Imagine you get home after a long shift at the office, and all you want to do is eat and watch that TV show everyone has been harping on about.
You get settled on your sofa, turn on the TV, open a streaming application—and to your surprise, you see an error message stating that the service is unavailable.
You might be upset for a few minutes and keep trying to refresh the app, but eventually you give up and watch terrestrial TV instead (the horror).
If that happened a few nights in a row, you might start looking for alternative streaming options. And there are plenty of other options for you to turn to in the streaming service space.
Now, imagine if it was your bank that had availability issues.
Your bank being available for you to use is essential for your day to day activities—being able to pay for lunch, travel and a round at the pub, plus getting paid your salary and being able to take care of bills.
A large UK bank had a few major issues in 2018. A system upgrade went wrong, and it cost them approximately £330m, and 16,000 lost customers.
What’s the difference between 99.9% vs. 99.95% availability?
To get from 99.9% to 99.95% allowed downtime, or unavailability, needs to drop:
Daily: from 1m 26.4s to 43.2s
Weekly: from 10m to 5m
Monthly: from 43m to 21m
Yearly: from 8h 45m to 4h 22m
Those may sound like small time differences, but even a tiny amount of downtime can have serious financial effects. A major software company experience just two hours of downtime in 2018, but it brought about $8 million dollars in revenue loss.
Why is availability so relevant to Onfido?
Availability influences all parts of our operations, from customer support to finances to customer onboarding.
If our service is unavailable, it means people lose access to services, transportation, entertainment and financial services. Any loss in service reflects directly on our clients, who are in effect trusting us with their brand.
This trust, if maintained, is why people recommend and renew with us. It is our greatest asset.
How do we ensure availability?
We have a special team who work around the clock in order to maximise our availability.
There’s a primary and secondary person who work on-call and are notified by either an automated or manual alarm in a system called ‘pagerduty’ if something affects our service availability.
If an issue is identified, they react quickly to resolve the issue. They use several different tools, from monitors, dashboards, logs, among others, to investigate the cause of the alarms and solve the problems, keeping customers in the loop via our status page.
We are also continuously working on improving our processes, infrastructure, deployments, etc. to ensure Onfido services are always available for our customers.