Cloud migration is no simple feat. It involves intense planning, resources and cooperation to innovate and optimize successfully.
But, no matter how much an organization plans ahead, there’s always a curveball.
“During a migration process, every cloud provider will try and make it as painless and as seamless as possible, and then on Day Two, you’re effectively left on your own,” said Clara Angotti, co-founder and president of Next Pathway.
While Amazon’s Jeff Bezos famously laments Day Two, favoring instead a permanent startup-like Day One mentality, the reality is that businesses must think about what comes after a migration and prepare for the challenges that arise once workloads are transferred.
CIO Dive spoke with five experts on how to navigate post-migration challenges so that tech leaders can spend more time innovating and less time problem-solving.
Cost
The first bill after migrating to the cloud will come as a shock to businesses persuaded by the cloud’s cost-saving abilities.
“I’ve seen numerous cases where the first day of cutover, the first month of cutover, it’s like, ‘oh my God, I didn’t realize it was gonna cost me this much,’” said Chetan Mathur, co-founder and CEO of Next Pathway.
While companies can save money with the cloud, it’s a different type of operating model. Legacy technology follows a capital expenditure model, while cloud technology runs on an operational expenditure model.
“You take pricing predictions and usage predictions and now you have to look at that in a very different way,” said Shawn Ahmed, chief marketing officer at CloudBees. “I think the bigger problem is just really knowing how to predict what your cost structure is going to be post migration.”
IT budgets have swung in favor of OpEx models since the pandemic, but that doesn’t make the first bill easier to swallow.
It’s important for businesses to remember that they are paying for increased speed of innovation, ability to scale and reach global markets in ways they couldn’t with legacy technology, according to Brian Adler, senior director of cloud market strategy at Flexera.
If businesses can prepare and at least be aware of the cost, they won’t be blindsided. Emily Lewis-Pinnell, VP and cloud practice leader at NTT Data Services, said leaders can create new internal roles focused on cloud cost management and monitoring.
Governance
Cost and governance were the biggest Day Two issues experts identified, and they go hand in hand. When leaders have good governance strategies and awareness, they are more likely to see cost optimization opportunities.
Governance is an issue that most experts feel businesses could nail down in front-end planning, but it’s a top CIO concern.
Companies that aren’t successfully executing pre-migration planning may lose sight of established protocols. Lewis-Pinnell described a Day Two clean-up scenario in which businesses identify holes or mishaps due to employees’ confusion during the process.
Not only does this cause a speed bump for modernization but it also brings into question security of the applications.
Adler described security as a Day Two problem of the past, given the current maturity and capabilities of cloud platforms. However, he said, when ambiguity on roles and responsibilities occurs within governance strategies, security will be an issue.
Analysis
Outside of pre-project planning, a Day Two analysis is the best way to clean-up, regroup and push forward.
Angotti said many companies migrate to the cloud in waves, so that on Day Two, “nobody’s resting on their laurels and saying, ‘let’s just watch things run.’ But [instead asking] what lessons did we learn? How can we improve the process? What’s our index workload look like? And then kicking off another migration initiative.”
Adler said analysis can be helpful to optimize cloud spending because businesses will see spend patterns and unlock new value.
A Day Two analysis won't make up for lack of planning, but will allow businesses to carve a clear path forward. Since cloud migration is not a one-and-done process, taking the time to continually evaluate workloads is vital.
“If you aren’t clear on what’s lacking, you need to get those answers,” said Lewis-Pinnell.
