There were many curious things about the winner of the Preakness Stakes on Saturday, the most obvious being the horse’s unusual name.

Lots of thoroughbred horses have weird names—it’s something that they’re known for, really—but Cloud Computing, bizarre as it sounds, is an absolutely perfect name for a winning race horse in 2017.

The name drew my attention when I first saw it trending on social media, but it was a comment from a key figure on the cloud revolution that got me thinking about it with a little more depth. Box founder Aaron Levie—who has banked his entire adult life on cloud computing, launching his company as a college student around the same time that Facebook was just getting off the ground—noted that the horse’s win, bizarre as it was, sort of hinted at a turning point for our broader understanding of the cloud.

Looking back over the past decade of trying to get the world to understand cloud computing, it's now obvious we were missing horse racing.

— Aaron Levie (@levie) May 21, 2017

While cloud computing is a fairly mature medium at this juncture, it perhaps isn’t getting used as much as it potentially could by many organizations. We’re not maximizing it.

It’s something that Levie’s comment hints at in a small way—the idea that selling “the cloud” as a concept came with years of unrecognized hard work. But it also hints that, despite all that progress, the cloud still has a way to go.

That’s something reinforced in a recent survey by the Uptime Institute, which—according to its roughly 1,000 respondents—found that nearly two-thirds of data that could otherwise be hosted elsewhere remains managed locally—a level that hasn’t fallen since 2014.

“Increased performance at the processor level, further expansion of server virtualization, and the adoption of cloud computing have all created an IT foundation that differs greatly from those seen just 5 years ago,” the Uptime Institute’s Matt Stansberry noted in a news release. “Through this change, enterprise-owned data centers have remained a central component.”

What’s keeping the giant data centers running? Simply, there are still a lot of places where the old data centers and server rooms still make a lot of sense. (Or, at least they feel like they do.)

There are settings where the cloud seems to slow things down. Just as an example: If you’re sharing huge files back and forth on a regular basis—say, high-resolution graphics or video files—there can be a point where going back and forth between an offsite server can cramp your style, rather than help maximize your organization’s freedom. Or perhaps your specific need for computing is so complex and narrow that adding the cloud to the mix really complicates things quite a bit.

That sounds like a pretty common boat, right? It was the subject of a recent piece in TechTarget, which implied that these factors are definitely there for many organizations that haven’t yet shifted toward the cloud. But it also pointed out another issue entirely: a lack of preparation and understanding of how the cloud can benefit the organization, beyond just financially.

This often means that leaders are pushing everything to be done through a single public cloud provider, despite the productivity benefits that might come with a more spread-out approach. Marc Clark, director of cloud strategy and deployment for the data warehouse firm Teradata, noted that this is a common issue among enterprise users.

“Even worse than moving somewhat blindly to the cloud, I often hear of companies having a single public cloud provider strategy,” Clark explained to TechTarget.

This is a problem, because as much as we’d like to imagine that one size can fit all, your association’s marketing team and your finance team simply have different needs and may be better served by different cloud infrastructures.

And this lack of understanding of specific organizational needs can create other issues as well. According to a recent survey from RightScale, more than half of respondents said that their most important cloud initiative was to optimize their existing cloud use in an effort to save costs. But much of that current spend is being wasted, according to the report: Respondents to the survey self-estimated that they waste 30 percent of their cloud spending, while RightScale says that the waste load is between 30 and 45 percent.

It’s worth pondering whether this wasted spend is the result of poor planning earlier in the process. If a cloud infrastructure was better built around an organization’s needs, it likely would be more cost-effective for the enterprise.

In comments to TechTarget, Baselayer Technology’s Chief of Staff Samir Shah spoke to this point by noting that “unanticipated utilization” of cloud resources can cause an increase in costs after the platform is implemented.

“The key is to develop a cloud-based strategy that allows for proactive decision-making instead of reactive decision-making,” Shah explained. “Keep in mind that the software supporting a company’s data infrastructure is critical to ensuring this proactive mindset.”

To put this into terms any organization can appreciate: Your association has a lot of different, sophisticated needs around each of your departments, and those needs often differ significantly based on the task at hand. Why shouldn’t this be the case around the cloud?

With more individual departments getting involved in tech decisions, this is a place where plotting things out ahead of time could save a lot of pain down the road. And it could maximize your chances for success with the cloud, too.

You want to win the race—and not with a photo finish, either.

Jockey Javier Castellano, front left, rode Cloud Computing to victory on Saturday at the Preakness Stakes. (via Facebook)