Whatever Happened to Five 9s Reliability?

Whatever happened to five 9s reliability? That’s the standard AT&T used to architect to for landline phone service.? In everyday language that translates to allowing for no more than five minutes of downtime per year.? At the time, a host of practices, policies and procedures were in place to support that standard, and it worked!? Yet, a few days ago, AT&T had an outage that lasted for 11 hours. For those who have had the privilege of using a landline phone, can you ever think of a time when your landline phone was down for 11 hours? I think not.? In fact, in that era, landline phones had been known to work even when other utilities were down due to natural disasters.? That’s how “bulletproof” the service was.? Meanwhile, AT&T’s outage was due to a bad software update, something that would have been unheard of in the five 9s era.

The thing is that five 9s reliability wasn’t just the standard that AT&T snapped to. Other “always on” services also aspired to that standard and designed for it. That includes internet service providers (ISPs), data centers, big tech websites, etc.? From the perspective of a customer, that gave peace of mind that those services would always be there when they needed it.?

Yet, it seems like especially since the pandemic we’ve entered a new era where service meltdowns are commonplace and have almost come to be accepted as the norm. ?Every other day we’re reading about a meltdown of some sort.? That includes meltdowns by phone service providers, airlines, air traffic control systems, and more. The bottom line is that a lot of these meltdowns are completely preventable.? It’s a matter of having a mindset to give customers what they want before they know they need it and then planning for it.? In this case what customers want is “always on” service, and they don’t even realize it or fully realize it until the service isn’t there.

Some would argue that it is cost prohibitive to build for five 9s reliability, but they should consider the cost of damage control and making it right for customers after one of these meltdowns occurs.? Would the $750M+ Southwest Airlines spent to recover from the 2022 holiday meltdown have been better spent architecting for five 9s ahead of time?? I bet it would, and that doesn’t even take into account lost shareholder value or the reputational hit that they took.? That doesn’t take into what it will take to regain lost customer trust.

Arguably five 9s isn’t the standard for every service or product, but every business should be well aware of what that standard is for their service or product then build to that and accept nothing less. The trust that will earn you with customers is almost priceless!