Marketing

Crisis Communication: 10 Essential Checks for Better Updates

Crisis Communication – Crisis Communication: What to Say When Your Site Goes Down

In this guide: Crisis communication when your website goes down: why communication matters, first message templates, updating customers, post-incident reports, and…

Every website goes down at some point — hosting outages, failed updates, DDoS attacks, expired domains, or hacks. Customers need accurate information about the impact and the next useful step. Avoid assuming that any particular message guarantees forgiveness or loyalty.

Responsible communication can reduce uncertainty, but its effect on trust depends on the incident and the response.

Why communication matters

When your site is down, customers:

  • Don’t know if it’s their connection or your site.
  • Worry about their data, orders, or payments.
  • Flood your support channels with the same question.
  • Assume the worst if you say nothing.

Clear communication:

  • Reduces support load — a useful update can answer repeated questions; the effect depends on the incident.
  • Protects your reputation — transparency signals competence.
  • Keeps customers loyal — they feel respected and informed.
  • Gives your team focus — fewer interruptions while fixing the problem.

First message templates

Your first message should go out quickly — ideally according to the severity, confirmed impact and the response plan. It doesn’t need all the answers.

Template 1 — Initial acknowledgment (social media or status page)

⚠️ We’re aware that [website/service] is currently unavailable for some users. Our team is investigating and working on a fix. We’ll share an update within [30 minutes/1 hour]. Thank you for your patience.

Template 2 — Online store

Our online store is temporarily unavailable. We are checking the affected functions and the status of existing orders. We will share verified information through [status channel]. We’re working to restore access and will update you by [time]. For urgent requests, contact us at [email/WhatsApp/phone].

Template 3 — Possible security incident (be careful and factual)

We’ve detected an issue affecting our website and have taken it offline as a precaution while we investigate. We’ll share more information as soon as we can. If you have questions, contact [email].

Rules for first messages:

  • Acknowledge the problem.
  • Say you’re working on it.
  • Give a time for the next update.
  • Provide an alternative contact channel.
  • Don’t speculate about causes or promise exact fix times you can’t guarantee.
  • Don’t blame your host or vendors publicly.

Updating customers

Where to communicate:

  • Status page (a separate service or subdomain not hosted on the same server) — see Best Uptime Monitoring Tools for tools that include status pages.
  • Social media — pin the update to your profile.
  • Email — for significant outages affecting customers’ accounts or orders.
  • WhatsApp/SMS — for critical services.
  • Support auto-replies — update chat and email auto-responses with the latest status.

Update template:

Update [time]: We’ve identified the cause of the outage ([brief, non-technical explanation, e.g., “a server issue at our hosting provider”]) and are working on a fix. Some pages may load slowly as service is restored. Next update by [time].

Resolution template:

✅ Resolved: [Website] is back online and working normally. The issue was caused by [brief cause]. We’re sorry for the disruption and are taking steps to prevent it from happening again. If you notice any problems, please contact [email].

Stick to your update schedule — even if the update is “still working on it.”

Post-incident report

For significant outages, publish or send a short post-incident summary within a few days:

  1. What happened: timeline from detection to resolution.
  2. Impact: who was affected and how (without exposing sensitive details).
  3. Root cause: in plain language.
  4. What we did: immediate fixes.
  5. What we’re changing: concrete prevention steps (e.g., staging for updates, better monitoring, backup improvements, hosting upgrades).
  6. Apology and appreciation.

Internally, run a blameless review: focus on process improvements, not individual blame.

If personal data was involved, follow applicable legal notification requirements and get professional advice.

Rebuilding trust

  • Follow through on prevention steps you announced.
  • Compensate when appropriate: service credits, discounts, or extended subscriptions for paying customers.
  • Thank your community for patience.
  • Show improvements: share when you’ve implemented changes.
  • Prepare for next time: write templates now, set up a status page, and assign who communicates during incidents.

Prevention helps too: monitoring, backups, staging, and security reduce how often you’ll need these templates. See What Is Uptime? and Online Reputation Management.

Further reading: Google SRE incident response guidance

Key Takeaways: Crisis Communication

  • Communicate verified impact and useful next steps without making unsupported assurances.
  • Publish a timely factual acknowledgment through the approved process.
  • Update on a schedule through a status page and social media.
  • Publish a clear post-incident report with prevention steps.
  • Prepare templates and roles before the next incident.

1. Assign communication responsibility before an incident

Decide who coordinates the response, who investigates the technical problem and who approves public updates. A small business may combine roles, but the responsibilities should still be explicit. Otherwise, several people can publish conflicting explanations while nobody owns the next scheduled update.

Google’s SRE incident response guidance distinguishes coordination, operations and communication roles. Use that principle proportionally rather than copying a large-company hierarchy into a tiny team. The practical objective is an accountable route for facts, decisions and messages.

Keep a fallback owner and handover route. A plan that depends on one person being online at all times is fragile. Record how a replacement can find the current incident state without relying on private messages or an unavailable administrator.

2. Define the confirmed communication scope

Start with what is known about the affected service and user journey. A failed checkout is not necessarily a complete website outage. A monitoring alert from one location does not establish that every customer is affected. State the observed scope and keep uncertain areas explicitly under investigation.

Separate the time of detection from the confirmed start of impact. Those timestamps may differ. Use a clear time zone and avoid filling a gap in the timeline with an assumed beginning. Preserve internal evidence so the public summary can be corrected if later investigation changes the assessment.

Our uptime monitoring guide explains why an alert and a complete diagnosis are different things. Good communication makes uncertainty visible without presenting guesses as established facts.

3. Write an acknowledgment without unsupported assurances

A first message should identify the affected function, the current action and where the next update will appear. It does not need a speculative root cause or an invented restoration time. Avoid statements about order integrity, payment safety or personal data unless the responsible team has verified the relevant facts.

A hypothetical acknowledgment could say: We are investigating reports that some visitors cannot complete checkout. The team is assessing the affected functions. Please use our status page for verified updates; our next progress update is planned for the stated time. This is a writing example, not a message actually sent.

Only publish operational instructions that have been approved. Telling everyone to retry payment can create confusion or duplicate attempts when the transaction state is unclear. Route affected users through an appropriate support process and keep individual order details out of public discussion.

4. Use a communication channel that remains accessible

Consider which parts of the system share dependencies. A status page on the same failed infrastructure may also be unavailable. A different hostname alone does not prove independence. Review hosting, DNS, authentication and publishing access as part of the channel plan.

Choose a primary source of incident information and make other updates point to it where suitable. This reduces conflicting versions without requiring identical wording everywhere. Record the approved accounts and publishing responsibilities before a disruption rather than improvising access while the site is down.

Atlassian’s communication guidance provides a useful framework for audiences, channels and update cadence. Your own plan should reflect the channels you can actually operate, not every option mentioned in a generic checklist.

5. Keep internal and public communication distinct

Technical investigation may involve logs, hypotheses and sensitive access details. A public update needs verified customer impact and useful guidance, not a raw copy of the incident chat. Do not expose tokens, internal addresses, customer records or exploitable details while trying to demonstrate transparency.

Maintain an internal issue record with evidence, decisions and ownership. Label hypotheses clearly so an unconfirmed theory does not become a public accusation. A vendor status notice can be cited accurately without claiming it proves the cause of every symptom your customers report.

Use a concise approval process that suits the severity. Avoid unnecessary bottlenecks, but ensure that factual and sensitive claims receive the appropriate review. Communication should support response work rather than pressure investigators into announcing certainty they do not have.

6. Set an update cadence you can honour

Choose the next progress time based on the incident and the available team. This is an update commitment, not a restoration guarantee. If the investigation is still underway, report the verified state and explain when another update is expected rather than inventing progress to fill the interval.

Keep timestamps and time zones clear. If several channels are used, record which messages were actually published and which remain drafts. A prepared caption or scheduled item is not the same as a visible public update, especially when a platform or integration is itself affected.

A hypothetical progress note might explain that investigation continues, the affected function remains unavailable and the next review is scheduled. Do not copy a resolved template merely because the team has applied a potential fix. Verification belongs between mitigation and the final message.

7. Verify recovery before resolving communication

Define the customer journeys needed to support a recovery claim. A homepage loading does not establish that login, payments, forms or notifications work. Test the affected functions through an appropriate controlled process and record the result and remaining limitations.

Our mobile testing checklist and WordPress SMTP guide provide complementary checks for presentation and notification paths. Use them when relevant rather than asserting that one positive status proves every layer recovered.

Distinguish restored, monitoring and fully resolved according to the actual response process. If a mitigation is temporary or some users remain affected, say so. A communication plan should make partial recovery understandable without overstating what the team has verified.

8. Publish a factual post-incident summary

When appropriate, explain the impact, confirmed timeline, mitigation and next actions. Separate the known cause from contributing conditions and unresolved questions. A useful summary need not expose sensitive technical material or pretend that every incident has a single simple root cause.

Assign owners and review dates to announced improvements. A promise to prevent recurrence completely is usually stronger than the evidence supports. Describe the practical change and its intended purpose instead, such as improving an alert route or testing a specific recovery journey.

Use the Google SRE postmortem reference for a learning-oriented approach. Keep individual blame and unverified accusations out of the report. The objective is an accurate account and better operational decisions, not a promotional story about a fictional flawless recovery.

9. Handle individual cases through suitable support

Public communication cannot resolve every affected order or account. Give people a monitored route for their specific concern and collect only the details needed for that process. Do not ask customers to post payment information or personal records in comments under an incident update.

Compensation and service credits depend on the applicable agreement and authorized business decision. Do not automatically promise refunds, discounts or extended subscriptions from a generic template. Confirm the actual policy and ensure that the support team can apply it consistently.

If an incident raises legal notification questions, involve appropriate qualified review for the actual facts and jurisdictions. This guide does not establish a universal deadline or certify that a particular message satisfies every obligation. Avoid using a website outage notice as a substitute for a required specialist process.

10. Rehearse communication without sending false alerts

Practice with clearly labelled fictional scenarios in a controlled internal environment. Do not publish a pretend live outage to customers simply to test the template. The rehearsal should verify ownership, access, approval and the ability to distinguish known facts from assumptions.

Include a handover between two people. Can the replacement identify the current status, next update time and unresolved decisions? If not, improve the record. Our project workflow guide can help organize responsibilities without prescribing one particular platform.

Review the plan after real incidents and material system changes. A new hosting arrangement, support channel or external integration can change the dependencies. Keep the communication process aligned with the service the business actually operates.

Relevant monitoring and website support

Promotional recommendation: KAWEDMA’s monitoring offering is an option to evaluate for the observation layer. Confirm current checks, alert channels and responsibilities directly. Monitoring can identify symptoms; it does not guarantee prevention, establish a root cause or write an accurate customer notice automatically.

KAWEDMA also offers website development and hosting through its main services site. Ask for a written support and recovery scope where appropriate. This promotional recommendation is not a claim that a plan provides round-the-clock incident response or guarantees uninterrupted service.

Communication facts
Communication facts. Original explanatory diagram.
Communication verification
Communication verification. Original explanatory diagram.
Communication review
Communication review. Original explanatory diagram.

A fictional communication handover exercise

Consider a small store that receives reports of failed checkout attempts. The team has not established the cause or the final state of every payment attempt. This is an educational scenario, not a report of an actual incident. The first task is to identify what has been observed and which questions require further investigation.

The coordinator records the affected function, the evidence available and the people responsible for technical review and public updates. The communications owner drafts a short acknowledgment that describes the investigation without saying all orders are safe. A separate support route handles individual cases without requesting private payment details in public comments.

Before publication, a reviewer checks every factual clause. Does the message distinguish a report from a confirmed scope? Is the next update time realistic? Is the alternative contact route actually monitored? Does the wording imply a restoration deadline that nobody approved? These questions improve the message without requiring the technical team to invent certainty.

The rehearsal then introduces a handover. A second colleague takes over the communications role and must find the current state, approved channel and next deadline from the record. If they need to reconstruct the incident from scattered private messages, the record needs improvement. The purpose is continuity, not testing whether somebody can memorize the whole event.

When a potential mitigation becomes available, the team records it as a change awaiting verification. The exercise does not mark the incident resolved until the agreed customer journeys are checked. Any remaining limits stay visible. A page loading successfully is not enough evidence to declare that every payment, account and email function has recovered.

Finally, the team reviews the draft summary and action register. Improvements receive owners and review dates, while unresolved questions remain unresolved. The exercise ends with a record of what the process demonstrated and what it did not test. No customer notices are sent, no refund is promised and no fictional result is published as a real operational success.

Repeat this communication rehearsal when the support team, hosting arrangement or critical integrations change. Keep the scenario clearly labelled so a practice document cannot be mistaken for a live outage announcement.

Z
zasma171

Writes for Hostreta about the topics covered in this article.

Comments (0)

Your email address will not be published. Required fields are marked *

Ad
Affiliate links
Tools we recommend
Protect Your Website with Automated Daily Backups
$2.09
Get it
the easiest way to manage all of your social media accounts
$29
Get it
Starter Website Get online, properly
$249
Get it
google workspace Business Starter
$3.75
Get it
Nord Vpn Cybersecurity. Built for Everyday Use
$8.99
Get it
More hosting options · affiliate
AI SiteBuilder
View plans
AI Application Hosting
View plans
Shared Hosting
View plans
Reseller Hosting
View plans
Managed WordPress · Rocket.net
View plans
Unmanaged VPS Hosting
View plans
Managed VPS Hosting
View plans
Virtual Dedicated Servers
View plans

We may earn a commission on purchases made through these links, at no extra cost to you.

Ad
Ad

$ top hosting picks

affiliate
★★★★★

KAWEDMA Website Services

See optionsCheck current plans and terms
  • Explore website setup services
  • Choose a scope that fits your project
  • Confirm features with KAWEDMA
Explore KAWEDMA →
affiliate link · we may earn a commission
★★★★★

Managed WordPress Hosting

See plansCurrent terms on provider website
  • Managed hosting for WordPress
  • Powered by Rocket.net
  • Compare current plan specifications
View WordPress plans →
affiliate link · we may earn a commission
★★★★★

AI SiteBuilder

See plansCheck current offer and renewal terms
  • Explore AI-assisted website creation
  • Review included tools and features
  • Choose a plan for your project
Explore SiteBuilder →
affiliate link · we may earn a commission

Never miss a new article

Join thousands of developers and builders getting a weekly digest of the best tech articles.