Here’s why Google Cloud support is now priced per person

One of the key messages out of Google Cloud Next last week: The company has gotten serious about serving enterprise customers in a way that meets their needs. A key component of that is a new set of support plans for customers of its cloud platform.

Instead of paying one big support fee, businesses will pay per user for support privileges, based on how quickly they will get a response. The new plan replaces a set of monolithic tiers that required businesses to pay a flat fee per month plus a percentage of their monthly product usage. 

In the past, Google wasn't exactly known for its support capabilities, but that's something the company has changed in recent years. Google engineers originally believed they could create great technology that was self-service and easy to use, and that would be enough for business customers, said Dave Rensin, the company’s director of customer reliability engineering.

"And we didn't appreciate — again, this is engineers, we make this mistake as a class all the time — the human dynamic," he said. "We didn’t appreciate the anxiety that people feel moving to the cloud because of the control they're giving up. We didn't appreciate any of those things. But that's changed."

The cheapest support tier costs US$100 per user per month and is meant for those folks who can get by with receiving a response within four to eight business hours of their request. Production engineering support is the next tier up and costs $250 per user per month for a one-hour response time on critical issues.

For absolutely critical issues, there’s on-call engineering support, which will let people page Google engineers about critical issues for $1,500 per user per month. Those requests get a 15-minute response time, 24 hours a day, seven days a week.

Over the course of 2017, Google will be working to move its customers over to the new support regime. As part of that, businesses will have to pick out a set of licenses for their employees.

Dave Rensin, Google’s director of customer reliability engineering, said that he recommends any team with a production application in Google's cloud pay for at least one member of their own on-call rotation to have access to the company's highest-priority support tier. That way, in the event something critical breaks, it's possible for the customer to have a "safety valve."

Beyond that, he also suggested customers not duplicate their support requests internally, so they get a faster response from Google. One of his hopes for the new system is that it will cut down on those duplicate requests, which would often result from the companywide support offerings.

"Now, we'll get 20 or 30 dupes of exactly the same issue at exactly the same moment. It creates a lot of friction for solving the first one," Rensin said. "If everybody storms the buffet at the same time, nobody gets fed."

One of the key benefits of the change is that customers won’t see their support bills spike in time with their usage of Google's cloud platform. Most customers will see their support bills go down as a result of the change, according to Rensin.

In his view, support is integral to Google's competitive stance in the public cloud market, where the company is facing down other titans, including Amazon Web Services and Microsoft Azure.

"If we want to win, we have to have the best technology — we just do — at the right price, but also we have to have a superior human relationship with people," Rensin said. "We have to be better, human-to-human, to deal with than the alternatives in the marketplace. That’s a must-have, or you can't win. So that's what we're going to go do."