Insights / Digital Foundation
Your website is down. What should you check first?
Before you update, restore or rebuild anything, find out which part actually failed and who looks after it.
The site won’t load. Someone on your team noticed. Or worse, a client did.
The instinct is to do something. Update the plugins. Restore last night’s backup. Call whoever built it and ask them to fix it now.
Hold on. A website depends on several systems, often run by different people: your own connection, the domain, DNS, the security certificate, the hosting and the site itself. Any one of them can make the site look down. Change things before you know which one failed and you can turn a short outage into a long one, or wipe out the evidence of what went wrong.
The question isn’t how to get the site back fastest. It’s what actually broke, and who is responsible for it.
Check in this order
Each step rules something out. Most of them take a few minutes and don’t change anything.
Write down what you see.
Note the exact error message, the time, which pages and which devices. Take a screenshot. “This site can’t be reached”, a browser security warning, a server error and a page with a broken layout point to different causes.
Check from somewhere else.
Try a phone on mobile data instead of the office Wi-Fi, or ask someone outside the business. An online “is it down for everyone” checker can help. If the site loads elsewhere, the problem is probably local: your network, firewall, browser or device.
Check the domain hasn’t expired.
Log in to the registrar and look at the renewal date. An expired domain can take down your website and your email at the same time. If you don’t know who holds the registrar account, that’s worth fixing as soon as the site is back.
Check DNS.
If the domain is active, ask whether anyone changed DNS recently: a host move, an email setup, a new tool that needed a record. One wrong record can send visitors to the wrong place. DNS changes take time to spread, so the site may work for some people and not others.
Check the certificate.
If visitors see a security warning instead of the site, the certificate may have expired or may not match the address. The site can be running fine underneath while browsers warn people away.
Check the hosting.
Look at your host’s status page or contact support. Servers fail, shared resources run out, accounts get suspended over a missed payment, and unusual traffic, including deliberate attacks, can overwhelm a server. Your host can see logs you can’t.
Check what changed on the site.
If the server responds but the site shows an error or a blank page, look at recent changes: an update, a new plugin or extension, a theme edit, new code. Conflicts between components are a common cause on platforms that use add-ons. Reverse the specific change, ideally on a staging copy first, rather than updating everything else.
Change one thing at a time, and only once you know why.
Broken parts aren’t a broken site
Missing images, a dead link or a form that won’t send are real problems, but they don’t mean the whole site is down. They usually come from a moved file, a renamed page or a third-party service that changed. Fix them specifically.
The same goes for security scans. A scan can flag known problems. It can’t find every fault, and a clean result doesn’t mean the site is healthy.
Don’t restore blind
A backup is a safety net, not a first move. Restoring replaces what’s there now. You can lose form entries, orders and content added since the backup, and if you don’t know the cause, you may restore the same fault.
Before restoring, know what failed, take a copy of the current state, pick the right backup and, where you can, test the restore on staging. Updating everything at once has the same problem: if the site comes back, you won’t know what fixed it, and if it gets worse, you won’t know what broke it.
Know who to call before it happens
Most of the stress in an outage comes from not knowing who owns which piece.
| If the problem is | Usually handled by |
|---|---|
| Your connection or network | Your IT support or internet provider |
| Domain expiry | Whoever holds the registrar account |
| DNS | Whoever manages DNS: registrar, host or a separate service |
| Certificate | Your host or website provider |
| Hosting | Your hosting provider |
| Errors after a change | Your developer or maintenance provider |
If several of those answers are “not sure”, that’s the gap to close. Hosting and ongoing care should come with a named person, a way to reach them and a plan for recovery. Here’s what that care should cover after launch.
An outage is stressful. Guessing makes it worse.
Find out what broke.
Call the right person.
Then change one thing at a time.