notice decide what you may review. Ask Legal when unsure.
- Keep the circle small, name the case neutrally, and keep it out of shared queues.
- Leave access alone until the case decides, unless harm is under way.
- Write what the logs show. Intent is for HR and Legal to weigh.
| Signal | Where it shows |
|---|---|
| Notice given, or a role ending | HR records only |
| Far more reads than usual | Object-level read logs, off by default |
| Data shared outside | The cloud's admin audit log |
| New ways to decrypt | Key policy and grant changes |
| Logging turned down | Trail, sink, and exclusion changes |
| Files copied to a removable drive | Windows event 4663, with Removable Storage auditing on |
The case
- 1. Triage Compare the person's activity with their own previous month. Check whether their access changed just before.
- 2. Open a case With HR and Legal: authority, scope, who knows, and a legal hold.
- 3. Preserve Extend log retention, lock the evidence bucket, and ask the mail and document platforms for their legal holds.
- 4. Collect Record who, when, where from, the hash, and where kept. Export queries with their text and time range.
- 5. Hand over A short memo: timeline, what data, how much, where it went, what access remains.
Watch out
- One signal is weak. Notice, then bulk reads, then sharing outside, is a pattern.
- Removing some access early can warn the person. Remove all of it together, timed with HR.
- The person's manager and close colleagues should not work the case.
Check it worked
Every item collected has a recorded custodian, time, source, and hash, and nothing the legal hold names can expire.