Your product will have outages. Servers fail. Databases corrupt. Deployment goes wrong. Third-party services you depend on break. It's not whether outages will happen. It's whether they're tomorrow or six months from now.
How you communicate during outages says more about your brand than your normal-day marketing does. Customers expect polish on good days; bad days reveal the underlying character. The brands that handle outages well end up with stronger customer relationships than they had before. The brands that handle outages badly damage relationships that were fine going in.
Here's the framework for outage communications and what separates good from bad responses.
What customers actually need during an outage
Three things, in this order:
1. Acknowledgment. They want to know the company knows. Silence during an outage is the worst signal. It suggests either ignorance or indifference. Customers will tolerate problems if they see the company is aware; they won't tolerate feeling ignored.
2. Information. What's happening. What's affected. What's not affected. How long it's expected to take. This isn't legal-CYA disclosure; it's the customer trying to make decisions about their own work.
3. Path to resolution. What the team is doing. What customers should do (if anything). When the next update will come. This converts anxiety into manageable waiting.
These three needs are the same across every customer and every outage. Brands that meet them well, in brand voice, build trust during outages. Brands that miss them. Even with technically excellent fixes. Damage trust.
The communication channels and what each is for
Status page. The canonical source of truth. Updated as the situation evolves. Customers check here when they want details. Should be styled and voiced like your brand. Most status pages look generic and feel disconnected from the rest of the brand.
Email to affected customers. Sent within the first hour for significant outages. Should be short, in voice, with the three elements above. Direct from a real person when possible (founder, head of engineering, customer success lead).
Public social posts. Used to acknowledge to the broader audience and to direct people to the status page. Posted in the brand's normal voice (don't switch to formal corporate-speak for outages).
In-app notifications. If users are in the product when an outage occurs, show what's affected and what isn't. Helpful banners, not modals that block work. Brand-voiced, not "Error Code 503" generic.
The voice principles for outage communications
Outage communications shouldn't sound different from your normal brand voice. The most common mistake: companies suddenly become formal, legalistic, and corporate during outages, abandoning the brand voice that built relationships in the first place.
Five voice principles that work:
1. Plain language about what's broken. "The dashboard isn't loading" beats "Customers are experiencing service degradation in the analytics module." The plain version is shorter, clearer, and more believable.
2. Specificity about scope. "About 30% of users are affected" beats "Some users may be experiencing issues." Customers can plan around specificity; vagueness creates anxiety.
3. Honesty about cause when you know it. "Our database failed over and the read replicas didn't promote correctly" beats "An infrastructure issue." Technical customers respect technical honesty; non-technical customers parse around the technical detail without being bothered by it.
4. Real-person ownership. Signed by a specific human, not by "The Team" or "Customer Success." Real names create accountability and humanize the response.
5. No empty apology language. "We sincerely apologize for any inconvenience this may have caused" is legal boilerplate. "This sucks and we know it. We're working on it" is brand voice. Pick the second.
What separates good outage responses from bad ones
Compare two responses to the same situation: a 3-hour outage affecting most users.
Bad response:
"We are aware of an issue affecting some users and are working to restore service. We apologize for any inconvenience this may have caused. We will provide updates as more information becomes available."
Why it's bad: vague ("some users"), passive ("an issue"), boilerplate apology, no timeline, no specifics, no human signature. The customer reading this learns nothing actionable.
Good response:
"Quick update: our main database failed over at 11:47am Eastern and the read replicas didn't promote correctly. Roughly 80% of users can't access the dashboard right now. Customer support and core APIs are unaffected. We're rolling back to a previous deployment and expect to be back up within 90 minutes. Next update in 30 minutes., [name], co-founder"
Why it's good: specific about what's broken (database failover), specific about scope (80%), specific about what's working (support, APIs), specific timeline (90 min, next update in 30), human signature.
Both responses take about the same time to write. The good one builds trust. The bad one damages it.
The post-outage communication
Once the outage is resolved, send a follow-up:
- Confirmation the issue is resolved
- Brief explanation of what happened (root cause if known)
- What's being done to prevent recurrence
- Specific apology for the disruption, with concrete details about the impact
For significant outages, a public postmortem within 48 hours. Engineering details, timeline, what failed, what worked, what's changing. This signals technical competence and transparency. Both brand-positive.
Many companies skip the postmortem because writing it surfaces uncomfortable details. The companies that publish postmortems consistently build reputation as competent and trustworthy operators, especially among technical buyers who read them.
The compounding brand effect
Outages, handled well, are brand-strengthening moments. The customer who watched you handle a 3-hour outage with clear communication, specific timelines, and accountable ownership trusts you more after the outage than before. They've seen you when the polish is off and the brand was still intact.
Outages handled badly are brand-damaging moments that linger. The customer who watched corporate-speak and vague updates during a problem remembers it for years. They tell their network. The damage compounds.
This is why outage communication is one of the highest-leverage brand surfaces, despite getting almost no attention in brand strategy work. Most brands invest heavily in good-day marketing and treat outage comms as legal-team territory. The brands that treat outage comms as brand surface area. Designed, voiced, and rehearsed before the first outage. Are dramatically ahead.
The preparation that matters
Before your first significant outage, prepare:
- A status page styled in your brand
- Template outage emails in brand voice (fill-in-the-blank format)
- A communication checklist for the on-call team
- A clear decision tree about who communicates what to whom
- A drafted post-mortem template
Two hours of preparation. Used dozens of times across the company's life. The version of your brand that customers see during outages is the version that decides whether they stay or leave. Make sure that version reflects the brand you want to be on your good days too.
Your brand kit, ready in 10 minutes.
Five quick taps. Free preview before you pay.
Start building free →