Sometimes the team is asked to investigate a suspicion of a suspicion -- to determine whether it’s worth anyone’s time to dig further. Sometimes the "breach" is actually a turf war between departments, branch offices, or system administrators, when someone uses a root password that they weren’t supposed to have. And sometimes the monitoring is just for distasteful behavior that, strictly speaking, isn’t security-related.
One of the most difficult tasks to do well with any kind of monitoring system is to support an HR investigation. Let’s take the classic example of an employee being suspected of visiting inappropriate websites using company systems. Most Web filtering reporting systems will list every single URL that the user’s browser accesses, and that’s a lot of them. Just think of how many ads, graphics, and frames are displayed on any given page, and throw in any automatic authentication processes behind the scenes.
The first task is to sort out which main site the user was actually visiting; the second is to determine which sites the user accessed on purpose. Determining intent in browsing behavior is not something that comes easily to a SIEM, but it’s crucial for any potential disciplinary action. Did the user just visit the Exotic Singles Club site, or did he log in there, indicating that this was more than a one-time thing?
Other "breach" incidents that monitoring can pick up include unscheduled code changes, strategic alteration of a legal document, employee naps at the keyboard, leaks to a reporter, accidental publication, magically appearing and disappearing servers (and I’m not even talking about virtual machines), and one person calling another a doodyhead on instant messaging.
Security monitoring can be supportive of the business as well. It can be used to prove to an auditor that actions were taken by a required deadline. I saw it used once in a grievance case to show that the company assumed an important email was successfully delivered by a certain time because it tracked all email bounces and didn’t get one for that recipient’s address. The right kind of security monitoring can help troubleshoot operational issues and collect information at a certain point in time that is no longer available anywhere else.
So when you’re designing your security monitoring system, make sure it supports more use cases than just "catching APTs." Gather requirements from your business areas, HR, legal, audit, and your colleagues in IT, because it can potentially be useful to all of them. Of course, you also need to be sure that your monitoring doesn’t clash with any data or privacy laws, but in all other ways, think bigger than "breach."
Wendy Nather is Research Director of the Enterprise Security Practice at the independent analyst firm 451 Research. You can find her on Twitter as @451wendy.