Online Reputation, Issues, and Crisis Management

Introduction

A brand’s reputation is the accumulated judgment that stakeholders form about its identity, conduct, competence, and social significance. In digital environments, reputation is especially dynamic. Customer reviews, employee posts, influencer commentary, search results, news coverage, platform algorithms, and private messages can all affect public interpretation of a brand. A single event may therefore move rapidly from an isolated complaint to a widely discussed issue or crisis.

Digital reputation management is not simply the practice of responding to negative comments. It is a strategic capability involving environmental scanning, evidence assessment, risk classification, stakeholder communication, decision rights, organizational learning, and trust repair. It requires managers to distinguish between legitimate criticism and misinformation, between an issue that is developing and a crisis that is already causing severe harm, and between a response that merely reduces visibility and one that addresses underlying causes.

The central managerial challenge is that organizations must act under conditions of uncertainty. Information may be incomplete, contradictory, or deliberately manipulated. At the same time, silence can be interpreted as indifference, while an inaccurate or defensive response can intensify the problem. Effective practice therefore combines speed with verification, empathy with accountability, and communication with operational correction.

Learning Objectives

By the end of this chapter, students should be able to:

1. Explain the nature and strategic value of online brand reputation.
2. Identify and interpret reputation signals across digital channels.
3. Distinguish complaints, misinformation, issues, and crises.
4. Assess reputational risk using structured criteria.
5. Design escalation, response, recovery, and stakeholder communication protocols.
6. Evaluate the role of leadership, governance, technology, and artificial intelligence in crisis management.
7. Apply post-crisis learning methods to improve organizational resilience.

1. Reputation as a Strategic Asset

Reputation differs from image and identity. Brand identity is what the organization seeks to express about itself. Brand image is the set of associations held by particular audiences at a given time. Reputation is a broader and more durable social evaluation based on accumulated evidence and stakeholder experience. Reputation includes perceptions of quality, reliability, fairness, responsibility, transparency, and capability.

Reputation has at least four strategic effects:

• It influences customer consideration, loyalty, and willingness to forgive mistakes.
• It affects employee attraction, retention, and morale.
• It shapes investor, regulator, supplier, and community confidence.
• It provides or removes strategic freedom during periods of uncertainty.

A strong reputation is not immunity from criticism. In some circumstances, highly visible brands receive more scrutiny. Rather, reputational strength increases the likelihood that stakeholders will interpret ambiguous information charitably, allow the organization time to respond, and believe credible corrective action.

Digital environments alter the economics of reputation in five ways. First, information is persistent: deleted posts may be archived or reproduced. Second, information is searchable: old controversies can reappear when a brand is under attention. Third, information is networked: content can move across platforms and audiences. Fourth, visibility is algorithmically mediated: engagement and emotional intensity may receive greater distribution than accuracy. Fifth, audiences are participatory: customers, employees, activists, creators, and automated accounts can all influence the narrative.

2. Reputation Signals and Social Listening

A reputation signal is any observable indication of stakeholder evaluation or emerging risk. Signals may be quantitative, such as a fall in review ratings, or qualitative, such as a credible allegation of unsafe practice. They may originate inside or outside official brand channels.

2.1 Main classes of signals

Monitoring should not be equated with collecting every mention. Effective social listening establishes a purpose, a defined scope, appropriate sources, and clear decision thresholds. A monitoring system should answer four questions:

1. What is being said?
2. Who is saying it, and which stakeholders may be affected?
3. Is the claim credible and materially important?
4. What decision, if any, should follow?

2.2 Interpreting volume and sentiment

Mention volume can identify unusual activity, but volume alone is not risk. A small number of credible allegations involving safety, discrimination, privacy, or fraud may be more serious than thousands of low-value complaints. Sentiment analysis can support triage, but automated sentiment classification often misreads sarcasm, cultural language, code-switching, and context. Human review is necessary for consequential decisions.

A useful conceptual measure is the weighted signal load:

Here, L is the overall signal load, sᵢ is the observed intensity of signal category i, and wᵢ is its managerial weight. Weights may reflect credibility, stakeholder importance, potential harm, velocity, and legal sensitivity. This is not a universal scientific law; it is a decision-support device. Managers must define categories consistently and avoid treating the result as an objective measure of reputation.

Another useful metric is the rate of change in relevant mentions:

where Vₜ is the volume during the current period and Vₜ₋₁ is the volume during the previous comparable period. A percentage change may be calculated as:

This calculation assumes that the periods, sources, query definitions, and measurement methods are comparable and that Vₜ₋₁ is not zero. It indicates attention dynamics, not truth, severity, or causation.

3. Complaints, Misinformation, Issues, and Crises

Classification determines the appropriate response. Misclassification produces two common failures: overreaction, which amplifies a minor matter, and underreaction, which allows serious harm to grow.

3.1 Complaint

A complaint is an expression of dissatisfaction about a product, service, interaction, policy, or outcome. Complaints are usually localized and may be resolved through customer service. However, repeated complaints can reveal a systemic issue. A complaint should never be dismissed merely because it is expressed emotionally.

3.2 Misinformation and disinformation

Misinformation is inaccurate information shared without demonstrable intent to deceive. Disinformation is deliberately false or manipulated information. The distinction concerns intent, which is often difficult to establish. Managers should focus first on evidential status and potential harm rather than prematurely assigning motive.

A false claim may be corrected with evidence, but correction strategy depends on reach, credibility, and risk. Repeating a low-visibility falsehood can unintentionally increase its visibility. Conversely, silence may be irresponsible when the claim concerns safety or urgent public action.

3.3 Issue

An issue is a developing matter that may affect stakeholder relationships, legitimacy, operations, or strategic objectives. Issues often have identifiable causes and may be influenced through policy, engagement, and corrective action. Examples include recurring accessibility complaints, criticism of labor practices, or concern about data use.

Issue management is proactive. It involves identifying an expectation gap, researching stakeholder concerns, assigning ownership, engaging affected groups, and changing practices where warranted.

3.4 Crisis

A crisis is a high-impact event or situation that threatens important organizational objectives, stakeholder well-being, legitimacy, or operational continuity and requires an urgent, coordinated response. A crisis is not defined solely by online popularity. A viral post may be harmless, while a low-visibility product safety failure may be a genuine crisis.

The following diagnostic matrix supports classification:

The matrix is a guide, not a mechanical decision rule. A single high-severity dimension may justify escalation even when other dimensions appear moderate.

4. Reputation Risk Assessment

Risk assessment combines evidence, potential consequences, and uncertainty. A practical model is:

where R is estimated risk, P is the probability of an adverse outcome, and I is the impact if that outcome occurs. The model assumes that probability and impact can be assessed on comparable ordinal or numerical scales. It should not be interpreted as a precise probability of reputational damage. In many reputation decisions, uncertainty, reversibility, stakeholder vulnerability, and speed of escalation must be recorded separately.

A richer assessment can use a risk register containing:

• the claim or event;
• source and evidence quality;
• affected stakeholders;
• potential legal, ethical, financial, operational, and reputational consequences;
• velocity and likely duration;
• current organizational responsibility;
• decision owner;
• containment actions;
• communication obligations;
• review time.

Worked application

Suppose a brand identifies three developing matters on a five-point scale, where 1 is low and 5 is high. Matter A has probability 4 and impact 2, producing R = 4 × 2 = 8. Matter B has probability 2 and impact 5, producing R = 2 × 5 = 10. Matter C has probability 4 and impact 5, producing R = 20. Matter C receives the highest numerical priority, but Matter B may still require immediate action because its impact involves a vulnerable stakeholder group. The numbers support discussion; they do not replace ethical judgment or legal advice.

5. A Governance Architecture for Escalation

Crisis readiness depends on governance established before an event. A social media team cannot be expected to determine product safety, legal exposure, employee rights, and public policy alone.

A robust architecture includes:

• Monitoring owners, who detect and document signals.
• Triage owners, who classify severity and evidence.
• Subject-matter experts, including legal, compliance, security, safety, human resources, and operations.
• An incident or crisis leader, who coordinates decisions.
• An executive decision-maker, who approves high-consequence actions.
• A communications lead, who manages public, employee, media, and partner communication.
• A documentation function, which maintains a decision log and evidence record.

Escalation criteria should be explicit. Immediate escalation is appropriate for allegations involving death or serious injury, child safety, discrimination, sexual misconduct, cybersecurity, privacy, fraud, regulatory action, widespread service failure, or credible threats of violence. Escalation is also appropriate when the organization lacks authority or expertise to verify the claim.

A decision log should record what was known, when it was known, who decided, what evidence was used, what assumptions were made, and when the decision will be reviewed. This protects accountability and enables learning.

6. Response Principles and Stakeholder Communication

A strong response is both communicative and operational. Communication cannot compensate for an unchanged harmful practice. The following principles are widely applicable:

6.1 Establish facts without delaying necessary protection

Organizations should separate verified facts, plausible but unverified claims, and known unknowns. When immediate safety is at stake, protective action may be necessary before the investigation is complete. The organization can state what it is doing without claiming certainty it does not possess.

6.2 Acknowledge impact and show empathy

An acknowledgment recognizes that stakeholders have experienced concern, inconvenience, loss, or harm. It should avoid performative language and should not imply that the issue is merely a public-relations problem.

6.3 Accept responsibility where warranted

Responsibility may involve direct causation, failure of oversight, inadequate controls, or delayed response. Legal review may be necessary, but legal caution should not become evasion. Statements such as “we are sorry you feel this way” often fail because they do not acknowledge the event or its consequences.

6.4 State specific actions and time horizons

Stakeholders need to know what the organization is doing now, what it will investigate, who is affected, and when an update will occur. Specificity increases credibility, provided that commitments are realistic.

6.5 Match the channel to the stakeholder

A public statement may be needed for broad awareness, while affected customers require direct notification, employees need internal guidance, regulators require formal reporting, and journalists may need a media briefing. One message should not be copied indiscriminately across every audience.

6.6 Correct misinformation proportionately

A correction should identify the accurate information, provide evidence where possible, and explain any action needed from stakeholders. The response should avoid repeating harmful claims unnecessarily. Platform reporting, search-result management, and legal remedies may be appropriate in cases involving impersonation, fraud, threats, or unlawful content, but takedown requests should not be used to suppress legitimate criticism.

A useful initial holding statement has five elements:

1. Recognition of the matter.
2. Concern for affected stakeholders.
3. What is currently verified.
4. Immediate protective or investigative action.
5. A specific time for the next update.

A holding statement is not a substitute for a final explanation. Its purpose is to reduce uncertainty while preserving accuracy.

7. Response, Recovery, and Trust Repair

Crisis management has three connected phases: containment, recovery, and learning.

Containment

Containment limits harm. Actions may include pausing a campaign, disabling a defective feature, recalling a product, securing systems, separating affected data, correcting a misleading post, or providing urgent customer support. Managers should consider whether the organization’s own scheduled content is inappropriate while stakeholders are experiencing harm.

Recovery

Recovery restores functioning and relationships. It may include refunds, replacement, compensation, remediation, independent investigation, policy reform, employee support, and transparent progress reporting. Recovery should be proportionate to harm and accessible to affected people, not merely to those who complain publicly.

Trust repair

Trust repair is not achieved through apology alone. Research and practice generally suggest that credible repair combines acknowledgment, explanation, responsibility, corrective action, and evidence of changed behavior. The correct balance depends on the organization’s responsibility. A brand that caused the harm should not use a purely defensive strategy; a brand falsely accused may need a firmer evidential correction while still recognizing stakeholder concern.

Reputation recovery can be assessed through multiple indicators: repeat complaints, customer retention, employee confidence, stakeholder interviews, media framing, regulator feedback, service performance, and the completion of corrective actions. A return to pre-crisis mention volume does not necessarily mean that trust has been restored.

8. The Role of Technology and Artificial Intelligence

Technology can improve detection and coordination but also creates new risks. Social listening platforms can identify unusual patterns, classify topics, map influential accounts, and alert managers to cross-platform spread. Natural-language processing can summarize large datasets, but it may reproduce bias, misclassify irony, or privilege dominant language patterns.

Generative artificial intelligence can help draft message variants, translate approved information, simulate stakeholder questions, and search internal records. It should not independently determine whether an allegation is true, make legal admissions, disclose confidential information, or target individuals for punitive action. Human review is essential for high-impact decisions.

AI-related reputation risks include deepfakes, synthetic reviews, automated harassment, fabricated screenshots, impersonation, and rapid content generation. Organizations should maintain provenance controls, verification procedures, access restrictions, audit logs, and clear approval authority. They should also distinguish between legitimate criticism and coordinated manipulation without assuming that all unusual activity is inauthentic.

9. Ethics, Law, and Governance

Reputation management raises ethical questions about privacy, freedom of expression, fairness, transparency, and power. Organizations should not collect more personal information than necessary, expose complainants to retaliation, or use surveillance to identify critics merely because they are unfavorable. Monitoring should comply with applicable privacy and data-protection requirements.

Important legal and regulatory areas may include advertising and consumer-protection law, product safety, employment law, defamation, intellectual property, data protection, accessibility requirements, financial disclosure, and sector-specific reporting duties. Jurisdiction matters: a response strategy lawful in one country may violate another country’s rules.

Managers should distinguish criticism from unlawful content. Legitimate negative reviews and whistleblowing may be protected or socially valuable. Fraudulent impersonation, threats, non-consensual disclosure of private information, and knowingly false commercial claims may justify platform or legal intervention. The organization should document the basis for any removal request and apply policies consistently.

Ethical crisis communication requires proportionality. It is inappropriate to use vulnerable individuals as symbolic evidence, shift blame to frontline employees without evidence, or exploit a crisis to promote unrelated brand content. Transparency should be meaningful: it should help stakeholders make informed decisions rather than overwhelm them with selectively disclosed data.

10. Illustrative Case: The Northstar Nutrition Recall

This is a fictional illustrative case designed for analysis.

Background

Northstar Nutrition sells ready-to-drink protein beverages through supermarkets and direct-to-consumer channels. Its brand positioning emphasizes scientific quality, active lifestyles, and responsible sourcing.

Problem

A customer posts a video showing an unopened bottle with an unusual odor and reports illness. Within two hours, several users claim similar experiences. A fitness creator with a large following suggests that Northstar is hiding a contamination problem. The company’s monitoring team detects a sharp increase in mentions, but laboratory confirmation is not yet available.

Analysis

The matter begins as a complaint but quickly becomes a potential product-safety crisis. The evidence is incomplete, but the possible impact is severe. The creator’s claim may be inaccurate, yet dismissing it would be imprudent. The organization’s first responsibility is consumer safety, not preservation of campaign momentum.

The crisis team classifies the situation as high severity because of the combination of potential physical harm, rapid spread, high stakeholder exposure, and possible regulatory involvement. It opens a decision log, contacts public-health and regulatory advisers, preserves samples, identifies affected production lots, and pauses distribution of the relevant lots pending investigation.

Approach and solution

Northstar issues a holding statement acknowledging the reports, advising consumers not to consume specified lot numbers while verification occurs, and providing a dedicated support channel. It directly contacts retailers and customers who purchased the identified lots. The company does not repeat the unverified claim that contamination has been proven. It states what is known, what is being tested, and when the next update will occur.

After laboratory testing confirms a quality failure at one facility, Northstar announces a recall, refunds affected customers, publishes testing information in accessible language, and commissions an independent review of supplier and quality controls. The chief executive addresses employees first, explaining reporting obligations and support resources. A later report identifies a control failure and specifies deadlines, responsible executives, and verification procedures.

Outcome

The brand experiences short-term sales loss and substantial customer-service demand. However, independent verification, direct notification, transparent progress reporting, and operational reform reduce uncertainty. Recovery is uneven: customers who experienced illness remain less trusting than customers who were merely inconvenienced. Northstar therefore continues remediation beyond the point at which online attention declines.

Learning points

1. A complaint can become a crisis when potential harm and velocity increase.
2. Safety decisions should not wait for perfect information when precautionary action is justified.
3. A holding statement should communicate action and uncertainty accurately.
4. Stakeholder-specific communication is more effective than a single generic post.
5. Recovery depends on changed controls, not only on persuasive language.
6. Post-crisis analysis must examine how the failure occurred, how signals were handled, and whether governance encouraged escalation.

11. Post-Crisis Learning and Organizational Resilience

A post-crisis review should be conducted after immediate pressure has declined but while records and memories remain available. It should be psychologically safe enough to surface inconvenient evidence. The purpose is not simply to assign blame; it is to improve the system while retaining accountability for negligent or unethical conduct.

A rigorous review asks:

• What happened, and what was the timeline?
• Which signals were available before the event?
• Were signals ignored, misclassified, or not connected across departments?
• Were escalation thresholds clear?
• Did incentives discourage reporting or encourage concealment?
• Which stakeholders were harmed or excluded?
• Were messages accurate, timely, accessible, and culturally appropriate?
• Which corrective actions are complete, independently verified, and still effective?

Organizations may use a timeline analysis, root-cause analysis, stakeholder interviews, complaint coding, and comparison of planned versus actual decisions. Root-cause analysis should go beyond “human error” to examine training, workload, technology, authority, incentives, supplier controls, and organizational culture.

Resilience is the capacity to absorb disruption, adapt, and improve without normalizing preventable harm. It depends on scenario exercises, contact trees, pre-approved principles rather than rigid scripts, backup communication channels, clear decision rights, and regular review of monitoring coverage.

12. Best-Practice Protocol

A graduate-level manager should be able to translate principles into an operational protocol:

1. Detect: Monitor relevant channels and document the original evidence.
2. Verify: Check source credibility, corroboration, context, and potential manipulation.
3. Classify: Determine whether the matter is a complaint, misinformation, issue, or crisis.
4. Assess: Evaluate harm, probability, velocity, responsibility, stakeholder vulnerability, and legal exposure.
5. Escalate: Notify the appropriate owners using predefined thresholds.
6. Protect: Take proportionate action to reduce immediate harm.
7. Communicate: Acknowledge concern, state verified facts, explain action, and set an update time.
8. Coordinate: Align customer service, employees, executives, legal advisers, regulators, and partners.
9. Recover: Remediate harm, compensate where appropriate, and demonstrate operational correction.
10. Learn: Conduct an independent review, update controls, and test whether improvements work.

The protocol should be adapted to industry, geography, organizational size, and stakeholder expectations. A regulated health, financial, transport, or food organization will require more formal reporting and specialized expertise than a small lifestyle brand, but the underlying logic remains applicable.

Chapter Summary

Online reputation is a strategic, cumulative judgment shaped by stakeholder experience and digital visibility. Effective management begins with structured monitoring of reputation signals, not with reactive posting. Managers must distinguish ordinary complaints from systemic issues, inaccurate information from credible evidence, and online attention from genuine crisis severity.

Risk assessment can support prioritization when its assumptions and limitations are made explicit. Escalation requires defined decision rights, subject-matter expertise, documentation, and clear thresholds for safety, legal, privacy, discrimination, and operational events. Crisis communication should be accurate, empathetic, specific, stakeholder-appropriate, and connected to real protective action.

Containment limits immediate harm; recovery restores services and relationships; trust repair requires credible responsibility and observable change. Artificial intelligence can strengthen monitoring and drafting but cannot replace human judgment, ethical governance, or accountability. Finally, post-crisis learning converts disruption into organizational resilience by examining signals, decisions, systems, incentives, and outcomes.

Key Terms

Reputation: A cumulative stakeholder judgment about an organization or brand.

Brand identity: The meanings and values an organization seeks to express.

Brand image: Audience associations held at a particular time.

Reputation signal: Observable evidence of stakeholder evaluation or emerging risk.

Social listening: Systematic monitoring and interpretation of digital conversations.

Complaint: An expression of dissatisfaction about an experience, product, or policy.

Misinformation: Inaccurate information shared without demonstrable intent to deceive.

Disinformation: Deliberately false or manipulated information.

Issue: A developing matter that may affect legitimacy, relationships, or objectives.

Crisis: A high-impact situation requiring urgent, coordinated action.

Triage: Initial classification and prioritization of a signal or event.

Escalation: Referral of a matter to higher authority or specialized expertise.

Holding statement: A temporary, accurate communication issued while facts are developing.

Containment: Action that limits immediate harm or spread.

Recovery: Restoration of operations, relationships, and stakeholder confidence.

Trust repair: Deliberate restoration of confidence through responsibility and changed behavior.

Risk register: A structured record of threats, evidence, consequences, owners, and actions.

Decision log: A contemporaneous record of information, assumptions, decisions, and review points.

Reputation resilience: The capacity to absorb disruption, adapt, and improve.

Review Questions

Short Answer Questions

1. How does reputation differ from brand identity and brand image?
2. Why is mention volume alone an inadequate measure of crisis severity?
3. Define misinformation and disinformation.
4. What are the five essential elements of an initial holding statement?
5. Why should a decision log be maintained during a crisis?

Long Answer Questions

1. Explain how digital media environments change the strategic management of brand reputation.
2. Compare complaints, issues, misinformation, and crises, using examples to show why classification matters.
3. Discuss the roles and responsibilities required in an effective crisis-governance architecture.
4. Evaluate the relationship between crisis communication, operational correction, and trust repair.
5. Explain the ethical and regulatory challenges involved in monitoring critics and correcting misinformation.

Analytical / Application Questions

1. A brand discovers a rapidly spreading allegation that its product caused illness, but laboratory evidence is incomplete. Develop a classification, escalation, communication, and containment plan.
2. Using R = P × I, assess two risks: Risk A has probability 4 and impact 2; Risk B has probability 2 and impact 5 on a five-point scale. Explain why the numerical result should not be the only basis for prioritization.
3. Design a post-crisis review for an organization whose online attention has declined but whose repeat complaints remain high. Identify the evidence, stakeholders, questions, and corrective actions required.

Multiple Choice Questions with Answers

Question 1

Which statement best distinguishes reputation from brand identity?

A. Reputation is what the organization intends to communicate, while identity is what stakeholders believe.

B. Reputation is a cumulative stakeholder judgment, while identity is what the organization seeks to express.

C. Reputation exists only on social media, while identity exists only in advertising.

D. Reputation is a short-term campaign message, while identity is a customer complaint.

Correct Answer: B. Reputation is a cumulative stakeholder judgment, while identity is what the organization seeks to express.

Question 2

Why is a small number of credible safety allegations potentially more serious than thousands of low-value complaints?

A. Safety allegations always receive less media attention.

B. Complaint volume is unrelated to stakeholder evaluation.

C. Potential harm and evidence quality can matter more than raw volume.

D. Low-value complaints are legally prohibited.

Correct Answer: C. Potential harm and evidence quality can matter more than raw volume.

Question 3

Which situation is most clearly an issue rather than an isolated complaint?

A. One customer reports a delayed delivery.

B. A customer asks how to return a product.

C. A recurring pattern of accessibility complaints indicates a policy problem.

D. A user makes an unrelated comment on a brand post.

Correct Answer: C. A recurring pattern of accessibility complaints indicates a policy problem.

Question 4

What is the primary purpose of a holding statement?

A. To provide a complete final investigation before any action occurs.

B. To reduce uncertainty with verified information and explain immediate action.

C. To prevent all criticism from being published.

D. To replace operational remediation.

Correct Answer: B. To reduce uncertainty with verified information and explain immediate action.

Question 5

In the conceptual risk model R = P × I, what do P and I represent?

A. Publicity and identity

B. Probability and impact

C. Platform and influence

D. Performance and investment

Correct Answer: B. Probability and impact

Question 6

Which action best demonstrates containment during a potential product-safety crisis?

A. Launching an unrelated promotional campaign

B. Deleting all negative comments

C. Pausing distribution of the potentially affected product lot

D. Waiting for online attention to decline

Correct Answer: C. Pausing distribution of the potentially affected product lot

Question 7

Why should AI-generated sentiment analysis be reviewed by humans in consequential cases?

A. AI cannot process any written language.

B. Sentiment systems may misread context, sarcasm, bias, or cultural language.

C. Human review guarantees that every criticism is false.

D. Automated tools are legally required to publish all private messages.

Correct Answer: B. Sentiment systems may misread context, sarcasm, bias, or cultural language.

Question 8

Which practice is ethically inappropriate in reputation management?

A. Providing direct notification to affected customers

B. Maintaining a documented decision log

C. Using surveillance to identify critics solely because they are unfavorable

D. Reporting credible safety concerns to regulators

Correct Answer: C. Using surveillance to identify critics solely because they are unfavorable

Question 9

What is the principal purpose of a post-crisis review?

A. To ensure that no employee can raise concerns in the future

B. To identify system weaknesses and improve future resilience while retaining accountability

C. To replace all monitoring with public relations campaigns

D. To prove that online attention is the only crisis measure

Correct Answer: B. To identify system weaknesses and improve future resilience while retaining accountability

Question 10

Which combination is most likely to support trust repair?

A. Silence, denial, and deletion of criticism

B. A vague apology without operational change

C. Acknowledgment, responsibility where warranted, remediation, and evidence of changed behavior

D. A new advertising slogan unrelated to the event

Correct Answer: C. Acknowledgment, responsibility where warranted, remediation, and evidence of changed behavior