ATT&CK Tactics

Overview

Tactics represent the adversary's tactical objective — the "why" behind an action. Each tactic answers a question: Why is the adversary doing this? To gain initial access. To escalate privileges. To move laterally. To exfiltrate data.

The MITRE ATT&CK Enterprise matrix organizes 14 tactics in a rough left-to-right attack progression, from pre-compromise reconnaissance through post-exploitation impact. Adversaries do not follow them sequentially — they loop, skip, and revisit tactics as needed. The ordering reflects a general operational flow, not a rigid kill chain.

Key Concepts

The 14 Enterprise Tactics

Each tactic has a unique ID (TA-prefix) and contains multiple techniques that achieve the tactical objective.

TA0043 — Reconnaissance

The adversary gathers information about the target before attacking. This includes identifying target infrastructure, employees, technologies, and vulnerabilities through passive and active means.

Examples: OSINT gathering, port scanning, social media research, DNS enumeration, job posting analysis.

TA0042 — Resource Development

The adversary acquires or builds resources to support operations. This happens before initial access and includes infrastructure setup, tool development, and capability procurement.

Examples: registering domains for phishing, purchasing VPS infrastructure, developing custom malware, compromising third-party accounts, obtaining code signing certificates.

TA0001 — Initial Access

The adversary gains a foothold in the target environment. This is the first point of entry — the transition from external to internal.

Examples: spearphishing attachments, exploiting public-facing applications, supply chain compromise, valid account usage, drive-by compromise.

TA0002 — Execution

The adversary runs malicious code on the target system. Execution is often combined with other tactics — the adversary needs to execute something to establish persistence, escalate privileges, or move laterally.

Examples: PowerShell, command-line interface, Windows Management Instrumentation (WMI), scheduled tasks, user execution of malicious files.

TA0003 — Persistence

The adversary maintains access across restarts, credential changes, and other disruptions. Persistence mechanisms survive system reboots and ensure the adversary can return.

Examples: registry run keys, scheduled tasks, creating local accounts, modifying startup scripts, implanting web shells, DLL hijacking.

TA0004 — Privilege Escalation

The adversary gains higher-level permissions. Escalation typically moves from standard user to administrator/root, or from local admin to domain admin.

Examples: exploiting SUID binaries, kernel exploits, token manipulation, abusing sudo misconfigurations, exploiting vulnerable services, leveraging group policy.

TA0005 — Defense Evasion

The adversary avoids detection throughout the operation. Evasion techniques operate alongside other tactics — the adversary evades detection while persisting, escalating, and moving laterally.

Examples: obfuscating scripts, disabling security tools, clearing logs, timestomping files, process injection, masquerading as legitimate processes, using living-off-the-land binaries (LOLBins).

TA0006 — Credential Access

The adversary steals credentials — passwords, hashes, tokens, tickets. Credentials enable lateral movement and privilege escalation without exploiting vulnerabilities.

Examples: dumping LSASS memory, Kerberoasting, brute-force attacks, keylogging, extracting credentials from configuration files, password spraying.

TA0007 — Discovery

The adversary explores the environment to understand what they have access to and what is reachable. Discovery maps the internal landscape — accounts, systems, network topology, security controls, and data locations.

Examples: network share enumeration, account discovery, system information gathering, group policy discovery, domain trust enumeration.

TA0008 — Lateral Movement

The adversary moves through the environment to reach additional systems. Lateral movement uses stolen credentials, exploitation, or legitimate remote access tools to pivot between hosts.

Examples: pass-the-hash, pass-the-ticket, remote desktop (RDP), SMB/Windows Admin Shares, SSH hijacking, WinRM.

TA0009 — Collection

The adversary gathers data of interest from target systems. Collection focuses on identifying and staging data before exfiltration — documents, emails, databases, credentials, and intellectual property.

Examples: data from local drives, screenshots, keylogging, email collection, data from network shared drives, clipboard data, data from information repositories.

TA0011 — Command and Control (C2)

The adversary communicates with compromised systems to control them remotely. C2 channels must blend into normal traffic to avoid detection.

Examples: HTTPS beacons, DNS tunneling, web service C2 (using legitimate platforms like Slack or GitHub), encrypted channels, domain fronting, protocol tunneling.

TA0010 — Exfiltration

The adversary steals data from the target environment. Exfiltration moves collected data out of the network through various channels, often with compression or encryption to avoid DLP systems.

Examples: exfiltration over C2 channel, exfiltration over alternative protocol, exfiltration to cloud storage, scheduled transfers, physical medium exfiltration.

TA0040 — Impact

The adversary disrupts, degrades, or destroys systems and data. Impact techniques achieve the adversary's final objective when destruction or disruption — rather than data theft — is the goal.

Examples: data encryption for ransomware, data destruction, defacement, denial of service, resource hijacking (cryptomining), account access removal.

Tactic Ordering

The matrix flows left-to-right in a general operational sequence:

Reconnaissance → Resource Development → Initial Access → Execution →
Persistence → Privilege Escalation → Defense Evasion → Credential Access →
Discovery → Lateral Movement → Collection → C2 → Exfiltration → Impact

Key points about ordering: - Reconnaissance and Resource Development happen before any target interaction (pre-compromise) - Defense Evasion applies throughout the entire operation, not just at one point - Exfiltration and Impact are often the adversary's final objectives, but not always — some operations focus purely on long-term access (espionage) - Adversaries loop back frequently — discovery leads to lateral movement, which leads to more discovery

ATT&CK Beyond Enterprise

ATT&CK covers multiple technology domains:

Matrix Scope
Enterprise Windows, Linux, macOS, cloud (Azure AD, AWS, GCP, etc.), network, containers
Mobile Android and iOS
ICS Industrial Control Systems

The Enterprise matrix is the most widely used. Each matrix has its own set of tactics and techniques tailored to the platform.

Practical Examples

Mapping a Simple Attack to Tactics

A typical web application compromise maps across multiple tactics:

Step Action Tactic
1 Scan target for open ports and services TA0043 — Reconnaissance
2 Exploit vulnerable web application TA0001 — Initial Access
3 Execute reverse shell TA0002 — Execution
4 Enumerate local system and users TA0007 — Discovery
5 Find SUID binary and escalate to root TA0004 — Privilege Escalation
6 Install SSH key for persistence TA0003 — Persistence
7 Dump /etc/shadow and crack hashes TA0006 — Credential Access
8 SSH to internal database server TA0008 — Lateral Movement
9 Export customer database TA0009 — Collection
10 Transfer data out via HTTPS TA0010 — Exfiltration

Using Tactic IDs in Reports

When writing pentest reports, tag every finding with its ATT&CK tactic and technique:

Finding: SQL Injection in /api/login endpoint
ATT&CK Mapping:
  - Tactic: TA0001 (Initial Access)
  - Technique: T1190 (Exploit Public-Facing Application)

This allows defenders to map your findings directly to their detection matrix and prioritize remediation.

References

Official Documentation