All Glossaries
1
min read
Table of Contents

Adversary Tracking and Attribution

What is adversary tracking and attribution?

Adversary tracking is the ongoing work of following a threat group over time: its infrastructure, tools, targets and habits. Attribution is the act of linking a specific attack or campaign to a specific group, and sometimes to the country or organization behind it.

The two depend on each other. Attribution is only possible because someone has tracked the group long enough to recognize its fingerprints. And tracking is what makes attribution useful, because knowing who is behind an attack tells you what they usually do next.

How does attribution work?

Attribution works by comparing evidence from an intrusion against what is known about tracked groups, and weighing how strongly the evidence points to one of them.

Analysts often organize this evidence using the Diamond Model of Intrusion Analysis, which links four elements of every intrusion: the adversary, their capability (tools and malware), their infrastructure (servers, domains) and the victim. Overlaps across those elements, such as reused infrastructure, shared code or a consistent choice of targets, build a case.

Attribution comes in levels of confidence and depth:

  • Technical attribution links activity to a tracked cluster, for example "this matches a cluster we track under a temporary designation."
  • Strategic attribution links that cluster to a sponsor or motive, such as a specific nation-state or criminal enterprise. This requires much more evidence and is often left to governments.

Why do threat groups have so many names?

Threat groups have many names because each security vendor tracks groups through its own visibility and uses its own naming system.

Mandiant uses numbered designations like APT29 and FIN7, plus UNC numbers for clusters not yet fully attributed. CrowdStrike uses animal names tied to the suspected country, such as Bear for Russia and Panda for China. Microsoft switched in 2023 to weather-themed names, such as Blizzard for Russia and Typhoon for China. As a result, one group known as APT29 is also called Cozy Bear and Midnight Blizzard.

These names do not always line up exactly, because each vendor's view of a group's boundaries differs slightly. Good threat intelligence platforms keep alias mappings so analysts can see that three reports describe the same actor.

Why does adversary tracking matter to the SOC?

Adversary tracking matters to the SOC because knowing the likely actor turns an isolated alert into a predictable story.

If an alert matches infrastructure tied to a group known for stealing credentials, then moving laterally with remote management tools, the SOC knows what to look for next. Mapping a group's known behaviors to MITRE ATT&CK techniques lets detection engineers build coverage for how that group actually operates, not just for the indicators it used last time. This is the top of the Pyramid of Pain, where defenses cost the attacker the most.

What are the pitfalls of attribution?

The main pitfalls of attribution are overconfidence, false flags and shared tooling.

Attackers sometimes plant misleading clues on purpose. The 2018 Olympic Destroyer attack on the PyeongChang Winter Olympics is a well-known example: its code contained deliberate false flags pointing to several different countries. Many groups also use the same commodity tools and rented infrastructure, so a shared tool is weak evidence on its own. Treat attribution as a confidence-rated assessment, and act on behaviors, not just on names.

Frequently asked questions

Do most organizations need attribution?

Most organizations do not need strategic attribution to a country. They benefit from technical attribution, because knowing which tracked group is likely involved helps predict its next steps and prioritize defenses.

What is the difference between a threat actor and an APT?

A threat actor is anyone conducting malicious activity. An advanced persistent threat (APT) is a well-resourced actor, often state-sponsored, that maintains long-term access to targets.

How confident can attribution be?

Confidence ranges from low to high depending on evidence. Analysts often use estimative language such as 'likely' or 'almost certainly' to communicate that confidence honestly.

Anomali and Adversary Tracking and Attribution

Anomali ThreatStream Next-Gen maintains actor profiles, aliases and ATT&CK mappings, and links them to indicators so that a match in telemetry comes with actor context attached.

Request a demo to see actor tracking in Anomali.

Read next

‍