Chapter 05

IR Communication

How an incident is communicated often determines how it ends. Poor communication during an incident multiplies damage — through rumor, panic, regulatory missteps, and legal exposure — that good technical work alone cannot repair.

Scenario

A hospital network discovers ransomware across 200 clinical systems at 2 AM. The on-call security analyst begins IR. By 3 AM, the hospital's normal Slack workspace is compromised — the attacker has read access to internal channels. The IR team, not yet aware of this, uses Slack to coordinate: sharing which systems are isolated, which contain PHI, and where their incident command is located. A nurse manager, alarmed by early-morning IT alerts, posts in a public hospital staff Slack channel asking what's happening. Someone responds "I heard it's ransomware." Three physicians screenshot that message and forward it to contacts. By 6 AM, a local news station has a tip. The hospital's legal team spends the next 48 hours trying to suppress news coverage while simultaneously managing the technical incident — all because early communications happened over a compromised channel in a disorganized way. The technical incident lasted 72 hours. The communications fallout lasted months.

Out-of-Band Communications

Out-of-band (OOB) communications are IR coordination channels that are separate from the corporate infrastructure potentially affected by the incident. If your email server is compromised, you cannot use that email server to coordinate the investigation. If your Slack workspace may have been accessed, you cannot use Slack to discuss sensitive investigation findings. Out-of-band communications are not a nice-to-have — they are a prerequisite for operating during an incident.

  OOB Communication Setup
  ══════════════════════════════════════════════════════════════════════

  WHAT TO ESTABLISH BEFORE AN INCIDENT
  ─────────────────────────────────────────────────────────────────
  Phone tree with personal cell numbers
  → Corporate phone directory is not OOB (if AD is compromised,
    phone directory may be inaccessible). Personal cell numbers
    must be collected in advance.

  Out-of-band group chat (Signal, WhatsApp, or similar)
  → Not on corporate infrastructure. Not integrated with corporate SSO.
  → Group includes: IR lead, key IR analysts, CISO, legal counsel,
    communications/PR, executive sponsor.
  → Keep this small. Each additional person is a potential disclosure.

  Out-of-band video (personal Zoom/Teams account or pre-provisioned
  incident bridge line)
  → "War room" calls happen here, not on corporate infrastructure

  Pre-established IR retainer contact information
  → Your external IR firm's emergency contact number — not their
    website, which you may not be able to reach during the incident

  COMMON FAILURE MODES
  ─────────────────────────────────────────────────────────────────
  Using corporate email to discuss a compromised email server
  Using Slack to coordinate while the attacker has Slack access
  Sharing investigation findings on internal wikis the attacker
    can access (Confluence, SharePoint) — attacker reads your progress
  Using compromised domain admin account for IR coordination
  Forgetting to onboard legal to OOB comms — legal finds out
    the same way the CFO does (from the CISO's update email)
Why "attacker reads your IR communications" is a real attack scenario

APT actors conducting prolonged intrusions routinely monitor IR activity once they detect that discovery is imminent. If your IR team is coordinating over email and the attacker has mailbox access, they read your containment plans before you execute them. If you're coordinating over Slack and the attacker has user credentials, they watch your pivot-finding in real time. This leads to a specific, documented behavior pattern: the attacker accelerates exfiltration or deploys ransomware before you can isolate them, because they can see your isolation plans forming. Requiring OOB communications from the first moment of confirmed compromise — not after you've established what's been accessed — prevents the attacker from using your own coordination channel against you.

SITREP Format: Structured Situation Reports

A SITREP (situation report) is a brief structured update about the current state of the incident. The purpose is to give stakeholders the information they need to make decisions or take actions without requiring them to wade through technical detail. A good SITREP can be written in 10 minutes and read in 3. A bad SITREP is either so technical it's incomprehensible to leadership or so vague it's actionless.

  SITREP Structure
  ══════════════════════════════════════════════════════════════════════

  HEADER
  ─────────────────────────────────────────────────────────────────
  Incident: [brief title — e.g., "Ransomware Incident - Manufacturing Network"]
  SITREP #: [sequential number — allows recipients to verify they have
              the most recent update]
  Prepared by: [name, role]
  Date/time: [UTC timestamp]
  Classification: [Internal Only / Legal Privilege / Restricted]

  SITUATION (2–3 sentences)
  ─────────────────────────────────────────────────────────────────
  What happened. What is confirmed. What is suspected but not confirmed.
  Current status (active, contained, in recovery, closed).
  Example: "On [date] at [time], ransomware was deployed across approximately
  200 endpoints in the manufacturing network segment. The initial access
  vector is under investigation; credential compromise via VPN is suspected.
  The incident is currently in the containment phase."

  SCOPE (bullets)
  ─────────────────────────────────────────────────────────────────
  • Systems confirmed affected: [number, type, location]
  • Data potentially accessed: [category, estimated sensitivity]
  • Business impact: [systems offline, services unavailable, revenue impact]
  • Geographic scope: [which sites, which regions]

  ACTIONS TAKEN (bullets)
  ─────────────────────────────────────────────────────────────────
  • Completed actions with timestamps
  Example: "[14:30 UTC] Isolated affected network segment via firewall rule"
           "[15:00 UTC] Engaged external IR firm [name]"
           "[15:30 UTC] Notified legal counsel"

  ACTIONS PENDING (bullets)
  ─────────────────────────────────────────────────────────────────
  • What is being done in the next 24 hours and by whom
  • Decisions needed from leadership (clearly called out)

  NEXT UPDATE
  ─────────────────────────────────────────────────────────────────
  "Next SITREP will be issued at [time] or upon material change."

Stakeholder Communication Tiers

Different stakeholders need different information at different times. Sending the wrong level of detail to the wrong audience creates problems: executives who receive raw IOCs cannot act on them; analysts who receive a board-level summary lack the detail to coordinate. Defining stakeholder tiers in advance prevents both scenarios.

TierAudienceWhat they needWhat they do not needCadence
Technical IR analysts, SOC, IT operations Full technical detail: IOCs, affected systems, investigation findings, containment actions, tool outputs Business impact narrative (they're building it) Continuous during active IR
Management CISO, IT leadership, business unit managers SITREP: scope, business impact, decisions needed, resource requirements Raw log output, technical IOC lists Every 4–8 hours during active phase; daily thereafter
Executive CEO, CFO, board Business impact summary, key decisions, regulatory exposure, public communications status Technical details; specific system names; investigation methodology At key milestones: confirmation, containment, eradication, closure. Major changes in scope.
Legal/Compliance General counsel, outside counsel, compliance team Notification obligations, data categories potentially accessed, timeline of events, preservation holds needed Speculative attribution; preliminary findings that may change As soon as data access is suspected; immediately before any external notifications
External Regulators, customers, partners, media Controlled by legal and communications; what is required by law or obligation Internal investigation details that create additional liability As required by regulation or obligation; not ad hoc
Discoverable written communications during IR

Written communications during an incident — emails, Slack messages, tickets, wiki edits — are discoverable in litigation. "What a mess, I can't believe we missed this" sent by the CISO in a Slack message at 3 AM becomes an exhibit in a class action lawsuit. "Our logging was clearly inadequate" becomes evidence of known security gaps that were not addressed. This does not mean IR teams should not communicate freely among themselves — they absolutely must. It means that factual communications (what happened, what was done, when, by whom) are appropriate, while editorializing about failures, assigning blame, or speculating about what could have been done differently should be reserved for verbal discussions or communications with legal counsel that may be covered by attorney-client privilege. Train IR team members on this distinction before the incident, not during it.

Media and Public Communications

When an incident becomes public — through a leak, a regulatory disclosure, or voluntary disclosure — the public communication becomes a separate and significant workstream. Security practitioners are not typically trained in crisis communications, and attempting to manage media response without professional help during an active technical incident is nearly always a mistake.

  IR Communication: Who Owns What
  ══════════════════════════════════════════════════════════════════════

  INCIDENT COMMAND (IR Lead / CISO)
  ─────────────────────────────────────────────────────────────────
  Owns: technical investigation, containment, eradication, recovery
  Decision authority: technical containment actions, IR firm engagement,
    evidence preservation, scope determination
  Communicates to: management stakeholders, legal

  LEGAL COUNSEL
  ─────────────────────────────────────────────────────────────────
  Owns: regulatory notification decisions, litigation holds, external
    disclosure content
  Decision authority: what to disclose externally and when
  Communicates to: regulators, external counsel, partner legal teams

  COMMUNICATIONS / PR
  ─────────────────────────────────────────────────────────────────
  Owns: external public statements, media response, customer notifications
  Decision authority: messaging, timing of public communications
  Communicates to: media, customers, public

  SINGLE SPOKESPERSON RULE
  ─────────────────────────────────────────────────────────────────
  One person — typically from communications, cleared by legal —
  answers external questions. Security analysts do not speak to media.
  IT staff do not speak to customers about the incident.
  Anyone who receives press inquiries routes them to the spokesperson.

  WHY THIS MATTERS:
  Inconsistent external statements create legal liability and extend
  the reputational crisis. A statement that "only 500 records were
  accessed" followed two weeks later by "we now believe 50,000 records
  were accessed" is dramatically worse than a more cautious initial
  statement of "the scope of data access is under investigation."

Partner and Third-Party Notification

Modern organizations are embedded in supply chains, service provider relationships, and technical integrations. When you are compromised, the attacker may have used your position to pivot to partners. When a partner is compromised, they may be the entry point to your environment. Notification obligations flow both ways, and managing them poorly damages long-term business relationships.

ScenarioNotification obligationTimingContent
You are compromised; attacker may have used VPN access to reach a partner's systems Notify partner; may be contractual obligation in your MSA/SLA As soon as pivot is suspected; before partner detects it themselves IOCs (C2 IPs, hashes, file names), timing of suspected lateral movement, your points of contact for coordinated investigation
You are compromised; partner data stored in your environment may be affected Notify partner; likely contractual and regulatory obligation Per contract/regulation; GDPR 72 hours from awareness of breach What data types were potentially accessed; your investigation status; steps taken to contain
Partner notifies you they are compromised; they had access to your systems Treat as potential incident in your own environment; begin investigation Immediately begin internal hunting for partner-related indicators Request partner IOCs; temporarily restrict or revoke partner access pending investigation
Managed service provider (MSP) compromise affecting your tenancy Notify your own legal/compliance; assess scope of MSP access to your data As soon as MSP notifies you or you discover independently Obtain full list of MSP access levels; assume all credentials they held are compromised
Mental model: communication failures multiply technical damage

A ransomware incident that is technically contained in 72 hours but communicated poorly can result in months of regulatory exposure, loss of customer contracts, and leadership changes. A ransomware incident that takes 2 weeks to contain but is communicated transparently and professionally often results in stronger relationships with regulators and customers, not weaker ones. The technical work determines whether the business survives the immediate crisis. The communication work determines how the business is perceived to have handled the crisis — which affects every relationship the business has going forward. IR practitioners who think their job ends with recovery are ignoring half the incident.

Q & A

When should OOB communications be established — before the incident or during it?

Before the incident, without exception. During an active incident, establishing OOB communications is difficult, slow, and inconsistently done. Under incident pressure, people default to familiar tools — which are the compromised tools. OOB comms must be established during Preparation: personal cell numbers collected into a physical or secure-access phone tree, a Signal or WhatsApp group pre-created with the core IR team, an out-of-band video bridge pre-provisioned and tested. The test is important — "we have a Signal group" is not sufficient if half the IR team has never used Signal and doesn't have it installed. Run the OOB comms channel in tabletop exercises so it is familiar when it is needed. If you arrive at an incident and OOB comms do not exist, establish them on personal devices immediately — before discussing anything sensitive.

What is attorney-client privilege and how does it apply to IR communications?

Attorney-client privilege is a legal protection that can, in some circumstances, protect communications between an attorney and their client from disclosure in litigation. When outside counsel directs or supervises an IR investigation (rather than the internal security team directing it), some of the investigation's findings and communications may be protected from discovery under the "work product doctrine" — the theory being that documents created in anticipation of litigation are protected. This is a complex legal area and the protections are not absolute — they can be waived by disclosure to parties outside the privilege relationship, and courts disagree on how broadly they apply to IR investigations. The practical implication: early in a significant incident, bring outside counsel in and have them be the client of record for the IR firm engagement if litigation is anticipated. Have factual communications directed through or to counsel. Do not rely on this as a strategy to hide the incident from regulators — regulators have mechanisms to pierce privilege — but it can protect candid internal investigation findings from plaintiff's counsel in civil litigation.

How do you communicate uncertainty during an incident without undermining confidence?

Distinguish clearly between what is confirmed, what is suspected, and what is unknown — and label each category explicitly in every communication. Stakeholders do not lose confidence when organizations acknowledge uncertainty; they lose confidence when organizations state things confidently that are later proven wrong. "We believe approximately 500 accounts were compromised, based on our investigation so far; we are continuing to assess scope" is a confident statement that acknowledges its own uncertainty. "500 accounts were compromised" stated as a conclusion that is later revised to 5,000 is a credibility-destroying communication failure. For regulatory communications specifically, explicitly qualifying scope estimates as preliminary is not weakness — it is standard practice that regulators expect. The failure mode to avoid is silence: not communicating because "we don't know everything yet" leaves stakeholders without the information they need to make decisions.

A junior analyst received a call from a journalist asking about the incident. What should they do?

The analyst should not provide any information to the journalist. The response is: "I cannot comment on that. Our communications team handles media inquiries. I will pass along your contact information." The analyst should then immediately notify the IR lead and communications team that a journalist has made contact, including the journalist's name, publication, and what they asked. This is not a failure — press contacts happening is expected in significant incidents. The failure would be if any analyst, under pressure or trying to be helpful, provided even apparently innocuous information like "yes, we're aware of an issue" or "I can confirm it happened on Tuesday." Every piece of confirmed information narrows the story and can be used without the organization's consent. All external communications about the incident route through a single spokesperson after legal approval. This should be stated explicitly to IR team members at the beginning of every significant incident.

How should the IR lead communicate a significant scope expansion (from 50 hosts to 500 hosts) to executive leadership?

Directly and quickly, without softening the number in a way that obscures its significance. Executive stakeholders need accurate scope information to make decisions about resource allocation, customer notification, regulatory disclosure, and public communications. A SITREP issued immediately on material scope changes is the mechanism — this is why SITREPs include a clause like "Next update upon material change." The communication should: state the revised scope clearly, explain briefly why the scope changed (investigation uncovered additional compromised systems, not that the original assessment was careless), describe what it means for business impact, state what additional resources or decisions are needed. Executives who receive accurate bad news quickly can make good decisions. Executives who receive softened or delayed bad news and make decisions based on it, only to have the picture worsen, lose confidence in the IR team's ability to assess reality. Delivering hard news promptly is a form of professional credibility.

What is a litigation hold and when should one be issued during IR?

A litigation hold (also called a legal hold) is an instruction, typically issued by legal counsel, to preserve all potentially relevant documents and data — suspending the normal document retention and deletion schedule. Once litigation is reasonably anticipated, the organization has a legal obligation to preserve relevant evidence. Destroying evidence after a litigation hold should have been issued (even if one was not formally issued) can result in sanctions against the organization in court. In an IR context: a litigation hold should be considered as soon as personal data breach, significant financial impact, or third-party harm is suspected. Legal counsel issues the hold. The security team's role is to (a) notify legal of the incident as early as possible, (b) not destroy anything — including rebooting and reimaging systems — until legal has assessed preservation needs, and (c) ensure that log sources with short retention windows are extended or exported before they expire. The tension between "we need to reimage this system to restore business operations" and "we need to preserve this system for litigation" is real and must be resolved by legal counsel, not unilaterally by the IR team.