Google's 100 minute Gmail outage this week is going to cost Google and every other vendor that is selling Software as a Service (SaaS) applications millions of dollars in sales.
The Gmail outage is the equivalent of a security software vendor suffering a major breach.
It's a big fat, black eye and credibility-killer for anyone and everyone selling SaaS or cloud-based services.
Google, in a post on its Gmail blog, admitted that the outage was a "Big Deal." So what caused the 100 minute meltdown? Believe it or not, regular maintenance. That's right. Google says it took "a small fraction of Gmail's servers offline to perform routine upgrades" and "slightly underestimated the load which some recent changes (ironically, some designed to improve service availability) placed on the request routers." Underestimated. That's the understatement of the century.
So this is all about not having request router capacity. Is Google serious? Try buying a few more request routers.
Anyway, the damage has been done. And it can't be undone.
David Koretz, founder and CEO of BlueTie, a Rochester, New York SaaS pioneer with a $4.99 per user per month fees for e-mail and calendaring, said that anytime someone in the SaaS industry has "reliability challenges it harms the whole industry."
"It creates a perception in the customers' minds that there is a reliability issue," says Koretz.
That said, Koretz said the service level performance of Google, BlueTie and other hosted e-mail services far surpasses the reliability of the average business hosting Microsoft Exchange on their own servers.
Koretz claims that the average service level performance for a business running Exchange is seven hours of downtime a month or a 99.1 percent service level compared with 99.99 percent for BlueTie.
"Is it a big deal?" asked Koretz rhetorically. "Yes! Is it frustrating? Yes! It is frustrating because it causes reliability concerns for the industry. But absolutely at the end of the day Google and BlueTie can do a better job of delivering reliability than the average Microsoft Exchange customer can do for themselves."
That may be true. But Perception is reality. And Google just changed the perception of millions of customers.
What impact do you think the Google Gmail malfunction will have on your SaaS business?
I'm amazed when I read articles like this, do you know of anything that has 100% uptime? If so, why would we ever have to replace anything. No matter what redundancy you put in place, you will not be able to cover every possibility. Whenever you have a SaaS business, you now open yourself up to the biggest possibility for an outage because you are dependent on the Internet, you have no control over it at all! If some contractor digs up a major fiber cable and knocks out the whole west coast (which has been done many times), to someone on the east coast who's Internet connection works just fine, but he can't access your servers on the west coast because of the break, his thought is that their servers must be down. In reality the servers were working just fine but no one could access them because of the cable break.
What is wrong here is the uptime claims by almost every SaaS company, "Koretz claims that the average service level performance for a business running Exchange is seven hours of downtime a month or a 99.1 percent service level compared with 99.99 percent for BlueTie." His servers may be up 99.99 percent, but is his service really that good for someone on the west coast whose ISP has only a 90% uptime performance?
While working with a client on selecting an ERP system for their business, one of the vendors, a SaaS ERP solution, was demonstrating their software, at the beginning of the demo they spoke of their 99.99% uptime, within one hour of the demo, they lost connectivity to their servers 3 times. This factored into the clients decision, the software was great but could they believe the vendors claim to 99.99% uptime when it crashed so many times in a short demo? The lost connectivity had nothing to do with the vendors servers, it actually was that day the clients ISP, but the fact that they made the claim and everyone in the room took out their calculators to really see if there would be 99.99% uptime made the vendor look untruthful.
Yes, 100 minute outages are frustrating, but instead of the media crying wolf every time something like this happens, they educate people on some of the reasons why these outages happen and sometimes they are not the vendors fault.