Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
An assumed breach exercise starts the test with the attacker already inside, skipping initial access so that the engagement measures internal detection, containment, and response instead.

What Is an Assumed Breach Exercise?
An assumed breach exercise is a security test that begins from the premise that an attacker has already obtained a foothold. Rather than spending the first week of an engagement trying to get in, the client provides the starting position: a standard domain user account, a managed laptop, or an agreed implant placed on a workstation. Everything after that point is tested properly.
The rationale is not that perimeters do not matter. It is that initial access has become cheap. Credentials harvested by infostealer malware are sold in bulk, phishing kits defeat one-time passcodes through real-time relaying, and a single unpatched edge device will do the job. Assuming the attacker gets in once is now the realistic planning assumption, and the interesting question is what happens in the hours that follow.
Perimeter-focused testing answers one question: can we be breached. Assumed breach answers several more useful ones.
The output is a timeline as much as a vulnerability list. Knowing that Kerberoasting went unnoticed for four days while a deliberately noisy share enumeration triggered an alert in nine minutes tells a security team more about its detection coverage than any number of configuration findings.
The scenario needs to be specific or the results are not comparable. Typical starting points, roughly in order of how much is granted:
The scenario also specifies what is not provided. If the tester is given local administrator rights on the first host, the exercise skips privilege escalation, which is usually the part worth testing.
| Exercise | Starting point | Primary question |
|---|---|---|
| Penetration test | External, unauthenticated | Which vulnerabilities exist and are exploitable? |
| Assumed breach | Foothold provided | How quickly do we detect and contain internal activity? |
| Red team | No assistance, objective-based | Can a determined adversary reach a defined crown jewel undetected? |
| Purple team | Collaborative, techniques run openly | Which detections fire, and how do we improve the ones that do not? |
Many red team engagements now include an assumed breach fallback clause: if the team has not achieved initial access within an agreed window, a foothold is granted so the remaining budget tests internal defences rather than being consumed by the front door. In the UK financial sector, the Bank of England’s CBEST and the industry STAR-FS scheme are threat-led and intelligence-driven, and their scenarios routinely include an assumed compromise leg for the same reason.
Two practices separate an exercise that improves defences from one that produces a report nobody acts on. First, keep a precise activity log with timestamps so the blue team can replay every action against their telemetry and find out what was recorded but not alerted on. Second, run a joint replay session afterwards, where each technique is repeated openly and detection engineering happens in the room. That converts a one-off finding into a durable detection rule mapped to MITRE ATT&CK.
At Privia Security, our Red Team Services run assumed breach scenarios with agreed starting positions and a full activity timeline for the defending team.
No. They answer different questions. A penetration test tells you which exposed vulnerabilities an attacker could exploit, which still matters for patching and for compliance obligations. An assumed breach exercise tells you what happens once someone is inside regardless of how they got there. Most organisations run both, with external testing on an annual or more frequent cycle and assumed breach work less often but at greater depth.
It depends on the objective. Keeping them uninformed produces a genuine measurement of detection and response, but risks a real incident response being triggered at cost. A common compromise is that two or three named people hold the details and can confirm that activity is authorised if an investigation escalates, while the wider team works the alerts normally.
The least that still represents a realistic compromise, which is usually a standard domain user account with no local administrator rights, on a device built to your normal standard with the full endpoint agent stack installed. Granting elevated rights at the outset removes privilege escalation from scope, and stripping the endpoint agents removes the controls you most want evaluated.
Typically two to four weeks of testing for a medium-sized estate, plus scoping and reporting. Shorter engagements can still be useful if the objective is narrow, for example confirming whether a specific attack path to a domain controller exists. Longer engagements allow slower, quieter tradecraft, which is what you need if the question is whether a patient adversary would be noticed.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.