Get all your news in one place.
100's of premium titles.
One app.
Start reading
inkl
inkl

FOUND: Why enterprises abandoning single cloud strategy - What They Never Told You

For most of the past decade, picking a cloud provider worked a lot like picking a bank. You chose one, built your entire operation around it, and switching later was expensive enough that nobody really planned for it. That logic is breaking down fast, and the breaking point has a specific date attached to it.

On May 7, 2026, a cooling failure in a single data-center hall at AWS's us-east-1 region knocked services offline for roughly 28 hours. Coinbase's trading systems went down for about seven hours, FanDuel couldn't process cash-outs in the middle of NBA playoff betting, and CME Group's trading tools failed.

It was the third major failure in that same AWS region since 2021. Each one had a completely different cause, including a scaling bug, a DNS race condition, and now a physical cooling system giving out under heat.

Why This Outage Got More Attention

What made this one sting more than the previous two wasn't simply the outage itself. It was the timing. A sports-betting platform going dark during a playoff game isn't a quiet infrastructure footnote; it's a headline, and headlines get read by people who don't normally think about which cloud region their vendors run on.

For many companies, the aftermath wasn't just a technical post-mortem. It became a harder conversation about whether running everything through a single provider is still a decision anyone can easily defend after watching a competitor's trading desk go dark for most of a day.

Companies looking to reduce this kind of exposure need to consider more than cloud-provider selection. Reliable connectivity between locations, cloud environments, and business systems is also important. An sd-wan managed service can help organizations manage network connectivity across distributed environments while supporting their wider infrastructure strategy.

The Math Finally Caught Up

Multi-cloud isn't a new idea. What's new is how ordinary it's become. Recent data shows 89% of companies now use some kind of multi-cloud setup, and 59% are actively running multiple public cloud providers rather than treating a second cloud as a backup plan gathering dust.

Just five years ago, that same conversation mostly happened in risk-management meetings nobody outside IT ever attended. Now it happens because executives have seen what a major cloud failure can do to revenue, customer trust, and operations.

The Cost of Diversifying Has Changed

The basic risks of single-cloud dependency were never a mystery. Every enterprise architecture review for the last decade has had a slide about them.

What's different is that the cost of doing something about those risks has dropped enough that ignoring them can be harder to justify. Companies don't necessarily have to move everything to another provider at once.

Instead, they can identify critical workloads, establish alternatives, and gradually build the ability to shift operations when circumstances require it.

Even the Cloud Giants Are Building Bridges to Each Other

Maybe the clearest sign of where this is heading is that AWS and Google Cloud, direct competitors fighting for the same enterprise dollars, have started building infrastructure to connect to each other.

Their new multicloud networking partnership lets companies set up private, high-speed connections between the two platforms in minutes rather than weeks, and Microsoft Azure is expected to join a similar arrangement later this year.

Why Competitors Are Supporting Multi-Cloud

Think about what that actually means. Two companies that spend billions trying to win customers away from each other are now building the plumbing to make it easier for those same customers to use both of them at once.

That's not generosity. It's an admission that enterprise customers were going to demand that flexibility regardless of what the providers preferred.

Customer Demand Is Driving the Change

This kind of move rarely happens because a vendor suddenly feels generous toward its competitors. It happens because enough large customers started asking for it that refusing started to cost more in lost deals than agreeing cost in lost exclusivity.

When Amazon and Google are willing to build each other a bridge, it's a reasonably reliable signal that demand for cross-cloud flexibility is no longer something providers can easily ignore.

What's Actually Driving the Shift on the Ground

For companies living through this shift firsthand, the decision rarely comes down to one dramatic outage. It's usually a collection of smaller frustrations that a major failure finally makes impossible to ignore.

A pricing negotiation can go sideways because a vendor knows you have nowhere else to go. A compliance requirement might demand data residency in a country where the primary provider has limited options.

A procurement team might also realize that nobody actually tested what would happen if the primary environment went down for a full day.

The Main Pressures Behind the Shift

Several factors are pushing enterprises toward more flexible infrastructure strategies:

  • Reducing dependence on a single cloud provider
  • Meeting data residency and compliance requirements
  • Creating realistic disaster recovery options
  • Improving negotiating power with cloud vendors
  • Supporting workloads in regions where one provider may not be ideal
  • Giving critical applications an alternative environment during failures

These pressures don't necessarily mean every company needs three cloud providers. They do mean enterprises are taking another look at how much dependence they are comfortable carrying.

The Skills Gap Makes Multi-Cloud Difficult

Running two or three cloud environments well is a fundamentally different operational problem than running one well. The skills, monitoring, failover testing, and cost optimization required across providers don't scale simply by hiring one more cloud engineer.

Each provider has different pricing models, security controls, APIs, and operational conventions. Teams have to understand those differences while still keeping the overall environment consistent.

Existing Cloud Expertise Doesn't Transfer Automatically

Most internal IT teams built their expertise around a single provider's ecosystem over years. That knowledge is valuable, but it doesn't automatically transfer when a second or third cloud gets added.

Engineers now have to understand how different platforms interact while keeping security, performance, and reliability consistent across them.

Hiring Makes the Problem Bigger

There's also a staffing reality that rarely gets said out loud. Engineers who deeply understand one provider's ecosystem are already difficult to hire and retain.

Finding or training a team that's genuinely fluent across two or three providers is a different order of difficulty. That's one reason many enterprises lean on outside expertise rather than trying to build the entire capability internally.

The Coordination Problem Nobody Talks About

Here's the part that rarely makes it into the executive summary. Multi-cloud doesn't just mean spreading risk; it means spreading complexity, and complexity has its own failure modes.

A company running workloads across three providers now has three different security models, three different sets of APIs, three different billing structures, and potentially three different groups of engineers responsible for separate pieces of the environment.

Why Multi-Cloud Management Matters

This is exactly why multi cloud management has become its own specialized discipline rather than simply an extension of whatever a company was already doing.

Enterprises moving in this direction increasingly need a single layer that can see and control workloads across providers consistently. Otherwise, they may end up stitching together several dashboards and hoping someone notices when something breaks at 3 a.m.

More Clouds Don't Automatically Mean More Resilience

Without proper coordination, a company can end up technically multi-cloud while still experiencing much of the operational chaos of managing several isolated environments.

The technology may be distributed, but the policies, monitoring, security controls, and recovery processes may not be. In that situation, having workloads on multiple clouds can create the appearance of redundancy without delivering the resilience the business actually expected.

Multi-Cloud Can Go Wrong Too

It's worth being honest about how often this goes wrong in practice. A company announces a multi-cloud strategy, migrates a handful of workloads to a second provider, and declares victory without building the operational muscle to actually manage both environments as one coherent system.

Six months later, nobody can say with confidence which provider is hosting what. Security policies may have quietly drifted apart, and the redundancy everyone assumed existed can turn out to be more aspirational than real.

Multi-Cloud Is an Operating Model

The answer isn't necessarily to avoid multi-cloud. It's to treat it as an operating model rather than simply a migration project.

The companies that benefit most are likely to be the ones that standardize how they monitor, secure, govern, and recover workloads regardless of where those workloads happen to run.

The Trade-Off Isn't for Everyone

None of this means single-cloud strategies are dead. For a smaller company with a simple workload and a tight budget, concentrating everything with one provider can still be the right call.

The complexity of managing multiple clouds well can genuinely outweigh the risk it's meant to solve, especially when the business doesn't have workloads that justify the additional operational overhead.

When Single-Cloud Still Makes Sense

A smaller business may not need multiple providers simply because the additional cost and complexity could outweigh the benefits.

For straightforward workloads, a single provider can offer simpler management, easier monitoring, and a smaller technical team. The important point is that the decision should reflect the company's actual risk rather than an assumption that multi-cloud is automatically better.

Critical Enterprises Face a Different Calculation

The calculation changes for companies where downtime directly affects revenue or customer trust. A trading platform, payment processor, betting service, or other business that genuinely cannot afford to go dark for a day has more reason to question whether concentrating everything with one provider is still the best risk decision.

The us-east-1 outage didn't teach anyone anything they didn't already know on paper. What it did was turn an abstract risk into a concrete memory that many boardrooms are still talking about months later.

That kind of memory tends to outlast a single news cycle. It's part of why the shift toward multi-cloud looks less like a passing trend and more like enterprises finally acting on something they had been quietly worrying about for years.

For organizations evaluating connectivity options as part of this transition, CypressTel offers networking solutions that businesses can consider alongside their broader cloud and infrastructure planning.

Sign up to read this article
Read news from 100's of titles, curated specifically for you.
Already a member? Sign in here
Related Stories
Top stories on inkl right now
One subscription that gives you access to news from hundreds of sites
Already a member? Sign in here
Our Picks
Fourteen days free
Download the app
One app. One membership.
100+ trusted global sources.