Roughly 96% of brand crises now spread internationally within 24 hours, and the window between a crisis starting and the public narrative getting set runs about two hours. Brands that respond inside that first two-hour window see 61% better sentiment recovery than ones that respond slower. Neither of those numbers leaves room for a team to figure out its plan in real time. This guide covers what to build before a crisis hits, the actual response sequence once one starts, and how to handle recovery once the immediate story has passed.
- Responding within two hours produces 61% better sentiment recovery than a slower response, and 96% of crises spread internationally within 24 hours, leaving almost no margin for a team without a plan already in place.
- The 15-20-60-90 timeline gives a concrete response cadence: acknowledge within 15 minutes, share a preliminary statement within 20, a detailed update within 60, and a full briefing within 90.
- A holding statement beats silence every time. Confirming awareness and that the team is investigating, even without full facts, is far better than saying nothing while a story fills the vacuum.
- Consistency across channels matters as much as speed. A social response that contradicts the official press statement amplifies distrust faster than a slow response ever could.
- Recovery is a distinct phase, not an afterthought. Sentiment tracking after the acute crisis passes shows whether trust is actually rebuilding or just quietly fading from view.
What to build before a crisis hits
A workable crisis plan needs five components in place before anything actually goes wrong, and building them mid-crisis is exactly how a slow response happens. Active monitoring across every relevant channel comes first, since a team cannot respond to something it hasn't noticed yet. Secured account access matters more than it sounds, since a compromised or single-person-controlled account becomes its own crisis during exactly the moment speed matters most. Defined roles with a clear escalation path, who drafts, who approves, who publishes, removes the friction of figuring that out under pressure. Pre-approved message templates for common scenarios, a product issue, an offensive post, a data concern, cut real minutes off the response time. And a publishing location, usually a newsroom page or pinned post, ready to go live fast gives the team somewhere to point people once a statement exists.
The five Cs framework, communication, consistency, compassion, control, and commitment, is worth keeping visible during plan-building specifically, since it's easy to draft a technically accurate statement that still reads as cold or defensive if compassion isn't deliberately built into the approval process alongside legal accuracy.
The first 90 minutes
The 15-20-60-90 rule gives a concrete cadence for the acute phase of a crisis. Acknowledge the situation publicly within 15 minutes of confirming it's real, even with a single sentence. Share a preliminary statement with the known facts within 20 minutes, being explicit about what is still unconfirmed. Provide a more detailed update with actual next steps within 60 minutes. Deliver a comprehensive briefing, the fullest account available at that point, within 90 minutes.
A brief holding statement that says the team is aware and investigating beats silence in every documented case, even when the full picture isn't ready. Silence reads as either not caring or not knowing, and both interpretations spread faster than an honest "we're looking into this" ever does.
Keeping every channel aligned
Social posts, press releases, email updates, and any leadership statement each reach a different audience segment, and a contradiction between any two of them does more damage than a slow response on its own. A social team posting a warm, apologetic tone while a legal-approved press statement reads defensive and hedged creates exactly the kind of inconsistency that amplifies distrust rather than calming it.
The practical fix is a single shared message document that every channel pulls from, updated in real time as facts change, rather than separate teams drafting independently and reconciling after the fact. Press coverage tracking alongside social monitoring during an active crisis shows whether the story reporters are running matches what the brand is actually saying, which is exactly the kind of gap that widens fast if nobody's watching both channels at once.
What not to do
A few specific missteps show up across nearly every documented brand crisis case study. Deleting the original post or comment, if it hasn't already gone viral in screenshot form, almost always backfires, since it reads as an attempt to hide rather than address the issue, and someone has usually already captured it regardless. Going quiet after the first acknowledgment, rather than following the 15-20-60-90 cadence through to a full update, lets the story keep moving without the brand's version in it. Responding defensively or arguing with individual critics in public comments tends to escalate rather than resolve, even when the brand's underlying point is technically correct.
A quick-service restaurant chain that faced a viral kitchen hygiene video is a useful contrast case: because roles were already assigned, PR issued a statement, customer service responded directly to concerned followers, and legal cleared the messaging within hours rather than days, which kept the story from running unanswered long enough to fully set the narrative.
Recovery and rebuilding trust
Every crisis follows a rough three-stage arc: detection, response, and recovery, and recovery is where most teams stop actively measuring even though it's arguably the most important phase to track. A sentiment dip during the acute crisis is expected and usually recoverable. Whether it actually recovers, and how long that takes, is only visible through continued monitoring after the news cycle has moved on and attention has dropped off.
Context-aware sentiment tracking through the weeks following a crisis shows whether trust is genuinely rebuilding or just fading from visible conversation while quietly persisting underneath. Those are very different outcomes that look identical on a volume chart alone, and only a sentiment trend line, tracked well past the point most teams stop paying close attention, actually distinguishes between them.
Catching it before it fully breaks
The single highest-leverage move in this entire process happens before the crisis is even confirmed: catching the early signal. A sharp, sudden negative sentiment shift frequently precedes the volume spike that makes a story impossible to miss, and that early window, sometimes hours before a story goes fully viral, is exactly when a calm, prepared response does the most good. Forum and Reddit monitoring specifically catches a meaningful share of early crisis signals, since a product complaint or safety concern often surfaces in detailed community discussion well before it becomes a viral post on a faster-moving platform. PR teams managing this cadence should build it directly into existing coverage review workflows rather than treating crisis monitoring as a separate, occasionally-checked system.
Common mistakes
- Waiting for complete facts before saying anything. A holding statement acknowledging awareness and investigation, sent within the first hour, consistently outperforms silence while facts get fully confirmed.
- Letting different channels draft independently. Social, press, and leadership statements need to pull from one shared, continuously updated source rather than reconciling contradictions after they've already gone public.
- Treating the crisis as over once volume drops. A quiet mention count doesn't mean sentiment has actually recovered, and stopping measurement too early misses whether trust rebuilt or just went underground.
- Building the plan during the crisis instead of before it. Roles, templates, and escalation paths decided under active pressure take far longer than the same decisions made calmly in advance.
- Ignoring early, lower-volume signals. A detailed forum thread or a handful of pointed comments often precedes the viral spike by hours, and that window is the best chance to respond before the story is fully set.
Frequently asked questions
Should a brand ever stay completely silent during a crisis?
Rarely, and almost never for long. A brief holding statement, even without full facts, consistently outperforms silence, since silence tends to read as either not caring or not knowing what happened, and both interpretations let the story fill in its own narrative.
Should the original post or comment that started the crisis be deleted?
Generally no.
How long does full sentiment recovery typically take after a crisis?
Varies significantly by severity and response quality, but a well-handled crisis with fast acknowledgment and consistent follow-through tends to show meaningful sentiment recovery within weeks rather than months, based on documented case studies. A poorly handled one, especially with inconsistent messaging across channels or a long initial silence, can extend recovery well past that window, and some brands never fully return to their pre-crisis sentiment baseline at all, particularly when the underlying issue points to a genuine, unaddressed pattern rather than a single isolated incident. That distinction, an isolated event versus a pattern, is often the single biggest factor separating a brand that recovers cleanly within a month from one still fielding the same complaint a year later, and it is usually visible in the mention data itself well before a formal post-mortem confirms it, provided someone is actually tracking sentiment closely enough during the weeks that follow to notice the difference.
Does a small business need a formal crisis plan, or is that overkill?
A scaled-down version is still worth having. Even a single-page document naming who responds, where the pre-approved holding statement lives, and which channels to monitor closes most of the gap a full enterprise crisis plan covers, without needing the same organizational overhead.
A brand crisis on social media moves faster than most organizations are built to respond, which is exactly why the plan has to exist before the first sign of trouble shows up. Build the five components, roles, templates, monitoring, secured accounts, a publishing location, this week if they aren't already in place, and run the 15-20-60-90 timeline as a drill at least once so the first real crisis isn't also the first time anyone's actually followed it.
Catch the early signal before the story sets
Mentient monitors news, social, reviews, and forums in real time with sentiment scored automatically, so a shift shows up before the volume spike does. Start free and see your own monitoring coverage today.
Start free trial
