When your website goes down, enquiries stop, orders fail and customers wonder whether you're still in business. Outages rarely come from nowhere: most have a handful of predictable causes, and most can be prevented or shortened with basic precautions. Here's what typically takes websites offline, and what to do about each.
How much downtime adds up
| Uptime | Downtime per month (approx.) | Downtime per year (approx.) |
|---|---|---|
| 99% | 7 hours 18 minutes | 3.65 days |
| 99.9% | 43 minutes | 8.8 hours |
| 99.99% | 4 minutes | 53 minutes |
The common causes
1. Hosting problems
Overloaded shared servers, hardware failures and provider outages. Prevent it with reputable hosting suited to your site, a CDN in front, and monitoring. Our guide on choosing web hosting helps.
2. Failed updates and code changes
A plugin update conflicts with your theme, or a code deployment contains a bug. Prevent it by testing updates on staging first, backing up before changes and having a quick rollback plan.
3. Expired domains
If a domain renewal fails, the website and email stop working. Prevent it with auto-renew, current payment details, an up-to-date contact email at the registrar and multi-year registration. Check yours with our domain registration lookup.
4. Expired SSL certificates
Browsers show security warnings that scare visitors away. Prevent it with automatic certificate renewal and monitoring of expiry dates.
5. DNS mistakes
A wrong DNS record during a migration or email setup can take a site offline. Prevent it by documenting DNS, making changes carefully, lowering TTLs before migrations and checking records after changes.
6. Traffic spikes
A sale, campaign or press mention can overwhelm under-powered hosting. Prevent it with caching, a CDN and hosting that scales, and by warning your developer before big campaigns.
7. Security incidents
Hacks, malware and denial-of-service attacks. Prevent it with updates, strong passwords and two-factor authentication, a firewall and regular backups. See our website security basics.
8. Third-party failures
Payment gateways, booking widgets, fonts or scripts from other services can fail and break pages. Prevent it by loading third-party scripts carefully and having fallbacks for critical features.
9. Database problems
Corrupted tables, full storage or overloaded databases. Prevent it with monitoring, regular maintenance and backups.
10. Resource limits
Running out of disk space, memory or bandwidth on your hosting plan. Prevent it by monitoring usage and upgrading before you hit limits.
Prevention at a glance
| Cause | Key prevention |
|---|---|
| Hosting | Quality hosting, CDN, monitoring |
| Updates | Staging tests, backups, rollback plan |
| Domain | Auto-renew, current details |
| SSL | Automatic renewal, expiry alerts |
| DNS | Documentation, careful changes |
| Traffic spikes | Caching, scalable hosting |
| Security | Updates, 2FA, firewall |
| Third parties | Careful loading, fallbacks |
Detect problems fast
Uptime monitoring checks your site around the clock and alerts you the moment it stops responding. Combine it with:
- Error monitoring for application problems
- SSL and domain expiry alerts
- Regular checks of critical journeys such as forms and checkout
What to do when your site goes down
- Confirm it's down for everyone, not just you
- Check recent changes: updates, deployments, DNS edits
- Check your host's status page
- Check domain and SSL expiry
- Restore from backup if a change broke the site and can't be fixed quickly
- Communicate: a status message on social media or email if the outage is long
- Review afterwards: find the cause and prevent it happening again
Good backups make step 5 possible; see our website backup strategy.
Setting up uptime monitoring
- Choose a monitoring service that checks from several locations.
- Monitor your key pages, not just the homepage: checkout, booking, contact.
- Check frequently enough to catch problems quickly.
- Send alerts to people who can act, by email, text or team chat.
- Monitor SSL and domain expiry dates as well.
- Add a public or internal status page for longer outages.
A post-incident review template
After any significant outage, spend 30 minutes on a short review. The goal is learning, not blame.
| Question | Answer |
|---|---|
| What happened? | A short description of the outage |
| When did it start and end? | Times, and how long customers were affected |
| How was it detected? | Monitoring, customer report, team member |
| What was the cause? | The root cause, not just the symptom |
| What fixed it? | The steps taken |
| What will prevent it happening again? | Specific actions, owners and dates |
Teams that review incidents honestly have fewer repeat outages, because each one leads to a lasting fix.
How we help
Our uptime and performance monitoring service watches your site around the clock and alerts our team immediately, while hosting management and updates and security patching prevent the most common causes.
The short version
Websites usually go down because of hosting problems, failed updates, expired domains or SSL, DNS mistakes, traffic spikes, security incidents or third-party failures. Use quality hosting, test changes, automate renewals, document DNS, plan for spikes, keep software secure, monitor around the clock and keep tested backups ready.
