Cloud and on-premises workloads have reached a balanced “equilibrium”
After years of debating, experimenting, and evolving, businesses are now dividing their workloads in a nearly even split between cloud environments and on-premises infrastructure. This is not by accident. It’s the product of years spent discovering the distinct advantages each solution brings to the table.
Some workloads thrive on the scalability of the cloud, while others remain firmly grounded on-premises due to the unmatched control and security it provides.
For years, the expectation was that enterprises would rush headlong into the cloud. Industry voices echoed with predictions of an overwhelming migration to the cloud, a transformative shift with no looking back. But today, that vision is tempered by reality—a reality that sees businesses choosing a strategic balance.
Michael Coté from VMware pinpoints this, describing an almost perfect 50:50 split, with about half of workloads residing in traditional data centers and the other half exploring the vast resources of the cloud. The real value for enterprises lies not in pursuing cloud over on-premises, but in using each as a strategic asset.
Cloud growth continues, driven by AI, but challenges remain
The demand for cloud isn’t slowing down; in fact, it’s fueled by the explosion of AI. AI needs enormous computational power and storage, resources that the cloud provides without breaking a sweat. Even with this momentum, some companies are hitting bumps along the road. Costs can creep up unexpectedly, and issues around data privacy and control make some executives hesitate.
Some businesses feel misled by marketing that painted cloud as the universal solution, only to find that the real benefits aren’t always as straightforward as advertised.
So, we’re seeing more discerning decisions from companies. They’re sizing up workloads to figure out which truly fit the cloud’s offerings and which are better kept close to home. This is backed by recent earnings from AWS, Microsoft, and Google, showing the spike in cloud use—AI being a major contributor.
Despite these impressive growth rates, there’s a clear sense of caution among executives, a sense that cloud isn’t a magic-fit solution for all.
“Lift-and-shift” migrations without refactoring increase cloud costs
Here’s where many companies get tripped up: they move applications to the cloud as-is, in a process known as “lift-and-shift.” It’s quick and easy, but without adapting these applications to leverage cloud-native features, costs often spiral. Instead of saving on operational expenses, these businesses find that their bills only grow.
Why? Well because these “lifted” applications are designed for traditional infrastructure, not optimized for the cloud’s unique architecture, making the cloud seem more like a costly experiment than a lean, efficient solution.
Let’s take AI as an example. Unlike traditional workloads, AI applications are usually built with the cloud in mind. They’re designed to tap into cloud capabilities from the start, which means they dodge many of the pitfalls of lift-and-shift.
There’s a catch though—even in the AI space, without careful planning, costs can get out of control. Some companies push into AI without a clear strategy, sinking money into projects that lack clear returns. When that happens, the cloud stops being an asset and starts feeling like a bottomless expense.
Actual movement of workloads back to on-premises remains low
With all these challenges, you’d think more companies would pull the plug on cloud and shift back to on-premises setups. But that’s not what’s happening. The idea of “repatriating” workloads may be tempting, but the reality is more complex.
Moving things back from the cloud can’t be done by simply flipping a switch—it’s an expensive, involved process, one that most businesses find simply isn’t worth the cost or hassle.
Even though many companies entertain the idea of bringing workloads home, those plans often stay just that—plans. Once businesses weigh the investment in time, resources, and potential disruption, they realize that sticking with cloud or hybrid setups makes more sense. That’s why, even as costs and control concerns surface, the number of companies actually moving back to on-premises remains low.
Cloud or on-premises decisions are influenced by human factors over technology
At the end of the day, the cloud-or-on-premises debate often comes down to people, not technology. You need the right talent to refactor applications, manage cloud resources, and really capitalize on what the cloud has to offer. Without that skilled workforce, companies are left making do with current setups or adopting a hybrid model simply because it’s what their teams can handle.
Companies are realizing that whether they stick with on-premises or push forward with cloud, there’s more to consider than computing power or cost. They have to carefully consider whether their teams can make the most of either option. That means this “people factor” will likely keep decisions complex, creating a flexible, case-by-case approach that may carry forward for years.
Final thoughts
Are you set up to make the cloud work for you, or is it just another expense eating into your margins? Think about it—carefully. Do you have the right people, strategy, and vision to leverage the cloud’s full potential while keeping a grip on costs, control, and adaptability.
In a world where AI and data dictate the pace, how you balance cloud and on-premises decisions could very well be the difference between leading or getting left behind. Are you ready to make that choice with intent, or are you letting it make itself?
