The 3 dumbest things enterprises do in the cloud

You’re going to make mistakes. I tell my enterprise clients that every week.

However, there are mistakes and there are mistakes that are more like self-inflected wounds. Here are three of the dumbest mistakes I’m now seeing enterprises make in the cloud efforts.

Dumbest mistake No. 1: Keeping the data on premises but the compute in the cloud

When helping clients plan their cloud efforts, I regularly hear, “My data is sacred, so we don’t want to put our data in the cloud. However, we’re paying too much for compute and datacenter space, so let’s place that on some public cloud.”

That is not a good move for a couple reasons. First, you’re going to hit a great deal of latency. In fact, I’ve never seen this kind of hybrid architecture work due to the lags. Second, security becomes way more difficult. In fact, you typically end up with more vulnerabilities.

Dumbest mistake No. 2: Firing staff working on the legacy systems too soon

Enterprises typically change their budgets around the use of public clouds, and publicly traded companies typically don’t want to have any upticks in expenses even during transitions. For cloud migrations, they budget for a zero sum game, and to do that they get rid of the staff that looks after the legacy systems—before moving the workloads to the public clouds.

That’s a huge mistake. Typically, significant cloud migrations take a year or more. You’re going to need your legacy systems during that time to run the business. So you still need your legacy staff for a good while. Moreover, you’re never going to completely get all your applications on the public cloud. Many applications should not move due to their economics, and others can’t move due to some limitations in the technology. So you still need some of your legacy staff for the long term. You’ll still save, but only over the longer run.

Dumbest mistake No 3: Overselling the benefits of the cloud

My, how the pendulum has swung! The people in enterprise IT who pushed back on the cloud just a few years ago are now aggressively embracing it. They see the writing on the wall.

But in a hype-driven frenzy, they are overstating the ROI that public cloud computing will bring. As a result, they are falling short in the eyes of the enterprise’s leadership.

The truth is that the mileage you get from cloud computing varies a great deal. That’s why I spend a great deal of time on the business case to tell enterprises exactly what they can expect. You have to do that, too.