Service Availability: Plan for the Inevitable

I remember the telephones when I was growing up—the analog telephones, first with the rotary dial, next with the touch tone. One cord went into the phone jack in the wall and the other cord was connected to the handset. The cool thing about those phones was unless there was some major city-wide problem that took down the telephone lines en masse, that phone was always usable and always worked. Even at companies, there was rarely any worry about downtime or network problems with them. There was no IT staff that was just dedicated to networked phone lines. No one had to worry about getting in touch with us. Those phones always worked. They were always “on.”

Now services are digital. And in the cloud. On servers. Which means that we can reach more people and we can accomplish more and we have more flexibility with use, but we can also have exponentially more problems. And for a company, service must always be “on.” And reliable. It has to be. Having a network go down is extremely expensive to any business on many levels; not only the physical cost of labor, hardware, software, and security is a money suck. Customer dissatisfaction also costs money.

If a customer is having a hard time with the company app, or there’s a problem accessing data and customer support can’t get answers, it is very easy for the customer to just up and leave and go to the competition.

The solution, of course, is not to go back to those old phones—the rotary version took almost a minute to dial even a local number—but to ensure that your systems are all “go” all the time. How do you do this?

Plan for the inevitable. You do need an excellent network IT staff, because unforeseeable problems do happen and there’s nothing you can do about that. However, you can reduce the blowback and damage from almost any problem if you have solid plans in place for when the bad stuff comes. Assume your system will be hacked at one point or another. Are you prepared? Coding mixes will cause program conflicts at some point. Do you have a way to resolve this? To prevent it?

While good planning, team coordination, constant testing, and dependable storage can’t save you from every service problem, all combined will reduce the issues and make the problem more manageable. And you do need all four. Because if you’ve got detailed plans for specific problems, your teams are constantly talking about What Happens When, you are always testing, testing, testing, and you trust your data storage, then when something bad happens—and it’s not an if, it’s definitely a when—the damage control should be minimal.

Starting to miss those analog phones yet? You could just sit by Ol’ Reliable and wait for someone to call …

Recent Posts

APM Best Practices to Deliver Big Performance Gains


Continue reading...