Dwell Time
How long an attacker goes undetected in a network.
What it is
Dwell time is the elapsed period between an adversary's initial compromise and their detection or eviction. It is a headline incident-response metric.
Why it matters
Longer dwell time means more time to move laterally, steal data, and entrench, so shrinking it through faster detection is a core program goal.
In practice
An intrusion discovered 90 days after initial access: a dwell time defenders aim to cut to days.
What to look for
- The earliest evidence of attacker activity, not the alert time.
- Gaps in log retention that hide when it started.
- Persistence created long before the detected activity.
Common mistake
Starting the clock at detection. Dwell time runs from the first foothold, which is usually weeks earlier.
Practice it nearby
CyberKit has no tool for this term itself. These are the closest places to practice it.
- Hunt Query Builder
Build a hunting query in KQL, SPL, LogScale or PowerQuery against documented field names, or start from a prebuilt hunt.
- Indicator Extractor
Pull every IOC out of a log or a report, then score the domains against published string heuristics.
- Report Generator
Write an incident report with a timeline, evidence and chain of custody, findings, indicators, ATT&CK mapping and queries. Export it as PDF, HTML or Markdown.
Read next
- Threat Hunting: Proactively searching for threats that evaded detection.
- Indicator of Compromise: Forensic breadcrumbs that a system was compromised.
- Lateral Movement: Spreading from the first foothold to other systems.