How Should Executives Lead a Cloud Transformation?
Sep 03, 2019
Some executives have developers try to build net-new systems of engagement without any workflows or controls, but the key to successful cloud migration is a common set of workflows, automated policies, and central shared services that developers can use to channel applications securely and successfully onto the cloud.
We often get asked these questions: “The world's going to be going cloud because I have to build these net-new applications to differentiate my business. What's the blueprint for doing that?” There's no single blueprint, but I think there are some principles.
The first of those is that you have to think about short-term quick versus long-term sustainable. And one option is “Hey, let's unleash all our teams to go and build applications quickly and pick the platform that they want to use for those application types.” And while that may be a relatively fast way of building some of these net-new applications, it doesn't really instill this machine-like engine that you need to create to achieve this over the longer-term.
In our experience, what we see most people doing is stepping back and making an investment and recognizing, “Hey, the world's going to be multi-cloud. I'm going to need multiple cloud platforms to be able to differentiate my business over the longer term. Those applications need to be secure, managed and appropriately networked.”
There's essentially a cloud foundation that you need to put in place. Google talks about this publicly, and I think it's accurate, the notion of a cloud foundation, which is to say, “Hey, let me step back and say there are fundamental roles in any organization. There are ops people, security people, networking people, and developers. How do I empower the development teams to be able to build net-new applications in a consistent way while satisfying the needs of the ops, security and networking teams that have to ultimately keep those applications running?”
So this idea of, how do I build this institutional capacity around a net-new application delivery onto cloud infrastructure? The paradox is I could go faster by just unleashing my development teams to build new applications quickly on the infrastructure of their choice.
But the long-term institutional capacity to do so requires stepping back and saying, “What is the foundational investment I need to create to enable my diverse application teams to build applications on premises or on Google or on Azure, on Amazon—and have the implicit sign-off of ops, security and networking? So that when those applications get there—if and when they start to become foundational to my company—I’m confident that they are built in the appropriate secure, manageable monitor-able way?”
I think that's the most common pattern we see is this shift from yes—rather tactical—let’s build net-new applications to—the more step back—I have to create this capacity for my organization to survive in the longer-term. The management consulting vernacular would be digital transformation. But I think the more practical reality-based question is, “How do I get good at building and delivering new applications in a sustainable way?”