Sooner or later, every website goes down. A hosting provider has an outage. A plugin update breaks something. The domain expires because a card on file stopped working. A traffic spike overwhelms the server. An attacker gets in. It’s not a question of whether it will happen, but when.
I’ve handled a lot of these moments for clients, often late at night and often with an anxious business owner on the other end of the phone. Over time, I’ve learned that the difference between a ten-minute blip and a two-day disaster usually isn’t technical brilliance. It’s preparation and a calm, methodical process.
This is the plan I follow. You can use it for your own site, or adapt it for the sites you manage.
Before anything happens: preparation
Most of the work in incident response happens before the incident. If you prepare well, the actual outage becomes a checklist rather than a panic.
Know where everything is
Write down, somewhere you can reach even when your website is down, the essentials:
- Where the domain is registered and who has the login.
- Where DNS is managed. It isn’t always the same place as the registrar.
- Who hosts the site, how to reach their support and where their status page is.
- How to access the server: control panel, SFTP or SSH.
- Where backups are stored and how to restore them.
- Who to contact for each part of the system, including developers and agencies.
Keep this in a password manager or a secure document, not only on the website or in an email account hosted on the same domain. If your email runs on the same domain and the domain fails, you lose access to the very emails you’d need.
Set up monitoring
You want to know about downtime before your customers tell you. External uptime monitoring checks your site every minute or few minutes from outside your network and alerts you by email, SMS or a chat app when it fails.
Monitor more than the homepage. A homepage served from cache can look fine while the checkout, login or contact form is broken. Useful checks include:
- The homepage, checking for a specific word that should appear on the page.
- A key page that isn’t heavily cached.
- SSL certificate expiry.
- Domain expiry.
- For stores, the cart or checkout page.
Have tested backups
A backup you’ve never restored is a hope, not a plan. Make sure backups are stored off the server, cover both files and database, and that you know how long a restore takes.
Step 1: Confirm the outage
When an alert arrives or someone reports a problem, first confirm that the site is actually down and for whom.
- Open the site in a private browser window.
- Try from your phone using mobile data rather than your office Wi-Fi.
- Use an online “is it down” checker or your monitoring service’s details to see whether it fails from several locations.
If it works for some people and not others, you’re likely looking at a DNS issue, a regional network problem, a caching layer or a firewall blocking certain visitors. If it fails everywhere, the problem is closer to your server or domain.
Note the exact time the problem started and what error appears. A blank page, a browser connection error, a 500 error and a “this domain has expired” page all point in very different directions.
Step 2: Read the error
The type of error tells you where to look first.
Browser cannot connect at all
Messages like “This site can’t be reached” or “DNS address could not be found” usually mean a DNS or domain problem, or the server is completely offline. Check:
- Has the domain expired? Look it up in a WHOIS tool or log in to the registrar.
- Have DNS records changed? Someone may have edited them, or nameservers may have been changed.
- Is the server running? Check your host’s control panel and status page.
Connection timed out
The domain resolves, but the server isn’t responding. The server may be overloaded, offline or blocked by a firewall. Check the hosting status page and server resource usage.
SSL errors
Warnings about an invalid or expired certificate mean the SSL certificate needs renewing or is misconfigured. Visitors can’t easily get past these warnings, so treat them as an outage.
500 errors or “critical error”
The server is running, but WordPress or PHP is failing. Usually a plugin, theme, PHP configuration or code problem. Check error logs.
502, 503 or 504 errors
The web server can’t get a healthy response from PHP or an upstream service. Common causes include PHP-FPM crashing or running out of workers, traffic spikes, slow database queries or a CDN unable to reach your origin.
Database connection errors
WordPress can’t reach the database. Check credentials, whether the database server is running and whether resource limits have been hit.
Content looks wrong or redirects elsewhere
The site loads but shows spam, defacement or redirects to another site. Treat it as a security incident, not just downtime.
Step 3: Check what changed
Most outages follow a change. Ask:
- Were plugins, themes or WordPress core updated recently, automatically or manually?
- Did anyone deploy code or edit files?
- Did the host make changes, such as a PHP version upgrade or server migration?
- Did DNS or CDN settings change?
- Did a payment for hosting or the domain fail?
An activity log plugin and your host’s notifications are invaluable here. If an automatic update ran ten minutes before the outage, you’ve probably found your cause.
Step 4: Stabilise first, investigate second
When a business site is down, the priority is getting it back online. Deep investigation can come afterwards. Depending on the cause, quick stabilising actions include:
- Rolling back a recent update. Deactivate the plugin that was just updated, or restore its previous version.
- Disabling all plugins temporarily through WP-CLI or by renaming the plugins folder over SFTP, then reactivating essential ones.
- Switching PHP versions back if the host just changed them.
- Restoring a recent backup if the cause isn’t clear and the site must be online quickly. Remember you’ll lose changes made since the backup, such as new orders or comments.
- Enabling a maintenance page at the server or CDN level while you work, so visitors see a friendly message instead of an error.
- Scaling up resources temporarily if traffic overwhelmed the server.
Before making changes, take a quick snapshot or backup of the current state if you can. If the problem turns out to be a security incident, that snapshot is evidence.
Step 5: Communicate
People cope much better with downtime when they know what’s happening. Silence breeds frustration.
- Tell stakeholders quickly. A short message to the business owner or team: what’s happening, that you’re working on it and when you’ll update them next.
- Tell customers if the outage is long. A post on social media or a notice on a status page helps, especially for stores and service platforms.
- Give realistic updates. “We’re investigating and will update you within 30 minutes” is better than an optimistic promise you can’t keep.
If your email runs on the same domain and is affected, use phone calls, messaging apps or social channels instead.
Step 6: Recover and verify
Once the site is back, don’t just glance at the homepage and walk away. Verify properly:
- Test key pages, forms, login, search and, for stores, a full checkout.
- Check the admin dashboard works.
- Look at error logs for continuing warnings.
- Confirm scheduled tasks, emails and backups are working.
- Check that monitoring shows the site as up from multiple locations.
If you restored a backup, work out what data may have been lost, such as orders placed during the gap, and whether any can be recovered from payment gateway records or emails.
Step 7: Find the root cause
After the immediate crisis, find out why it happened. Read logs, review the timeline and identify the real cause, not just the symptom.
“The site was down because PHP ran out of memory” is a symptom. “A plugin update introduced a memory leak on product pages, and our server had no headroom” is closer to a cause. “We apply updates directly to production without testing” might be the root cause.
Step 8: Prevent it happening again
Write a short summary of the incident: what happened, when, how it was detected, what was done, how long it lasted and what will change. Then make the changes:
- Test updates on staging before production.
- Add monitoring for whatever failed silently.
- Increase server resources or add caching if capacity was the issue.
- Set domain and SSL auto-renewal, and check payment methods.
- Improve backups or document restore steps more clearly.
These short reviews are how a site becomes more reliable over time. Every incident should leave you better prepared for the next one.
A simple incident checklist
Keep this somewhere you can find it quickly:
- Confirm the outage from several locations.
- Note the time and the exact error.
- Check domain, DNS and hosting status.
- Check what changed recently.
- Take a snapshot of the current state if possible.
- Stabilise: roll back, disable, restore or scale.
- Communicate with stakeholders and customers.
- Verify everything works after recovery.
- Find the root cause.
- Write a short review and implement improvements.
Frequently asked questions
How quickly should I respond to downtime?
As quickly as your business needs. For a store, every minute matters. For a small brochure site, an hour may be acceptable. Decide in advance what response time you need, and set up monitoring and support to match.
Should I always restore a backup when the site breaks?
Not always. Restoring is reliable, but you lose recent changes and may not fix the underlying cause. If the cause is clear, like a just-updated plugin, rolling back that change is usually faster and loses nothing.
What’s the most common cause of downtime?
In my experience, plugin and theme updates applied without testing, hosting resource limits, expired domains or SSL certificates, and security incidents. Most are preventable with staging, monitoring and good renewal habits.
Do small sites need an incident plan?
Yes, even a simple one. Knowing where your domain, hosting and backups are, and having monitoring in place, turns a stressful outage into a manageable task.
Final thoughts
Downtime is inevitable, but chaos isn’t. Prepare by documenting where everything lives, setting up monitoring and testing backups. When something breaks, confirm it, read the error, check what changed, stabilise first and communicate clearly. Afterwards, find the real cause and fix it. Each incident handled this way makes your site, and your process, a little stronger.




