It's a Wednesday afternoon. A customer at one of your restaurant locations records a 14-second video of something that went wrong. Maybe an employee said something racist. Maybe a hair ended up in someone's pasta. Maybe a manager yelled at a kid. The customer posts the clip to TikTok. By the time you finish your meeting at 5pm, the video has 2.3 million views. By 7pm, your brand's name is trending. By 9pm, three news outlets have reached out for comment. By the time you wake up Thursday morning, ChatGPT has already started referencing the incident when anyone asks about your restaurant.

If you're a marketing director, founder, or comms lead at a U.S. business with any kind of public profile, this scenario keeps you up at night. And honestly, it should. The crises that took weeks to escalate ten years ago now reach saturation in hours.

We're Crepo Media, and we work with companies navigating exactly these moments. By the time you finish reading, you'll know what makes 2026 crises different, why the first few hours matter more than ever, and what a real crisis communication plan should include before you ever need one.

What Makes Modern Crises Different

Four things separate today's crisis environment from what it looked like even five years ago.

Velocity

By October 2025, social media had reached 5.66 billion active user identities worldwide, effectively placing corporate reputation in front of most of the planet. In 2026, that exposure has turned reputation into a real-time risk, with crises capable of igniting and escalating across global platforms in a matter of hours. A complaint on X gets screenshotted to LinkedIn within 30 minutes, lands on Reddit by hour two, and shows up on local TV news by the end of the day.

AI amplification

This is the new wrinkle most companies haven't grasped yet. AI search means that a viral negative thread on Reddit doesn't just reach Reddit users. It gets cited in ChatGPT responses, Perplexity answers, and Google AI Overviews for months afterward. A bad week on social media used to fade. Now AI systems keep referencing it long after the initial news cycle ends.

Public scrutiny

Audiences expect transparency that wasn't standard a decade ago. They want named accountability, real apologies, and visible follow-through. Corporate-speak that worked in 2014 gets ratioed in 2026.

Permanent digital memory

Screenshots live forever. Wayback Machine archives press statements. Twitter threads get embedded in news articles that sit on the first page of Google for years. The crisis doesn't end when the news cycle moves on. It becomes part of how people search you, find you, and decide whether to trust you.

The First 24 Hours Matter Most

Here's the part most American business owners get wrong. They wait too long.

The old playbook said you have 24 hours to respond. The current reality is different. In 2020, a 24-hour response was acceptable. In 2026, if you haven't responded within 2–4 hours, the narrative is already set.

The 2-hour rule: in 2026, the window between a crisis starting and the public narrative being set is roughly two hours.

Two hours. By the time you've called a meeting to talk about it, the story has often already calcified in the public mind.

A few things worth doing in those first hours.

Acknowledge

A brief holding statement saying "we're aware of the situation and looking into it" beats silence every time. You should acknowledge a crisis publicly within the first few hours, even if you don't have all the facts yet. A brief holding statement that confirms you're aware and investigating is far better than silence.

Verify

Before you make any specific claims, get the facts. If an employee did something wrong, find out who, when, and what actually happened. If the allegation is false, get the evidence in hand before you push back publicly.

Communicate

Once you have a verified position, get it out across every channel where the conversation is happening. Your website. Your social accounts. Direct outreach to journalists who've started reporting.

Monitor

Set up real-time tracking of mentions, sentiment, and where the story is spreading. You can't respond to what you can't see.

Centralize messaging

Every employee, every spokesperson, every executive needs to be saying the same thing. Conflicting statements from different people at the same company turn a manageable crisis into a permanent one.

Mistakes That Make Crises Worse

The way companies blow up their own crisis response usually follows recognizable patterns.

Silence

The most common mistake. Founders think saying nothing buys them time. It doesn't. It cedes the narrative entirely to critics. Silence is never a safe option. A pre-prepared plan with designated spokespersons, approved messaging templates, and a multidisciplinary response team turns a chaotic situation into a managed process.

Emotional reactions

When the founder personally responds to a viral tweet at midnight with sarcasm or anger, it almost always makes things worse. Screenshots of that response become the new story.

Inconsistent messaging

Your CMO says one thing on LinkedIn. Your customer service team says another on X. The CEO gives a contradictory quote to a reporter. Now there are three versions of your response circulating, and audiences pick whichever one fits the narrative they already believe.

Delayed statements

The "we'll respond when we have all the facts" approach worked when news cycles ran on 24-hour newspapers. It doesn't work in a 2-hour viral cycle.

Defensive corporate-speak

The infamous United Airlines moment in 2017 became a textbook case. United's initial response was criticized for its lack of empathy, referring to the incident as "re-accommodating" passengers. Following public backlash, the CEO issued a more sincere apology, and the airline implemented policy changes to prevent similar incidents. "Re-accommodating" instead of "we dragged a paying customer off our plane." That choice of language cost United more than the actual incident did.

A Fictional Example

Let's walk through a hypothetical. This is a made-up scenario for illustration, not a real incident.

Picture a regional restaurant chain called something like Honeycomb Diner, with 28 locations across the Carolinas. A customer at the Charleston location records a 22-second video of a manager appearing to lose his temper with a young server who'd dropped a tray. The clip catches the manager using language that sounds dismissive of the server's age and accent.

The video goes up on TikTok at 2pm on a Saturday. By 5pm it has 800,000 views. By 8pm it has 4 million views. The hashtag #HoneycombDiner starts trending. Local news in Charleston, Greenville, and Raleigh pick up the story.

The company's first response is a generic statement posted to Facebook at 11pm Saturday. It says "We take all employee concerns seriously and are investigating the matter internally." No names. No specifics. No acknowledgment of what was visible in the video.

The internet eats them alive. By Sunday morning, the original video has been stitched 14,000 times. New accusations from former employees start surfacing. Yelp is flooded with one-star reviews across all 28 locations. Two corporate partners pause their relationships.

Now picture the alternative version. Same incident. Different response.

The company's social listening tool catches the video at 3pm Saturday, an hour after posting. By 4pm, a holding statement goes up across all channels: "We've seen the video circulating about our Charleston location. The behavior shown is not acceptable. We're investigating right now and will have a fuller response within hours." By 7pm, the CEO posts a video. She names what she saw in the clip. She announces the manager has been suspended pending investigation. She commits to a published timeline for next steps. She apologizes directly to the server.

The crisis still happens. The story still spreads. But by Sunday morning, the narrative has shifted from "Honeycomb is a toxic workplace" to "Honeycomb is handling this the right way." Local TV runs follow-up stories that include the CEO's response. Yelp reviews stay manageable. Corporate partners hold steady.

Same incident. Different preparation. Different outcome.

How AI Search Changed Reputation Risks

This is the piece most crisis planning hasn't caught up on yet.

When somebody Googles your company in 2026, they don't just see the news articles. They see AI-generated summaries. Total search impressions increased approximately 49% since the AI Overviews launch, with click-throughs declining nearly 30% and Google maintaining over 90% market share.

Those AI summaries pull from whatever the AI considers authoritative. If a crisis got significant news coverage, that coverage becomes part of how AI describes your company. Forever, or close to it. The crisis doesn't fade when the trending hashtag stops trending. It becomes baked into your digital reputation in a way that's harder to dislodge than old-school news cycles ever were.

A potential customer in 2027 asks ChatGPT "is Honeycomb Diner a good place to eat?" The AI pulls from news coverage and review sentiment. If 2026's crisis was poorly handled, the AI's answer reflects that for years. If it was handled well, the response coverage gets weighted alongside the original incident, and the summary reads differently.

Indexed controversies last longer than they used to. A story that ran in three local outlets used to disappear after a month. Now the URLs stay indexed, the AI keeps surfacing them, and a customer doing research five years later still encounters the original story.

This raises the stakes on response quality. A bad crisis response in 2026 doesn't just cost you next quarter's sales. It costs you years of digital reputation that compounds against you.

Why Prepared Brands Recover Faster

Companies that recover from crises share traits. They didn't figure these out during the crisis. They had them in place beforehand.

Protocols

There's a written document somebody can pull up that says what happens in the first hour, the first day, and the first week of a crisis. Who calls whom. Who signs off on statements. Who talks to journalists.

Spokesperson training

The person who'll appear on camera knows how to handle hostile questions, stay on message without sounding robotic, and apologize without sounding scripted. This kind of training takes months to develop. You can't acquire it during the crisis.

Media preparedness

The brand already has relationships with key journalists in their industry and region. When the crisis hits, those journalists are willing to take a call from the company, hear the company's side, and report fairly. Cold relationships during a crisis almost never work.

Documentation

Every claim about operations, sourcing, certifications, or compliance can be backed by paperwork that's accessible within minutes. If your sourcing, claims, approvals, and customer care systems are weak, viral misinformation will expose those weaknesses. If your documentation is strong and your communication is disciplined, you have a real chance to contain the damage.

Internal alignment

Employees know what's happening. They're not finding out about the crisis from Twitter while at work. They have guidance on what to say if customers ask them about it directly.

What a Crisis Communication Plan Should Include

If you're building one from scratch, a working framework includes the following elements.

  • A defined crisis tier system. Not every problem is a Level 10 crisis. Sort potential issues into tiers (a customer complaint vs. a product recall vs. a leadership scandal) and have different response protocols for each.
  • A response team roster. Who's on the crisis team? Names, roles, phone numbers, backup contacts. This list should fit on one page and live somewhere everyone can find at 2am on a Sunday.
  • Pre-approved spokespersons. Usually the CEO, sometimes a designated communications lead, sometimes legal counsel. Each person has clearly defined situations in which they speak.
  • Holding statement templates. Pre-drafted holding statements for the most likely crisis scenarios, ready to be customized and pushed live within 30 minutes.
  • Channel protocols. Which channels you communicate through, in what order, with what cadence. Your website should always be the source of truth where your full statement lives. Social channels link back to it.
  • Monitoring infrastructure. Real-time tools tracking brand mentions, sentiment shifts, and the spread of viral content.
  • Stakeholder communication chains. Customers, employees, investors, partners, regulators. Each has different information needs in a crisis. The plan should specify who reaches each group, when, and with what message.
  • Recovery protocol. The plan shouldn't end with the immediate response. It should outline how the brand rebuilds trust over the weeks and months after the initial event.
  • Legal review processes. When does the legal team get pulled in? Which statements need legal sign-off? How do you balance the comms team's urgency with legal's caution? Disagreements here lose hours you don't have.

How Crepo Media Supports Crisis Response

At Crepo Media, the crisis work we do for clients tends to fall into three phases.

Before the crisis

We help companies build their crisis communication infrastructure. The plan, the templates, the protocols, the spokesperson training, the monitoring setup. This is the work that pays off when something actually goes wrong. Most clients come to us after a near-miss, when they realized they weren't ready.

During the crisis

We help draft holding statements, coordinate messaging across channels, manage media relations, and provide a senior comms voice during the most pressured hours. Having outside counsel during a crisis matters because internal teams are often too emotionally invested or politically constrained to make the right calls.

After the crisis

This is where most companies underinvest. The recovery work — including reputation rebuilding through earned media, content strategy that reshapes the digital narrative, and longer-term work on owned media and AI search optimization — determines whether the crisis becomes a footnote or a permanent stain.

If you want to see how we think about crisis communication in the context of broader PR work, our Digital PR services page covers our broader approach. We're not the only firm doing this kind of work, and depending on your situation, you might need a specialist crisis firm or an in-house communications hire instead of an outside agency. The thing that matters most is having someone, somewhere, who's thought about this before the crisis arrives.

The Honest Take

Most American companies don't have a real crisis communication plan. They have a Google Doc somebody made in 2022 that nobody's looked at since. They have an assumption that "we'll figure it out when it happens." They have a CEO who thinks crisis comms is overhyped.

Then it happens. And they spend the next six months wishing they'd built the infrastructure when things were calm.

The cost of preparation is small. A few weeks of focused work. A retainer with a comms partner. Some training sessions for the leadership team. Maybe a monitoring tool subscription.

The cost of being unprepared in 2026 is enormous. Lost customers. Lost partnerships. Lost talent. A digital reputation that follows the brand for years through AI search results that keep surfacing the worst version of what happened.

Reputation damage spreads in hours now. Recovery takes years. The brands that come out the other side of a crisis with their credibility intact aren't lucky. They were ready before they needed to be. Build the plan now, before you need it. Future-you, in the middle of whatever crisis is coming, will be grateful.