Skip to content
Tutorials Center
My learning

Topics

Lesson 32 of 37 ยท Metrics and Frameworks

MITRE ATT&CK and Detection Coverage

Learn how a SOC uses MITRE ATT&CK, the Cyber Kill Chain and NIST CSF to describe attacker behavior, map detections and find coverage gaps, with a coverage calculator.

Security teams need a shared language. When one analyst says "the attacker persisted with a scheduled task" and another says "they set up a cron job", are those the same behavior? When a report describes a campaign, how do you check whether your detections would catch it? Frameworks answer these questions. This lesson introduces the most important ones for a SOC, with the focus on MITRE ATT&CK, and shows how to turn it into a measurable detection coverage map.

MITRE ATT&CK

ATT&CK is a public knowledge base of real-world attacker behavior, maintained by the MITRE Corporation. It organizes behavior into:

  • Tactics: the attacker's goal at a stage ("gain initial access", "avoid detection").
  • Techniques: how they achieve that goal ("phishing"). Techniques have IDs such as T1566.
  • Sub-techniques: more specific variants, written like T1566.001.
  • Procedures: how a particular group or tool really does it.

The Enterprise matrix has fourteen tactics, roughly in the order an intrusion unfolds:

flowchart LR
    R["Reconnaissance"] --> RD["Resource Development"] --> IA["Initial Access"] --> EX["Execution"] --> PE["Persistence"] --> PR["Privilege Escalation"] --> DE["Defense Evasion"]
    DE --> CA["Credential Access"] --> DI["Discovery"] --> LM["Lateral Movement"] --> CO["Collection"] --> C2["Command and Control"] --> EXF["Exfiltration"] --> IM["Impact"]

Attackers do not follow the order strictly: they loop, skip and repeat tactics.

ATT&CK vocabulary

Tactic The goal, such as Persistence
Technique The method, such as Scheduled Task
Sub-technique A specific variant
Group Who uses it, with procedures

IDs like T1053 let everyone mean the same thing.

1 / 2

A few well-known techniques (always confirm IDs and names on the official site, since the knowledge base is updated regularly):

ID Technique Tactic Typical data source
T1566 Phishing Initial Access Mail logs, gateway
T1078 Valid Accounts Initial Access, Persistence and others Authentication logs
T1110 Brute Force Credential Access Authentication logs
T1059 Command and Scripting Interpreter Execution Process creation logs
T1053 Scheduled Task/Job Execution, Persistence Process, task and cron logs
T1003 OS Credential Dumping Credential Access Endpoint telemetry
T1021 Remote Services Lateral Movement Authentication, network
T1071 Application Layer Protocol Command and Control Proxy, DNS, network
T1070 Indicator Removal Defense Evasion Audit and log-clear events
T1486 Data Encrypted for Impact Impact Endpoint, file activity

Mapping detections

For each detection rule, tag the ATT&CK technique(s) it covers, the data source it needs, and a status: working, untested, noisy or planned. Then ask: which techniques that matter to us have no working detection? That answer drives the detection engineering backlog.

Important cautions:

  • Having a rule is not the same as having coverage. A rule that never fired may be broken, or the attack may never have happened. Test it by replaying logs or running an approved simulation.
  • Techniques have many variants. One rule for T1059 covers some scripts, not all.
  • Not all techniques matter equally. Prioritize by the groups that target your industry and by how your environment is built.
  • A heat map of 100% green is a lie. Nobody covers everything.

Try it: calculate coverage

This stand-in holds a toy rule list and a list of priority techniques, and prints which are covered, which are only partially covered and which are not. The data is invented.

Python Detection coverage by tactic

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

The result is a to-do list: write detections for credential dumping, remote services, command and control and ransomware behavior, and fix the noisy and untested ones. See alert tuning.

Try it: map an incident to techniques

Using the framework in reports makes incidents comparable. Here an incident narrative is tagged, then summarized by tactic to show the attack's shape.

Python Tag an incident timeline with techniques

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Other frameworks you will meet

  • Cyber Kill Chain (Lockheed Martin): seven stages from reconnaissance to actions on objectives. Older and more linear than ATT&CK, useful for explaining an attack to non-specialists.
  • Diamond Model: describes an intrusion event through four features: adversary, capability, infrastructure and victim. Good for linking events into campaigns.
  • NIST Cybersecurity Framework: organizes an organization's security program into functions (Govern, Identify, Protect, Detect, Respond, Recover in version 2.0). The SOC mostly lives in Detect and Respond.
  • NIST SP 800-61 and the SANS model: incident handling processes (incident response lifecycle).
  • MITRE D3FEND: a companion knowledge base of defensive techniques, mapped to attacker behaviors.
  • CIS Critical Security Controls and ISO/IEC 27001: prioritized controls and management standards used by auditors.
  • Sigma: a vendor-neutral rule format that carries ATT&CK tags with each rule.

Use frameworks as maps, not scoreboards. The goal is to find and close real gaps against real threats, not to color a matrix green.

Common mistakes

  • Counting every rule tagged with a technique as full coverage.
  • Ignoring data sources: a technique cannot be detected without the logs that show it (see the log collection chapter).
  • Copying someone else's priority list instead of choosing techniques relevant to your organization.
  • Treating ATT&CK as a compliance checklist.

Check your understanding

Quick check ยท 4 questions

of 4 answered You scored / 4 ยท perfect!

  1. 1. In ATT&CK, what is the difference between a tactic and a technique?

    Tactics are goals such as Persistence; techniques such as Scheduled Task are methods.
  2. 2. A detection rule has never fired. What can you conclude about coverage?

    Silence can mean no attack or a broken rule. Replay logs or simulate safely to verify.
  3. 3. Why can a technique be impossible to detect even with a good rule?

    Detection depends on the necessary telemetry; missing data means a visibility gap.
  4. 4. Which framework is mainly about a security program's functions such as Detect and Respond?

    NIST CSF groups security activities into functions; the SOC covers Detect and Respond.

Summary

  • ATT&CK gives a shared vocabulary of tactics, techniques and sub-techniques based on observed attacker behavior.
  • Tag detections with techniques, data sources and test status to measure real coverage and prioritize gaps.
  • Map incidents to techniques so they can be compared and reported consistently.
  • Other frameworks (Kill Chain, Diamond Model, NIST CSF, D3FEND, CIS, Sigma) serve different purposes; treat all as maps, not scoreboards.

Next: Alert Tuning and Detection Quality