LOOP is built on modern, resilient cloud infrastructure, using Azure, Cloudflare CDN, and best-practice deployment processes designed to minimise downtime and reduce risk. We isolate services, use caching, scaling and continuously monitor sites to make outages extremely rare.
But... no cloud service can ever guarantee 100% uptime - you'll have seen the headlines from the large AWS, Azure and Cloudflare outages in 2025. We were fortunately only impacted by one of these, but it's important for you to be prepared.
Most outages are quick to resolve and only cause minimal disruption (minutes) - we typically say:
- 99% of issues are self-healing as sites will trigger a restart
- ... of the remaining 1%, the majority of those are quick to debug and resolve with simple troubleshooting steps (on our side)
- ... which leaves us with a small 0.001% chance of wider disruption (where outages might be hours rather than minutes)
The short version - outages cause disruption, but with a little prep you can turn them into opportunities.
Typical emergency timeline
- With 50+ brands on LOOP, we can quickly see if it's an wider issue that's affecting multiple clients
- We have Statuscake alerts that ping all sites every few minutes > if a site goes down then we typically spot it before you :-)
- If it's only affecting one client,
- We quickly jump in to the hosting portal and CMS to trace the cause and attempt troubleshooting steps (republishing pages, clearing caches, restarting the site)
- If we can't get the site back up and running in under 10 mins, then we'll reach out to you
- If it's a larger issue across multiple clients
- We escalate to suppliers
- We check across sites to see any patterns / issues
- We look at DownDetector / Reddit / X to determine if this is a wider issue
- If it's down for more than 10 minutes or so we send an email to clients that are impacted (to your chosen contacts)
- From there, we will keep you posted with regular updates and feedback as needed.
Typical LOOP packages include support in UK office hours, with best-efforts support out-of-hours. With any site outages, we get pinged of these on the company Slack messaging channels so endeavour to look and resolve these as quickly as possible, subject to availability.
Checklist to have ready
- Keep a log of direct booking links to your ticketing provider that you can point customers so that they can still book straight from emails / socials
- Have an email campaign / discount ready to send and promote on socials in prolonged outages
- Check common status pages first (Azure, Cloudflare, Umbraco Cloud)
- Save our emergency@semantic.co.uk email address to easily reach us > the phones get busy of course, but rest assured that we'll keep you updated throughout.
Making the most of the opportunity
- Use the outage as a reason for emailing your subscriber list
- Maximise engagement on socials
- Send additional comms once it's resolved to maximise reach + engagement
After it's resolved
- Keep an eye out for messages from us with any more follow up
- Review your comms and effectiveness of those
- Check key website pages for any leftover glitches (e.g. homepage, news, events, tickets, contact)
We are always working to evolve LOOP infrastructure to deliver the most resilient and reliable attraction websites possible.
The October Azure outage was the first supplier outage (in over 3 years) that took all of the LOOP sites offline for several hours, yet the clients who activated campaigns, emails and social posts turned it into an opportunity. Some even saw a boost in sales by using the moment as an extra touch point with their audiences.
So in summary, we'll work tirelessly to provide the most flexible, resilient cloud infrastructure setup for attractions, but when the worst does happen, try to make the best of it and use it as an opportunity.
