The ultimate metric of success
When defining cloud migration success, it’s critical to keep in mind that mere cloud adoption is not the goal. Businesses don’t adopt cloud just to adopt cloud. It’s the transformative journey you undertake to achieve the goal on the other side.
The ultimate goal? The ultimate metric? It’s not to get out of your data centers or drop your MTTR (or, it shouldn’t be).
The ultimate metric is customer value.
Consider Netflix. Their data centers were already groaning under heavy demand in early 2008. When a corrupted database took their entire DVD operation down for three days in early 2008, something had to give, and company leaders made the fateful decision to migrate to the public cloud.
Looking back, the decision was a simple one. “It wasn’t core to our business to build and operate data centers,” says Ruslan Meshenberg, Director of Cloud Platform Engineering from 2011-2017. “It’s not something our users get value from. Our users get value from enjoying their entertainment. We decided to focus on that and push the underlying infrastructure to a cloud provider like AWS.”
The focus on customer value also led Netflix away from a simple lift-and-shift approach. “You end up moving all the problems and limitations of the data center along with it.” Instead, Netflix chose to return to first principles and take “the cloud-native approach, rebuilding virtually all of our technology and fundamentally changing the way we operate the company.”
With everything Netflix had to do, its migration took seven years, but the results have spoken for themselves. By the time they completed their cloud migration, they boasted eight times as many customers as when they started, and monthly streaming hours had increased 1,000x. In January 2016, Netflix’s cloud infrastructure enabled them to expand service to an additional 130 countries simultaneously.
Talk about creating customer value.
How many, how well, how much?
Customer value may represent the ultimate success metric, but in many ways it’s more of a guidestar. It’s the point you’re aiming for, but you need other ways to map the course and track the progress of your cloud adoption journey.
We could probably spend days (or weeks, or months…) running through the various ways to measure your cloud efforts, but channeling them into three key pillars can keep you moving without losing focus on the ultimate goal of customer value: How Many, How Well and How Much.
What is it?
How many workloads or applications — how much of your business — have you migrated to the cloud?
Why does it matter?
Think of private data centers like a massive corporate gravity well. It’s difficult to break free of their pull, and modest projects rarely achieve the necessary escape velocity. Some applications might break free, but they’re more like the probes launched by NASA and other space agencies. Easy to celebrate, fun to cheer on, but fundamentally isolated.
Within the data center gravity well, businesses are constrained by the limits of hardware. Expansion means racking new servers, and scale means maintaining them all the time. Every new move requires capital expenditures, hamstringing your ability to innovate or scale at speed.
Migrating your applications to the cloud is all about creating a second, separate gravity well. And as you move more and more applications, cloud’s gravitational pull gets stronger and stronger. When cloud’s gravity exceeds that of the data center, you cross a natural tipping point from pushing projects to the cloud to pulling them from the data center. This is where the flywheel of innovation can really start cranking.
Depending on the scale of your business and your data center footprint, this process can take years. It took Netflix seven years. Capital One, which will close the last of its data centers during 2020, is on a similar seven year trajectory.
Our free eBook, 7 Common Cloud Adoption Mistakes, helps you avoid the pitfalls that can hold businesses back when they go cloud.
What is it?
How well-architected are the workloads you’re migrating to the cloud?
Why does it matter?
There’s more to successful cloud adoption than just taking all your stuff out of data centers and tossing it into the cloud. This lift-and-shift approach just takes what you’re already doing and puts it somewhere else. It can be a useful stopgap measure, but it doesn’t take advantage of cloud’s true capabilities.
Truly thriving with your cloud adoption involves a lot of rethinking and rearchitecting your systems to take advantage of the cloud. The AWS Well-Architected Framework lays out five pillars:
- Operational excellence
- Cost optimization
- Performance Efficiency
Understanding how each pillar works, how they work together, and using them to optimize your applications for the cloud does represent an upfront investment of time, training, and effort, but it will also pay dividends with a more successful, sustainable, and cost-effective cloud adoption.
You can track your progress toward well-architected solutions by measuring key metrics related to the five pillars, with a particular focus on compliance, resilience, and elasticity. How many availability zones are you using? Is your data encrypted? How well? Is your team using resource tagging consistently? These and tons of other metrics can be pulled out of API log trails, and much of the process can be automated using open source tools like Cloud Custodian.
Learn the principles of building secure, efficient, cloud-enabled apps with AWS Community Hero Mark Nunnikhoven’s course, Mastering the AWS Well-Architected Framework
What is it?
How much are you spending (or not spending) in the cloud? How cost-effective are your workloads?
Why does it matter?
The migration from data centers to cloud entails more than shifting where your applications are running. It entails shifting your mindset. Data centers require massive upfront capital expenditures, and ongoing costs to operate, maintain, and upgrade hardware systems in addition to the software running on them.
With cloud, all of that goes away. You only pay for what you use. Consider the case of HBO. When they release a hotly anticipated episode of Game of Thrones, or Watchmen, or Westworld, demand spikes. HBO even refers to this traffic pattern as “the wall”.
In years past, “the wall” has crashed streaming services and driven fans to distraction. Supporting such peaky demand with a legacy data center would be horrendously expensive, assuming it’s even possible. You’d have racks of servers running at 1-2% utilization most of the time, just waiting to soak up the demand surges. Instead, HBO started deploying its programming through AWS, using EC2 instances and auto-scaling. When that started facing strain, they moved to Kubernetes to handle the load better while also improving utilization.
The trick to optimizing your cloud costs, then, is to figure out how to use only what you need, and architect your systems accordingly. This requires getting out of the legacy data center mindset and into a cloud state of mind, using well-architected frameworks and best practices as a guide.
How to How…
How Many. How Well. How Much. There’s one more critical measure that you might be considering. How are you going to get all of the “how” done? Or rather, who is going to get all of the “how” done?
Ultimately, cloud adoption isn’t a technology challenge. It’s a people challenge. And if you want to succeed, you have to start with your people.
That’s why companies like Verizon, Capital One, NAB, and many more have invested so much in talent transformation, from providing training to forming study groups, opening dedicated time for learning, and measuring certifications. They’re building the “who” to get the “how” done. And you can, too.
Ready for your transformation to blast off?
Two-thirds of organizations say they don’t see the benefits they expected with the cloud. A Cloud Guru can help get your organization to the cloud (and its benefits) faster. ACG for Business provides everything you need to transform your talent — and your business.