Last week I took a road trip to meet with some of our customers, and talk about the business benefits they are looking to realize from their cloud investments. At one client, we were discussing the well-documented ability of cloud providers to deliver a new release every four to six months, and how best to incorporate newly released functionality - using the example of salesforce.com Chatter - into the organization. It was at this point that our client - the CIO - threw a curve ball: “those rapid release cycles can be a problem.”
I got ready to hit that ball out of the park by talking about the importance of participating in release previews, and gathering advice from customers participating in the beta release, as well as the importance of organizational change management. That’s when he went on to explain that on an earlier cloud project their e-commerce system went down because one of their on-premise software partners was not able to keep up with the rapid changes in the cloud solution.
Nasty nasty curve ball. Our experience has been that cloud integration is generally easier than on premise integrations because of the service-oriented architecture foundation on which most cloud solutions are built. So all I could think about was how frustrating it must have been for that CIO: feeling held hostage by a vendor who provided a key element of his e-commerce infrastructure when all he was trying to do was improve his customer experience and introduce new products. The amount of incremental revenue he was missing out on while he waited for his traditional on premise vendor to catch up with his cloud vendor crossed my mind too. While thinking about what it must have been like to integrate that application into their environment the first time around, I took a weak swing and said something about how integration is perennially the hardest part of any substantial implementation effort.
I’ve been replaying that conversation in my head ever since, and now I’m ready to step into the batter’s box again. Yes. Not being able to adapt your business processes and introduce new products and services because your application architecture is too brittle to handle it is a big problem. No question. But being able to move so fast that your on-premise apps can't keep up is a pretty high class problem to have.
This increase in agility is one of the primary reasons that organizations are choosing to move their business onto the cloud-based applications and platforms. In the cloud, applications can be configured and re-configured to adapt to changes in business processes or to the introduction of new products and services, with minimal impact to the future upgrade path. And there are cloud integration platforms with pre-built and configurable integration templates to help knit applications together. Sure there may be the occasional on-premise app that makes for a challenging integration, but in the long run, moving to the cloud means that IT can focus on innovation instead of infrastructure.
So what is a CIO to do when some vendors who comprise the application architecture can’t reach cloud speed soon enough? Waiting until every solution provider in your portfolio is cloud-based is not the answer: the competition will have feasted on you by then. A more proactive approach is to conduct a portfolio review with three aims in mind:
- Which applications in my portfolio can be migrated to the cloud platform (or platforms) of my choice? (Hint: it’s not just your least strategic apps.)
- Of those that remain, which have vendors with a roadmap that will support my move to the cloud? Which have substitutes in the cloud ecosystem that I should consider?
- What is the best path to the cloud for my organization?