Personalisation Starts at the Car Park: What JLL Data Shows
74% of people prefer brands that recognise them, says JLL. How car parks can welcome returning visitors, reward loyalty and extend your brand.
6 days ago
Just after 7am on Tuesday (20 October), one of Google Cloud’s three London zones went dark. Several of the UK’s best-known parking apps were offline for the morning. Hozah’s car parks kept running, with no transactions lost.
Here’s what happened, how our systems responded, and what we’re improving next.
What happened in the Google Cloud London outage?
Google splits its London data centre into three zones: A, B and C. On the morning of Tuesday 20 October, zone B failed. For Hozah, that took out 10 of our 33 servers and 6 of our 15 database copies in one go.
That’s roughly a third of our infrastructure, gone without warning, but because we built our systems to run across all three zones, most of it fixed itself.
- Hozah Permits moved its main database to a healthy zone in 80 seconds.
- ANPR, Enforcement, Hozah Account and Hozah Console rebuilt themselves in under half an hour, with no manual intervention.
- Hozah car parks continued to function: the cameras kept reading plates, pay machines went to manual, and SSV tablets recorded 1,434 entries to be sent once we were fully back online.
For our drivers and landowners, their parking experience continued as normal, with little to no on-site disruption. Our systems are built to handle moments like this, so the outage had minimal impact on the ground.
How our engineers recovered Parking without losing data
One system didn’t recover on its own (the Parking database behind AutoPay). Google’s auto-move got stuck for an hour and then failed. The quick fix would have been to restore the backup from the night before, but that would have lost four hours of data. We weren’t going to do that.
Instead, our engineers switched the Parking database over to a live copy in a healthy zone, so no data was lost. Pay-on-foot, SSV and vehicle lookups were back online by 09:55, while Google was still in the midst of its own outage.
A big thank you to our engineering team, who made the right call under pressure and moved fast to put it right. Their quick thinking is a big part of why our drivers and landowners barely noticed the disruption.
A tough morning across the industry
There were plenty of bigger names that had a much tougher time. The BBC reported that bank customers were having problems, and some of the most well-known parking apps were down as late as 10:32.
Outages like this can hit anyone, and Tuesday was a hard morning for many teams. But it shows why resilience must be designed in from the start, not added later.
That’s how we build our systems at Hozah. If a cloud provider goes down, a car park can’t just stop functioning. People still need to park and pay, and it’s our clients’ names on the gate. On Tuesday, our car parks were still reading plates, the pay machines were on manual, and no transactions were lost.
What we’re improving next
Every incident teaches us something, and we’re already acting on Tuesday’s lessons:
- Making the Parking database move itself, just as Hozah Permits and our other systems already do, so it no longer depends on Google’s automatic move.
- Improving how the SSV tablets recover so that they reconnect and send saved entries more smoothly next time.
When something goes wrong in the wider cloud, our job is to make sure the people using our car parks barely notice. On Tuesday, that’s exactly what we did.
Is your car park ready for the next outage?
Outages like Tuesday’s don’t come with any warning, and the next one could last longer. If you’ve got a car park, then it’s worth asking yourself what would have happened on your site that day. Would people have been able to pay? Would you have lost transactions/revenue? With Hozah, the answer is built in rather than bolted on.
Book a 20-minute call with us, and we’ll explain in detail how our platform keeps your car parks going, even when the cloud goes down.
