External Crisis Communication

Status pages use precise, present-tense language that communicates impact and current state without over-promising resolution timelines. Common phrases include "We are investigating an issue affecting…," "Our team has identified the root cause and is working on a fix," "Service has been restored and we are monitoring for stability," and "We will provide an update in 30 minutes." The writing should be factual and calm, avoiding jargon while being specific enough to help customers assess their own exposure.

Effective outage apologies follow a structure of acknowledgement, impact statement, current action, and next update time. Phrases such as "We sincerely apologise for the disruption this has caused," "We understand how critical this service is to your operations," and "We take full responsibility and are committed to preventing a recurrence" are standard in professional crisis communication. Avoid vague language like "we're sorry for any inconvenience" as it can feel dismissive; specificity demonstrates that the company understands the real impact.

A customer-facing post-incident report summarises what happened, why it happened, how long it lasted, what was affected, and what steps are being taken to prevent recurrence. Key vocabulary includes root cause, contributing factors, detection time, resolution time, impact scope, and remediation actions. Unlike an internal post-mortem, the external version avoids blame language, internal system names, and overly technical detail, focusing instead on customer impact and forward-looking commitments.

Standard severity vocabulary includes "partial outage" (some users or features affected), "full outage" (service completely unavailable), "degraded performance" (service available but slower than normal), "elevated error rates" (more requests failing than expected), and "intermittent issues" (problems occurring sporadically). Status pages typically use a traffic-light system: operational, degraded performance, partial outage, and major outage. Using consistent, pre-agreed terminology avoids confusion and helps customers self-triage.

Planned maintenance announcements typically include phrases like "We will be performing scheduled maintenance on [date] from [time] to [time] UTC," "During this window, [service] may be unavailable or experience degraded performance," "No action is required on your part," and "We apologise in advance for any inconvenience." Good announcements specify which services are affected, provide a precise maintenance window with timezone, and include a contact point for urgent questions.

Data breach notification letters must clearly state what data was accessed, when the breach occurred, when it was discovered, what steps the company has taken to contain it, and what customers should do to protect themselves. Regulatory requirements (such as GDPR's 72-hour notification rule) often dictate specific language obligations. Phrases include "We are writing to inform you of a security incident that may have affected your personal data" and "We have taken immediate steps to secure our systems and prevent further unauthorised access."

In technical communication, an "incident" is a broad term for any unplanned event that disrupts or degrades service, including security events, performance degradation, and partial failures. An "outage" specifically refers to a complete or near-complete loss of service availability for a defined user group. Companies typically use "incident" for internal tracking and reserve "outage" for customer-facing communications when the impact is severe and widespread, as "outage" carries stronger implied severity.

External crisis communication should use a calm, authoritative, and empathetic tone that avoids panic while taking the situation seriously. The register should be formal but not bureaucratic—clear sentences, active voice, and no unexplained technical acronyms. Teams are advised to write for a "non-technical executive stakeholder" as the mental model: someone who needs to understand business impact without needing to understand the technical details. Avoid minimising language such as "just a small issue" or "nothing to worry about."

Resolution language on status pages is typically brief and definitive: "This incident has been resolved," "Service has been fully restored as of [time] UTC," or "All systems are now operational." A good resolution update also references the incident start time, acknowledges the total duration, and links to or promises a post-incident report. This closure message is as important as the initial alert because it signals to customers that monitoring is active and that the team follows through on communication commitments.

PR crisis vocabulary for tech companies includes terms like holding statement (an initial brief statement buying time while facts are gathered), media briefing, spokesperson, off the record, on background, press release, official statement, and damage control. Internal teams also use terms like crisis playbook, communications lead, escalation path, and approval chain. The phrase "no comment" is generally discouraged in favour of a holding statement that demonstrates awareness and active response.