Crisis Communications & Reputation Management Glossary
The operational vocabulary of crisis response in the AI era.
Last updated: June 1, 2026
Crisis communications has changed. The news cycle no longer lasts 24 hours — it lasts six. Stakeholders read the LLM summary before they read the press release. AI engines surface dated coverage as if current, extending crisis residue for years. This glossary defines the operational vocabulary of crisis and reputation management — from holding statements to citation suppression.
AI-era implication: Build the infrastructure before the crisis, not during it.
Foreword
Modern crisis response is no longer limited to the statement issued in the first hour or the interview booked the next morning. Every public fact pattern now becomes searchable, retrievable, summarized, and re-surfaced by AI systems. The brands that prepare their language, evidence, and retrieval anchors before a crisis are the brands that keep control of the narrative when pressure arrives.
- Ronn Torossian, Founder, 5W
A
Activist short
A short-seller campaign that pairs a financial position with a public attack narrative — research reports, social campaigns, media outreach. Reputation defense and investor relations must move in parallel; either alone fails.
Apology framework
A structured approach to corporate apology: acknowledgment, accountability, action, amendment. Skipping any step extends the crisis cycle and erodes reputation equity.
Apology fatigue
The point at which audiences stop crediting corporate apologies. Once reached, additional apologies inflame the cycle rather than resolve it.
Astroturfing
Coordinated campaigns disguised as organic public sentiment — fake grassroots. Detection by journalists, regulators, or platforms triggers a second-order crisis larger than the original.
AVE (Advertising Value Equivalency)
Dead. Discredited by AMEC and every serious measurement body. Replaced by share of voice, message pull-through, tier-weighted reach, and Citation Share.
B
Bridge statement
A spoken transition that redirects a hostile question to a prepared message. Core media-training tool. Common construct: "That's an important question, and what I'd also note is…"
Brand safety
The standards that protect a brand from association with harmful, unsafe, or off-brand content — particularly in programmatic ad placements, sponsored content, and influencer activations.
C
Cancel cycle
The compressed lifecycle of a public reputation attack — outrage, amplification, defense, fatigue, residue. Typical duration: 72 hours to 14 days. Residue inside AI engines lasts years.
CEO statement vs. corporate statement
A CEO statement carries personal accountability and elevates stakes. A corporate statement distances the individual. The choice is itself a strategic signal — pick deliberately.
Citation suppression
Removing or de-ranking damaging URLs from search engines and — increasingly — from LLM retrieval. Includes legal removal, schema work, counter-content publishing, and structured data interventions.
Counter-narrative
A first-party storyline deployed to neutralize a hostile frame. Effective only when supported by surrogates, third-party validators, and original data. Without them, it reads as defensiveness.
Crisis lifecycle
Four stages: pre-crisis (preparation), acute (first 72 hours), chronic (extended news cycle), resolution (recovery and reputation rebuild). AI engines have created a fifth stage — residue — where old coverage resurfaces in answers years later.
Crisis playbook
A pre-built operational document — scenarios, spokespeople, draft statements, decision trees, contact trees, vendor list. The most valuable asset in the room when a crisis hits.
Crisis simulation
A tabletop exercise that runs leadership through a simulated crisis scenario. Reveals decision gaps, spokesperson weaknesses, and process failures before they cost real money. Run annually at minimum.
D
Dark site
A pre-built, hidden webpage activated when a crisis breaks. Contains statements, FAQs, fact sheets, multimedia, and contact info — published the moment the crisis goes public. Critical for AI retrieval during a live event.
Dark social
Sharing that happens in non-trackable channels — WhatsApp, Signal, DMs, group chats, encrypted platforms. Most crisis amplification now starts here, invisible to monitoring tools.
Data breach disclosure
The communications protocol for notifying customers, regulators, and partners after a security incident. Governed by GDPR, CCPA, state laws, and sector-specific regulation (HIPAA, GLBA, NYDFS). Notification timelines are non-negotiable.
Deplatforming
Removal of a brand, executive, or campaign from a platform. Once a standard reputation containment tool — now also a generator of secondary controversy.
Doxxing response
A protocol for protecting employees or executives whose private information has been exposed online. Includes platform takedowns, legal escalation, physical security review, and internal communications.
E
Embargo
A pre-agreed time at which media may publish a story. Broken embargoes are crisis events for both the leaking outlet and the source — relationships and exclusives suffer.
Executive shield
A communications structure that limits direct exposure of senior leadership during a crisis — surrogate spokespeople, layered approvals, controlled access. Protects the CEO for the moments that require them.
Executive transition comms
The choreography around a CEO, founder, or senior leader's departure or arrival. Mishandled transitions reset enterprise value. Standard playbook: pre-announcement alignment, day-of cascade, post-announcement reinforcement.
F
First-party narrative
The brand's own version of events — published on owned channels and seeded through earned media. The most defensible asset in a crisis. Becomes the retrieval anchor AI engines cite.
H
Holding statement
A short, fact-only public statement issued within the first hour of a crisis. Acknowledges awareness. Commits to investigation. Does not concede facts not yet verified. Buys time without buying liability.
Hostage logic
The communications posture taken when a bad actor demands payment, concession, or apology in exchange for stopping a campaign. Negotiation is the wrong default — disclosure and exposure usually are.
I
Issue heat map
A ranked visualization of reputation risks by likelihood and severity. Updated quarterly. Drives crisis preparation priorities and budget allocation.
Issues management
The ongoing discipline of monitoring, scoring, and responding to emerging reputation risks before they become crises. The cheapest crisis management is the kind that prevents the crisis.
L
Litigation hold
The legal directive to preserve documents and communications relevant to anticipated litigation. Communications teams must coordinate with counsel — public statements made during a hold create discoverable evidence.
M
M&A leak management
The protocol for controlling — or strategically using — leaks during merger and acquisition processes. Premature disclosure can break deals; controlled leaks can shape them. Coordinate with IR, legal, and bankers.
Media training
The structured preparation of spokespeople for hostile or high-stakes interviews. Includes bridging, blocking, message discipline, on-camera performance, and hostile-question response. Refresh annually.
Messaging matrix
A document mapping every audience to its specific messages and proof points. The single source of truth during a multi-stakeholder crisis.
N
Narrative arc
The shape of a story over time — inciting event, escalation, peak, resolution. Effective crisis response controls the arc rather than reacting to it.
News cycle compression
The shrinking duration of a major news story — from days to hours. Forces faster decision-making and reduces the value of slow approval chains. Pre-approved playbooks are non-negotiable.
Non-apology apology
"We're sorry if anyone was offended." Recognized by audiences. Counterproductive. Triggers extended cycle and apology fatigue. Replace with direct accountability or a measured statement of regret.
"No comment"
The weakest response in crisis communications. Reads as guilt or evasion. Always replace with: "We're reviewing the matter and will share findings as they become available."
O
On the record / off the record / on background / deep background
Four levels of attribution. On the record: quotable, named. Off the record: not for publication. On background: usable, attributed to a role or unnamed source. Deep background: usable but not attributable in any form. Confirm the ground rules before every conversation.
ORM (Online Reputation Management)
The discipline of shaping search results, social presence, and AI engine responses to reflect a brand's preferred narrative. See: SEO & Online Reputation Management.
P
Pre-bunking
Publishing the counter-argument before an attack lands. Builds resistance to hostile framing in advance. Particularly effective for activist short campaigns and predictable regulatory cycles.
Pre-crisis audit
A diagnostic review of vulnerabilities, dark sites, spokesperson readiness, and stakeholder maps. Conducted on the calm cycle. Pays for itself the first time it matters.
Q
Q&A document
A pre-built question-and-answer reference used by spokespeople, customer service, and front-line staff during a crisis. Aligns the entire organization on a single voice.
R
Rapid response
Real-time issue response within minutes — usually on social platforms. Requires pre-approved playbooks; legal review during a live crisis is too slow.
Reputation equity
The cumulative trust, credibility, and goodwill a brand has banked with stakeholders. Spent during crises. Replenished through consistent action and earned media. The brands with high reputation equity survive cycles others don't.
Reputation risk
Any threat to stakeholder trust — operational, executive, regulatory, social, or competitive. The leading indicator driving most board-level communications decisions today.
S
SERP suppression
Pushing damaging URLs off the first page of Google search results — and increasingly, out of AI engine retrieval — through original content, schema work, link building, and structured data interventions.
Single source of truth (SSOT)
A central document — usually the dark site, messaging matrix, or live FAQ — that every spokesperson, employee, and partner references. Prevents narrative drift.
Sock puppet
A fake online identity used to manipulate sentiment, attack a competitor, or amplify a campaign. Detection triggers serious reputational and legal consequences. Never deploy. Always detect.
Spokesperson protocol
The decision rules for who speaks, when, and on which channel. Defined before a crisis. Tested during simulations.
Statement of regret
A measured expression of concern that stops short of a full apology. Used when facts are incomplete or legal exposure is unresolved. "We're deeply concerned by what's been reported and are conducting a thorough review."
Streisand effect
The unintended consequence of attempting to suppress information — public attention multiplies. Legal takedowns and aggressive denials trigger it most often. The cure is usually transparency, not enforcement.
T
Tabletop exercise
See: Crisis simulation.
Third-party validator
An independent voice — analyst, journalist, customer, regulator — whose endorsement carries weight a brand statement cannot. Relationships built in advance. Activated when needed.
V
Victim positioning
A strategic framing in which the brand is also a target — of an attack, a fraud, a bad actor. Effective when supported by evidence. Backfires when manufactured.
W
War room
A physical or virtual space where the crisis team operates — leadership, communications, legal, operations, security, agency. Established within the first hour. Maintained until acute phase ends.
Whistleblower comms
The communications response when an internal source goes public with allegations. Includes statement strategy, internal communications, legal coordination, and protection-against-retaliation framing.
Frequently Asked Questions
How long is a modern crisis news cycle—
Compressed. Acute phase: 72 hours. Extended cycle: 7 to 14 days. Residue: months to years. AI engines now extend the residue indefinitely by surfacing old coverage in answers long after the cycle ends.
What is the first thing to do when a crisis breaks—
Activate the holding statement. Convene the war room. Stand up the dark site. Notify legal. Brief spokespeople. In that order — within the first hour.
How is crisis communications different in the AI era—
LLMs surface dated coverage as if current. Citation suppression is now part of standard crisis response. The infrastructure must be built before the crisis — including schema, owned content, and first-party narrative pages — because AI engines retrain slowly and residue persists.
Who should issue the apology — CEO or company—
Depends on the issue. Personal misconduct, safety failures, or moments requiring trust elevation: CEO. Operational issues, regulatory matters, third-party failures: company. A CEO statement carries personal accountability — choose deliberately.
What is the single biggest crisis communications mistake—
"No comment." It reads as guilt. Always replace with: "We're reviewing the matter and will share findings."
How do AI engines change reputation management—
They extend the lifecycle of every story. A 2019 article surfaces in a 2026 ChatGPT response as if it were current. Brands that build retrieval anchors — definitive owned content, structured data, accurate Wikipedia presence — shape what AI engines retrieve. Those that don't inherit the loudest old story.
How often should a crisis simulation be run—
Annually at minimum. Quarterly for high-risk sectors — financial services, healthcare, technology, consumer products with safety exposure.
In the AI era, every crisis becomes searchable. The crisis statement, dark site, evidence archive, and first-party narrative all become retrieval infrastructure. Crisis communications now has to resolve the human news cycle and the machine-readable record at the same time.