Client Website Down? The 15-Minute Agency Playbook

Q: What should I tell a client when their website is down?

Send an immediate acknowledgment: 'Hi [Name], we've detected an issue with [website] and are actively investigating. I'll update you within 15 minutes. No action needed from your side.' Then update every 15 minutes until resolved. After resolution, send a summary with cause, fix, and prevention steps.

Need to act right now?

🔍 Check if the site is actually down 📢 Create an emergency status page 🔒 Quick SSL certificate check 🌐 Verify DNS resolution

The first time a client's site went down on my watch, I spent 20 minutes checking the wrong things because I didn't have a checklist. Now I follow this exact playbook every time. It's called The 15-Minute Agency Incident Playbook, and it works whether this is a single freelance client or your 50th agency account.

In 80% of the agency outage cases we've seen, the root cause is one of 5 things: expired SSL, failed deployment, DNS change, hosting issue, or plugin conflict. Not exotic. Just systematic. This playbook helps you find which one, fast.

The 15-Minute Agency Incident Playbook

1 ⏱ Minute 0-2: CONFIRM

Verify the Outage Before You React

Don't panic. Don't start restarting things. First, confirm what's actually broken:

Check the site from a different network (mobile data, not your office WiFi)
Use an external uptime checker to test from multiple locations
Is the whole site down, or just one page/feature?
Is it a timeout, an error code (500, 503), or a DNS failure?
Can you reach the server at all? (SSH, hosting panel)

Why this matters: If the site is up from external locations but your client can't reach it, it's a DNS cache or local ISP issue—not your emergency. If it's a 503, the server is alive but overwhelmed. If it's a DNS failure, restarting the server won't help. Each symptom points to a different cause.

2 ⏱ Minute 2-5: COMMUNICATE

Send the First Message to Your Client

Rule of crisis communication: the first message doesn't need answers. It needs acknowledgment. "We're aware and investigating" buys you 30 minutes of trust. Silence buys you nothing.

The First Client Message:

"Hi [Name], we've detected an issue with [website] and are actively investigating. I'll update you within 15 minutes. No action needed from your side."

Send this within 5 minutes. Every time. Even at 2am. Even if you have no idea what's wrong yet. The client needs to know you're on it.

If the outage affects end users, create an emergency status page in 60 seconds—free, no signup. It gives you a public URL to share and shows you're handling the situation professionally. For more on writing clear updates, see our guide on writing status page updates.

3 ⏱ Minute 5-10: DIAGNOSE

Run the Quick Diagnosis Checklist

Work through this list in order. In most agency scenarios, you'll find the cause within the first four checks.

Quick Diagnosis Checklist

1. Is the server responding? — Try SSH or the hosting panel. If the server itself is unreachable, it's a hosting-level issue.
2. Is DNS resolving? — Use the DNS checker. If DNS is wrong, the domain might have expired or nameservers changed.
3. Is SSL valid? — Use the SSL checker. An expired certificate shows a browser warning that looks like "the site is down" to most clients.
4. Was there a recent deployment? — Check your CI/CD logs, Git history, or hosting deploy log. A bad deploy is the most common cause of sudden outages.
5. Is the database reachable? — If the site loads partially or shows generic errors, the database connection might be the issue.
6. Plugin or theme conflict? — Especially for WordPress/CMS sites. Check if anything was auto-updated recently.

4 ⏱ Minute 10-15: ACT

Fix Based on What You Found

You've identified the likely cause. Now act on it:

→ Bad deployment: Rollback immediately. Fix forward later, when the site is back up.
→ Expired SSL: Renew the certificate. If using Let's Encrypt, run the renewal command. Most hosting panels have a one-click renew.
→ DNS issue: Check the registrar. If the domain expired, renew it (can take up to 48 hours to propagate). If nameservers changed, revert them.
→ Hosting outage: Check the provider's status page. If it's on their end, you can't fix it. Communicate clearly to the client that it's a hosting-level issue and provide their ETA.
→ Server overload: Restart the web server. If it keeps crashing, check error logs for the root cause (memory leak, traffic spike, runaway process).
→ Plugin conflict: Disable the last updated plugin via FTP/SSH if you can't access the CMS admin.

If you can't fix it in 15 minutes, that's okay. The goal is to have a diagnosis and a plan. Escalate to the hosting provider, the developer who made the last change, or the plugin vendor. What matters is that you know what's wrong and can communicate that clearly.

5 ⏱ Minute 15+: UPDATE

Keep the Client Updated Every 15 Minutes

Even if nothing has changed, update the client every 15 minutes. "Still investigating, we've narrowed it down to X" is infinitely better than silence.

Update your status page with the current state
Message the client with what you know and what you're doing
Document every step you take (you'll need this for the post-mortem)
If it's a hosting issue, share the provider's status page link with the client

Once the site is back up, verify from multiple locations before declaring it resolved.

6 ⏱ Within 24 hours: POST-MORTEM

Send a Professional Incident Report

This is what separates agencies that lose clients from agencies that build trust. Within 24 hours, send a brief incident report covering: what happened, when, what you did, and how you'll prevent it next time.

The Resolution Message:

"[Website] is back online as of [time]. The issue was [brief cause]. I'll send a full incident report tomorrow with steps we're taking to prevent this from happening again."

The full incident report doesn't need to be long. One page is enough:

Timeline: when the outage started, when you were notified, when it was resolved
Root cause: what specifically went wrong
Impact: how long was the site down, was data affected
Resolution: what you did to fix it
Prevention: what changes you're making to prevent recurrence

Need a template? Use our post-mortem template.

What If You Knew Before the Client Did?

This playbook is your fire extinguisher. But you shouldn't need to fight fires if you have a smoke detector.

Imagine this scenario instead: your monitoring alerts you at 2:17 AM that a client site returned a 503. You check from your phone, confirm it's a hosting issue, and send the client a message before they even know there's a problem. By 2:45 AM the hosting provider resolves it. The client wakes up to your message saying "there was a brief outage overnight, we caught it immediately, here's what happened." That's the difference between reactive and proactive.

What proactive monitoring gives you:

Instant alerts when any client site goes down—before the client notices
SSL and domain expiry warnings so certificates never expire on your watch
Status pages per client so they can check status themselves instead of calling you
Client logins (from Agency Pro) so clients access their own dashboard on your branded domain

The full setup guide: Website monitoring for agencies — the business playbook.

Frequently Asked Questions

How fast should an agency respond when a client website goes down?

Acknowledge within 5 minutes, even if you don't know the cause yet. The first message doesn't need answers—it needs acknowledgment. "We've detected an issue and are investigating" buys you 30 minutes of trust. Silence buys you nothing.

What are the most common causes of client website downtime?

In about 80% of agency outage cases, the root cause is one of five things: expired SSL certificate, failed deployment, DNS change or propagation issue, hosting provider outage, or plugin/theme conflict. Not exotic problems—just systematic ones that a checklist can catch quickly.

What should I tell a client when their website is down?

Send an immediate acknowledgment: "Hi [Name], we've detected an issue with [website] and are actively investigating. I'll update you within 15 minutes. No action needed from your side." Then update every 15 minutes until resolved. After resolution, send a summary with cause, fix, and prevention steps.

How can I find out about client downtime before the client does?

Set up uptime monitoring that checks each client site every 1-3 minutes from multiple locations. Tools like PerkyDash send instant alerts via email, Slack, or Discord when a site goes down. From Agency Pro plans, clients get their own login to check status themselves instead of calling you.

This Playbook Is Your Fire Extinguisher. PerkyDash Is Your Smoke Detector.

Monitor all your client sites—and from Agency Pro, give them their own login so they check status themselves instead of calling you. SSL expiry alerts, DNS monitoring, instant notifications, branded status pages. Free 14-day trial.

Start Free Trial See Agency Plans

Related Guides

Pillar Guide