Alas, there are way too many ways to get it wrong. So here are some of the most outstanding mistakes you should avoid.
Think before you move
It's simple — really, it is. Look before you leap. Nonetheless, a survey by THINKstrategies and INetU a few years ago on cloud migration found that 70% of respondents admitted they had to change their cloud design before deployment.
But, wait, there's more! More than half had to change their plans within the first six months. Overall, 43% of the cloud projects failed or stalled, and almost half needed more budget in those critical first few months.
Those numbers sound about right to me. All too often, someone in the executive suite decides the cloud is magic and will make all things bright and beautiful. That's not the way this works. Moving or expanding to the cloud only works if you do your homework first and then deploy.
All or nothing
I've also seen way too many companies decide if they're going to move to the cloud, they're going to move everything. I call this the kitchen sink problem, as in "we're going to migrate everything and the kitchen sink server."
Just don't do that.
You can't forklift all of your infrastructure and services at once. It just won't work.
Take your time. Consider all the issues and then slowly and methodically start moving your IT stack to the cloud.
Focus only on saving money
Yes, the cloud really can save you money. But don't think that just because you're moving from a traditional IT capital expenditure (CAPEX) model to the cloud's operational expenses (OPEX) model, more cash will automatically show up on the bottom line. It doesn't.
It's a different business model. What the cloud gives you — in theory — is more flexibility. That may mean you'll save money you can call out in the annual report. But, if you don't do it carefully, a cloud can cost you every bit as much — or even more — than running your own data centers.
All clouds are created equal
Oh, no, they're not. Clouds vary wildly in their capabilities, costs, and purposes.
It's not just that Microsoft Azure could work better for you than Amazon Web Services (AWS). Each cloud comes with all its strengths and weaknesses. In addition, your business comes to the cloud with different requirements than the one next door, so make sure you match your needs to what each cloud can provide.
Find a path that works for you and stick to it
Finally, even after you have your first cloud project migrations under your belt, make sure you've learned the lessons from those projects. I've seen far too many companies migrate one project using one method, then another migration with still another approach. Before you know it, you've reinvented your cloud architecture so many times that your spiffy new 21st-century IT infrastructure is a tangled mess.
The goal isn't to recreate an IT mess in the cloud. Once you've found a way that works right for your company, keep using it. Do not reinvent the wheel.
All five of these mistakes have one thing in common: They show people are not doing their homework.
Before making any significant IT changes, look carefully at what others have done before, what you want as a result of the change, and then — and only then — make your cloud move. | | |
