How does EDR actually work?
EDR records what happens on each device: which processes start, what they change and which network connections they make. When that behaviour matches an attack pattern, EDR can stop the process and cut the device off from the network. That is how it catches attacks that use legitimate admin tools or stolen passwords, where nothing on disk looks like a known bad file.
EDR is only as good as the people watching it. An alert that nobody reads protects no-one. Because penetration testing is also part of our work, the team reviewing your alerts knows how attackers behave once they reach a device.
A common example of what this looks like in practice: if a credential-theft tool is launched from a downloaded file on a laptop, the tool is blocked and the device isolated automatically. Our team then confirms nothing else is affected before releasing the device.
Which level of service suits you?
The FAQ below defines EDR, MDR and XDR. The practical difference is who acts on an alert and how widely you look:
| Option | Who acts on an alert | Suits |
|---|
| EDR | Your own IT staff | Organisations with IT staff who can review alerts |
| MDR | A security team, on your behalf | Most organisations without their own security team |
| XDR | A security team, correlating several sources at once | Organisations with several security tools and more complex environments |
We recommend one based on how sensitive your data is, who would act on an alert, and how much risk you’re prepared to carry.
How does endpoint security fit with ASD Essentials?
In ASD Essentials, EDR feeds the visibility and response domain. It is not one of the strategies in ASD’s current Essential Eight maturity model, which ASD Essentials will replace, but it supports several of them and helps you see when other controls fail. During rollout, we also help you apply hardening that maps directly to ASD’s strategies: application control, restricting Microsoft Office macros and user application hardening.
Endpoint security is strongest alongside other controls. Vulnerability management finds the missing patches attackers look for, user and identity management limits what a compromised account can reach, and managed security puts all of them under one team.