Amazon Web Services (AWS) recently announced that it would begin billing customers by the second for its popular EC2 cloud computing service. Although the move is welcome news for many organizations, it raises new questions about flexibility and the risks of vendor lock-in.

Vendor lock-in explained

When AWS launched in 2006, customers were charged by the hour for EC2 instances. Since then, usage totaling less than 60 minutes has been rounded up to an hour, resulting in cost inefficiencies for customers.

Amazon’s per-second billing model is a clear challenge to Google and Microsoft’s per-minute billing models. But it’s also a strategic response to rising market demand for more flexible cloud computing solutions.

And that’s where vendor lock-in enters the picture.

As cloud technology vendors compete to give customers more options and services, IT organizations are starting to take a closer look at the opportunity cost of locking into a specific provider for their cloud computing needs.

Think of it like this: Buying an iPhone instead of an Android phone might seem like the right decision today. But if Google releases a new suite of must-have services in six months, your experience using those services will be more painful than it would be if you had chosen an Android device. Your decision to buy an iPhone has, to a degree, locked you into Apple services.

In essence, that’s the dilemma cloud services customers face. Amazon’s new pricing model is just another reminder of the costs associated with vendor decisions.

You can’t avoid lock-in. But you can minimize its impact

Everyone is locked into something. When you choose between AWS, Azure or even iOS and Android, you’re making a decision to tie your business (or your mobile device experience) to a specific provider and the slate of services, features and options they have to offer.

But although a level of lock-in is inevitable, you can minimize its current and future impact on your organization. The trick is to analyze your options and make smart decisions based on a set of key considerations.

  • Existing Assets—Evaluate your current technology environment and determine which provider makes the most sense according to your existing assets and services. For example, if you are primarily a Microsoft shop, then the cost of locking into Azure for cloud computing services may be less than the cost of locking into AWS or Google Cloud Platform.
  • Go-forward Business Strategy—Your go-forward technology and business strategies should always inform your cloud computing decisions—and your technology strategy should always follow your business strategy. By selecting the cloud platform that is most closely aligned with your business goals and priorities, you can reduce friction and protect your ability to achieve critical business outcomes down the road.
  • Interoperability and Portability—Interoperability is an obvious concern. It’s untenable to commit to a cloud platform that doesn’t play well with other solutions in the organization. But portability is just as important when selecting a cloud vendor. In fact, one of the reasons why many organizations are considering containers is to improve portability between platforms.
  • Innovation—In the current marketplace, cloud vendors differentiate themselves by releasing innovative services and features like Amazon’s per-second pricing. If you’re interested in maximizing access to innovation, then you should consider a multi-vendor strategy (see my article “Creating a More Strategic Multi-Cloud Environment”). But if you’re committed to a single-platform approach, you will need to determine whether each vendor’s innovation trajectory aligns with your organization’s needs.

Cost is always a factor in technology selection. But in this case, it’s not just the cost of the service that you need to consider. You also need to evaluate the cost of migrating to a different platform if you choose the wrong platform. So, be smart—analyze your requirements, evaluate the marketplace, and make the right decision the first time.