As demand grows, data centers continue to face the challenge of transforming their operations from mere cost centers into profit centers. This fireside chat, led by Stephen Warren, CTO of Data Center Dynamics, features insights from industry experts Dan Moore of CyrusOne and Stuart Chambers of MCIM by Fulcrum Collaborations. Together, they explore the complexities of managing vast amounts of data and the strategies necessary to leverage it effectively.
The Challenge of Data Management in Data Center Operations
As data continues to grow exponentially, organizations struggle to utilize it effectively. The conversation highlights several key points:
- Data Overload: Operators are inundated with data from various sources, making it difficult to extract actionable insights.
- Evolving Roles: The role of operations has shifted from basic task execution to a more strategic focus on customer needs and data utilization.
- Customer-Centric Approach: Understanding customer requirements and providing transparent data flow is essential for building trust and enhancing service delivery.
Key Strategies for Transformation
To turn operations into profit centers, the experts emphasize several strategies:
Automate Business Processes:
Automating data transmission and reporting can streamline operations and improve response times.
Benefits: Reduces manual errors and enhances efficiency.
Implement Asset-Based Accounting:
Understanding the total cost of ownership (TCO) is crucial for informed decision-making.
Key Metrics: Track maintenance costs, failure rates, and operational data to optimize asset management.
Leverage Predictive Maintenance:
Transitioning from reactive to proactive maintenance can significantly reduce costs and improve reliability.
Data Utilization: Use historical data to predict failures and schedule maintenance effectively.
Enhance Customer Self-Service Options:
Providing customers with access to real-time data empowers them to make informed decisions.
Transparency: Customers can monitor their usage and performance metrics, fostering trust and collaboration.
The Importance of Collaboration
The discussion underscores the necessity of collaboration between operators and customers:
- Shared Responsibility: Operators must communicate effectively with customers to manage shared resources and ensure optimal performance.
- Consensus Building: Achieving agreement among multiple customers in a co-location environment is vital for implementing new technologies and strategies.
Data as a Resource
The experts agree that data is a powerful resource that can drive profitability when managed effectively. Organizations must:
- Invest in Data Management Tools: Equip teams with the right tools to analyze and interpret data.
- Foster a Culture of Data-Driven Decision Making: Encourage all team members to leverage data in their daily operations.
- Embrace Continuous Improvement: Regularly assess and refine data strategies to adapt to changing market conditions.
As the data center industry continues to grow, the ability to harness data effectively will be a key differentiator for organizations looking to thrive in a competitive landscape. By transforming operations into profit centers, companies can not only enhance their bottom line but also provide greater value to their customers.
Full Transcript:
Stephen Warren, CTO of Data Center Dynamics:
Welcome, everyone. My name is Stephen Warren. I’m the CTO of Data Center Dynamics, and it’s an absolute pleasure to have you with us on this Management and Operations Channel where we’re really looking at this fireside chat, Dollars from Data: How to Turn Your Operations from a Cost Center to a Profit Center.
That’s something we’ve been dealing with all along in this industry, and we all know it—the amount of data is just getting incredibly large. How do we deal with it? It remains very much underutilized and, simply not used the right way if it is. But it remains a huge opportunity for all of us, and I think this is really important.
We’ve got two of the most wonderful subject matter experts joining us: Dan Moore, Senior Director of Business Operations at CyrusOne—a global, wonderful player in this industry who has helped so much in educating us. We also have Stuart Chambers, Director of Strategic Services at MCIM, which is a Fulcrum company—please note that.
And I would like to say, from all of us at DCD and the entire community, thank you so much to MCIM and Fulcrum for actually sponsoring this.
I’m going to turn it over to you guys right away. The first question that comes up, of course, is: How do you turn operations into a profit center? You guys even said it when we were chatting before—you’re dealing with so much data these days. How do you even deal with it? I mean, structured data, unstructured data… and I even heard someone say “quicksand data.”
So with that, Dan, can I please turn it to you first, sir? From the CyrusOne perspective—and you, you’ve had a lot of experience. How do you do this? What’s your perspective on it? Because it’s extremely complicated.
Sir, to you, Dan.
Dan Moore, Director of Business Operations at CyrusOne
It is—and good morning, Stephen. Thank you for the opportunity to join this panel and this conversation, as it’s extremely appropriate at a time of where we are within the industry and how we move forward with this explosion of data that we have at our fingertips.
Operations was historically thought of as just folks who did tasks and walked around buildings. That has evolved—and it’s had to evolve. We’ve had to get smarter, we’ve had to gain enhanced training and skills, and we’ve had to really put our customer at the core focus. We always have, but even more so today.
We now look at what we can do with this data to help our customers and our environments be smarter, and to allow them to operate more effectively—and with safeguards and parameters around what they’re doing.
So really, that’s one of the key things I’d like to cover today is just that.
Stephen:
So Dan, I’ve gotta ask you, you’ve got a tremendous number of customers globally, the biggest to the most critical. Is this something that you’re hearing from your customers also, Dan?
Dan:
Yeah, it is something that they’re very interested in, the data. A lot of it is, I know we’re going to touch on SLAs and transparency.
Even when we talk about SLAs, right? They’re legal parameters that are put in place, and we used to provide them just to prove that we didn’t violate something—around temperature and humidity, or access, etc. But what it’s really evolved into is more of a transparent partnership of data flow.
We want to make sure they have the data they need to make their decisions, as well as being transparent with them about operational parameters within their environment.
The data requests are tremendous. They also used to just be stagnant Excel reports—“Here you go, here’s a little temperature and humidity chart. See? We didn’t do anything wrong during this time.” But now it’s evolved.
We want to be more engaged with them, explain operational things and situations, and help them understand what’s going on in their environments.
Stephen:
Stuart, I don’t even know how to begin where Dan left off. You’ve got to be smarter, you’ve got to be more transparent. I mean, he hit it all.
To you, Stuart, what are your thoughts on this in terms of what needs to be changed, how are people reacting to this, do we even understand it? Is there some idea that we’re just doing it wrong?
Stuart Chambers, MCIM by Fulcrum Collaborations
You’re exactly right, Stephen. Dan is a hard man to follow. But first, thank you for letting us sponsor this conversation—because it’s a real one.
To your earlier point, I think you could ask any operator on the deck plate, and they’d tell you—they’ve been drowning in data in these data centers for the past two decades. From operations manuals to maintenance procedures to customer escalations—that’s what they deal with every day.
So, to all of Dan’s points, and to the broader question of how you turn an operations center into a profit center—it is going to revolve around the data. You have to be able to bridle this. That’s why I like talking with Dan—because Dan gets down to very specific techniques, and he already hit on some of them. I want to unpack those here.
You can’t get away from what I call the big three. The first of those is automating the workflow process—or the business process. As Dan just said, you have to be transparent with customers. They need to see the information to build trust. But there’s a time component to that—if they get it late, it doesn’t mean as much.
So, as one of the first tenets is that you have to automate the transmission of data within your business process. And Dan, I’m sure you could tell me—that’s not easy to do when you’ve got a million different customers screaming at you from all angles, saying, “I need it now!”
So that’s at least the first one, Stephen—you’ve got to automate the business process around your data. That’s going to be key to turning it into a profit center.
Dan:
Yeah, Stewart hit on it right. It really does come down to: how do we get that data? Data is everywhere—so how do we harness it? What is the appropriate data to focus on?
We’re going to get into all that and try to really navigate through it, because it’s a vast sea of complexities—if you let it be.
So yes, just again, automating things, and I know it’s not exactly “dollars from data,” but even automating simple things like rounds and readings makes a big difference.
Instead of the old days where a technician or operator would go around with a clipboard and a piece of paper—something no one ever looked at again after it was scanned and uploaded—now that data goes into a database, and we trend data.
Most of the data points we’re looking at aren’t even in our BMS, because those are already logged. But by capturing other parameters, we can identify potential issues before they become problems.
That’s the whole point of rounds and readings, right? To use all of your senses to ensure that the customer environments are safe and operating appropriately.
Stephen:
Wow, that was great. Customers and you, everyone’s impacted here. So let’s get into this proactive reactive element of how you even look at this. Asset based accounting, life cycle management, what are our goals in this? Dan can I turn it to you on this one?
Dan:
As I’ve progressed in my career, this category—this topic—is one that’s near and dear to my heart. As I’ve moved more from the day-to-day into a broader operational business view, it’s become even more important.
We’ve obviously partnered with Stewart and MCIM on this journey, and asset-based accounting was one of our primary focuses when we made that transition.
We need data—cost data, operational data—to allow us to understand how our assets are performing. And that, of course, ties directly back to the customer base.
Without knowing how much we’re spending, how many failures we’ve had, how much we’re spending on maintenance—both proactive and reactive—we’re kind of in the dark. We end up relying on platitudes, assumptions, or feelings. But in this business, in this world we operate in, we can’t make decisions based on feelings. Data is what drives those decisions.
Asset-based accounting becomes the foundation for a strong, effective lifecycle asset management tool.
Just because—and I’ll get into some examples—just because an asset hits a certain age, say the OEM says a piece of equipment has a 15-year end-of-life expectation, doesn’t mean you replace it wholesale at 15 years. And I say this with love for all my OEMs out there—this isn’t a jab.
What it means in today’s data-driven world is: we take that input—“Hey, we’re heading toward that 15-year mark”—and we ask, what has our spend been? How many breakdowns or repairs have we done? Then we use those multiple inputs to make an informed decision about how to spend our capital.
Asset-based accounting, through lifecycle asset tools, will also reveal when that “15-year” asset fails in year five. Based on the number of failures and the associated costs hitting your OpEx, you can make a smart decision to invest CapEx to replace that piece of equipment.
Because ultimately, that’s not only the smart business decision—it’s the best decision for our customers. We always keep them front of mind, ensuring that all our active equipment is operationally ready and meeting all redundancy requirements.
Stephen:
We’re going to come back to you in a moment, Dan, on this one—specifically, what you’re looking at. But Stuart, I’ve got to ask you the exact same thing: asset-based accounting, lifecycle management, overwhelming amounts of data… and everyone’s got a different business model. How do you deal with all of that and get to what Dan was just talking about—identifying the best data to drive value and make informed decisions? Dan said it perfectly.
Stuart:
He did, and his comments are not uncommon at all, Stephen. We talked about number one being automating your business process. Number two is automating your monitoring—and deciding exactly what you’re monitoring at the asset level. Dan hit this in spades.
We’ve all heard “what gets measured gets managed,” right? So, from that overwhelming wealth of data, you must distill what the cost impacts are. And you have to do that at the asset level. This is a 20-year problem for most infrastructure assets.
As Dan said, there’s a lot of front-end intellectual work—design decisions, equipment selection—that locks in your capital costs. But then there’s the other side of total cost of ownership: expense management. How is it maintained? How is it repaired? You have to be able to measure that not just at an asset level, but at a classification level. And if you’re really good, you can do it at a model level.
This allows Dan to make really smart decisions around things like battery replacements. For example, am I shifting to a new battery technology like lithium over something like VRLA? With the right data, over a certain period of time, he can prove to his customers that a technology shift brings cost benefits without sacrificing risk mitigation or redundancy. He’s got the data to back it up.
So, you’ve got to be measuring this. If you don’t, you’re flying blind when it comes to capital management—replacement cycles, and so on.
One last example I’ll bring up—because I’ve been through this decision with Dan a few times—is insourcing versus outsourcing. He’s got extremely capable technicians who know the plants like the back of their hands. But it’s still a financial decision. He has to know what the market can provide—what OEMs offer, how that compares to internal teams, and how to measure those support packages. That clear lens starts with asset-based accounting, just as Dan said.
Stephen:
Again, I’ve got to ask, Stuart—how do you get to that TCO thinking? You said it: what gets measured gets managed. What should we be monitoring? You’re looking at a lot of major companies. I’m going to ask Dan the same question, but first—how do I get to TCO, and what should I be measuring?
Stuart:
This is a really good question, and I think it has two parts.
A great example is predictive maintenance. That’s a big way to influence your TCO. But predictive maintenance isn’t prescriptive—it doesn’t follow a fixed set of rules. There are certain things you do measure, like transformer oil as part of a predictive maintenance package.
But the other side is looking at your data and identifying what’s an edge—what patterns emerge. That’s why we now have titles like “data scientists” on data center teams. These folks are trained to distill patterns and correlations from the noise.
We begin asking questions like: what is my failure rate after intrusive maintenance? Does equipment experience more failures after being shut down and restarted? Those are things someone has to analyze in addition to the regular prescriptive metrics.
Stephen:
Thank you, thank you very much—and by the way, thanks for bringing up data scientists and giving us all a heart attack. Finding them is like trying to find a cloud engineer!
So Dan, I’ve got to ask you—how do you even manage space, power, time, and cost to deliver client requirements? There’s so much to unpack here. I’m just leaving it to you. Please.
Dan:
No, yeah—I’d like to get to that. But first, I just want to touch on something Stuart said about TCO.
TCO is an important metric because it’s what our OEMs provide us when we make purchasing decisions. In my conversations with those folks, they rarely get positive feedback. Like most of us, it’s usually just negative feedback. And a lot of it is platitudinal—feelings and vague impressions. But TCO should be based on factual data.
If an OEM estimates a TCO over X number of years, and we can bring data to them—not just negative data, but positive data too—like, “Hey, we estimated failure rate would be X and cost would be Y, but our data shows it was actually better than that,” then we’ve got something valuable. That creates a full-circle relationship.
Now, regarding space and power—that’s a key area where data really helps us manage more effectively.
These days, while some customers are expanding, many are also consolidating locations. That’s a big movement right now. They want to identify what space they currently have.
And with all the new technologies coming out—densification, liquid cooling—we’re seeing changes. Traditional air cooling in co-lo locations can only go so far. But by leveraging liquid cooling technologies, we can extend the life and usability of those spaces.
I’m actually working with a customer right now who doesn’t need more space. They just want to swap out 15 kW racks for 35 to 50. That’s the evolution we’re seeing.
But to do that, we need to ask: do we have the cooling capacity? Do we have the chiller and power capacity? And not only that—how is it going to affect their neighbors?
We’ve got to be Mr. Rogers here. Everyone has to be a good neighbor. If we let customer X increase their density, does that drain resources from the rest of the property? Do we have the capacity to support that without negatively impacting others?
That’s ultimately what an operational team does. We manage all those parameters to make sure all customers are treated equally and aren’t negatively affected by their neighbors.
Data allows us to make those informed decisions.
Stephen:
Absolutely perfect. Stuart, I’ve got to bring it back to you—and by the way, I see you nodding, so help me out here. Go.
Stuart:
I could barely contain myself—because Dan just hit the third point!
Going back to our original question: how do you turn operations into a profit center? It’s data-driven decision-making.
And like Dan said, that’s not just internal—it’s not just an operator game between the OEMs and the maintenance teams. It extends to the customers. That’s the level it takes to stay competitive today.
Think about power reports, density reports, smart hands requests, cross-connects—customers are making those decisions daily. Do I go hyperscale? Do I stay in co-lo? What happens if I install a containment solution—how will that affect my neighbors?
Once you’ve automated your data pipelines, and once you’ve automated what you’re monitoring, you can make data-driven decisions internally—and then let your customers do the same.
That’s when the magic happens. When Dan’s customers log into his portal, they can see what his team is capable of, what their solutions are, and make smart decisions on the spot. That’s when things really start to scale.
It’s a big price of entry to get to that point. It takes a lot of foundational work. But it’s worth it. TCO is one of the best ways to prove that it’s working.
Stephen:
So, Stuart, I’ve got to ask—going back to Dan’s point about the layers: layers of cost, layers of efficiency, layers of need. He gave a great example of a customer moving from air to densification. How do you deal with all those layers of requirements, especially since every customer has a different point of view?
Stuart:
Yeah, I think we touched on it earlier—but you’ve got to give customers a time-sensitive view of the data. They’re going to judge how it impacts their deployment in real time. Rarely do I see a customer—Dan, you can chime in if it’s different—completely shift from air cooling to immersion cooling. It doesn’t just happen at the snap of a finger.
They need to build trust in that technology. They need to trust the partner who’s going to operate and maintain it. So, they start small. They run pilot programs. They move incrementally. And once they see the data confirming their assumptions, they move forward. If they don’t get that data from a partner, they’ll find someone who can give it to them.
So, to the earlier point—you’ve got to be looking at the data internally, and you’ve got to be thinking, “How can I pipeline this efficiently to my customers so they can make the decisions they need to make?”
Dan:
To your point, Stuart, yeah—they don’t just jump in with both feet. They do their due diligence. They run pilots.
I’m working with a customer right now on a pilot program, verifying the technology works and confirming that all the promises made by hardware vendors actually hold up in practice. But even throughout that pilot, it’s about sharing data. Everything that we anticipated and forecasted—when it occurs as designed, that’s what builds trust.
Stephen:
Fantastic. It’s so true. Let me jump to the next point. Dan, you set this up so well, I should be sending you a check.
Like Dan was saying, it’s a shared environment. There’s the customer’s needs, but also the operator’s responsibilities—CyrusOne’s responsibilities. I’ll throw this to you, Dan. You’re managing this shared environment and the shared issues that come with it, every day, around the world. Customers are constantly changing. There’s the intersection of business needs, customer requirements, SLAs—there’s a lot to balance.
Dan:
It is a complicated world. In a shared environment, you might even have multiple SLA levels. So, you have to manage to the lowest common denominator—you work to the lowest threshold to make sure everyone is covered. In doing that, others may benefit from that same level of protection.
To make sure we’re good neighbors, we leverage our BMS data, our EPMS data, to highlight issues within customer environments. For example, if a customer has 60-amp receptacles and each is pulling 35 amps, there’s no failover capacity. That’s a shutdown risk. So, we take that data and proactively alert them: “Hey, you’re trending toward a point where, if something fails—like a power supply or a component—it’s going to impact you.”
And if they’re overutilizing power, a failure might not just affect them. It could cause an upstream issue—a cascading failure that impacts other customers, such as through an RPP (Remote Power Panel) or a daisy-chained setup. So we have to protect not only that specific customer, but everyone in the data center.
We also monitor for zero-amp alarms. If a circuit was running at 15 amps and suddenly drops to zero—something may have gone wrong. We don’t always know exactly what’s happening in each customer’s environment, but we are proactively reaching out when we see anomalies.
We’re not just a landlord. We’re monitoring, supporting, and doing our best to keep everything operating as designed. That means alerting customers to changes they may not yet be aware of, so they can avoid broader impacts to their systems and applications.
Stephen:
I like that, Dan. You’re not just a landlord—you’re a true partner to the entire industry. What you all do at CyrusOne is just incredible.
Stuart, let me bring it to you—back to this concept of shared environments and shared issues. I like how Dan framed it—because impacts can happen upstream, downstream, and sideways. These issues touch every neighbor. How do you deal with that, especially with so many challenges at once? I’m surprised Dan still looks so young!
Stuart:
Amen to that! And I cannot overemphasize how intense those issues become, day-to-day. Having worked on white space floors, things like failover reporting—giving customers a heads-up when they’re approaching limits—or even just noting when a deployment is under-densified, are everyday tasks.
It’s a moving target. Customers rack and stack all the time. You have to stay on top of it.
But another key thing I’ve seen Dan do well—and it’s absolutely necessary in this space—is not just being a landlord, but being a mayor.
Let’s say a new technology comes out—a new UPS design, for example. If Dan can adopt it, it might lower costs for customers in a co-location environment. But to enable that, he needs consensus. That means getting buy-in from potentially a hundred different customers sharing that space.
And if you don’t have the data to prove the benefit, you’re not going to get them all on board. That’s why it’s critical for Dan to give customers access to the same real-time data he’s using. That transparency allows him to bring consensus to the table and unlock efficiencies in co-lo environments. And that’s hard to do.
Stephen:
Sorry to interrupt, Stuart—but this is something we’re hearing from the industry too. You’re talking about unlocking value, but so many customers are looking to Dan saying, “Help me with this. Give me everything.” So I’ve got to ask—how do we balance short-term and long-term solutions? Isn’t it a joint effort, with Dan and CyrusOne working alongside customers? They’ve got to share what’s happening, and customers have to share their goals.
Stuart:
You hit the nail on the head, Stephen. That’s exactly why I use the term “mayor.” Dan is in the middle of that ongoing push—his customers want to get more efficient, but also stay safe. And everyone has a different strategy to achieve that.
Those strategies intersect in the white space. Someone has to mediate—and good mayors like Dan do it transparently, with data.
When he implements a new strategy—like raising temperatures on the floor in line with ASHRAE standards—he can show everyone that it doesn’t impact reliability. That builds trust, because they can see it in real time through their portal. That kind of visibility makes it possible to evolve quickly.
Stephen:
I’ve got to ask you again, Stuart—customers want to reduce problems and risk while increasing efficiency. Dan’s trying to do the same. So how do you operationalize this when some customers aren’t ready? Or when they’re overwhelmed by data?
Stuart:
That’s a fair question, Stephen. And it goes back to something you said—showing them that the answers are already there. They’re just stranded in some black box or buried in a database.
That stranded value is what we need to unlock. It’s not some miracle feature of a software that tells you to “move your server to this position.” The answers are in the data they already have.
It’s about aligning the trajectory and building the pipeline—getting the right data to the right decision-maker, so they can act.
Stephen:
So, from MCIM’s perspective—you’re dealing with a lot of different customers and different data requirements. Is it possible to support all those varying needs?
Stuart:
Absolutely—100%. But the key is having a plan. If you don’t, the data will manage you.
You need to be proactive. Without a structured approach, your technicians will drown in data. You’ll see SLAs drop. But if you have a strategy, here’s what happens:
First, identify your key data points. Go back to the basics—automate business processes to exchange data quickly between stakeholders. Don’t let it get stuck in boxes or delayed in transmission. Automate.
Second, focus on what to measure. What’s actually important? Distill it down, get agreement on those metrics, and manage based on that.
Third, present it clearly—in dashboards, reports, alerts. Let people make data-driven decisions. The software doesn’t make the call—the people do. Executives making capital plans, technicians on the floor responding to alerts… they all need access to the right data.
For example, a technician sees that the chiller temperature is trending high. His mobile device notifies him—he investigates. That quick action prevents a potential incident that could impact every customer on the floor.
So yes—it’s not just about executives or customers. It’s technicians, maintenance teams, OEMs—everyone needs to be part of your data strategy. And you need a platform to make that possible.
Stephen:
Good thing you’re around, Dan! I’m going to hand it back to you.
This shared environment—customers constantly changing their requirements, new technologies coming in.
Dan:
Well, you know, going back to the “mayor” analogy—it’s about open communication. It’s that constant dialogue between the business and the customers. Where are they headed? What are their concerns and issues—not just in their environment, but in their business? How can we work together to meet those new goals?
For example, if a customer says, “We want to move to the cloud,” we can talk about on-ramps. If they want to densify, we can discuss our experience with the relevant technologies. It’s those ongoing conversations that keep us informed and also challenge us to go out and explore new technologies and situations. We lean on our industry network to understand how others have managed similar transitions.
That kind of collaboration, transparency, and consistent communication is how we keep the “neighborhood” running smoothly—everyone supported, and ultimately, happy in their environments.
Stephen:
Happy—and that’s no easy task. Stuart, over to you.
Stuart:
Not to cut you off, Stephen—dangerous, I know—but to make this real, a great example is something many operators are concerned about: stranded capacity.
No one intentionally designs stranded capacity. Your A&E firm didn’t mean for it to happen. But it becomes a consensus issue among customers. One customer might have deployed with a certain expectation or technology, and it didn’t work out. Now they’re underutilizing their allocation, while another customer in the same data hall is struggling to expand.
If you have a good “mayor” in the middle—someone who understands what each customer is doing and can say, “Are you using what you need? Could you scale back?”—and you’ve got the data to back that up, then you can repurpose that stranded capacity for someone else.
We see this often with contiguous space challenges, customers needing to move around in the data center. Having a good partner who knows the environment and can mediate those needs—that’s how you prevent stranded capacity.
Dan:
Yeah, and something we’ve talked about, Stuart, is how we communicate that to our sales team. How do we alert them to available power and space in a data hall, building, or even an entire campus?
Whether it’s for new customers or, more commonly, expanding existing customers, we need to know, for example: there’s 5,000 square feet in that hall, but it’s broken up—3,000 here, 2,000 there—not contiguous. Or, there’s 750 kW available in a hall, but it’s split across both sides.
So we use data to manage space and power across the environment and understand how to best match availability to customer needs—whether that’s new business or expansion within an existing cage.
Stuart:
Let me zoom out on that. What Dan’s describing is hundreds of customers in a white space, generating and throwing off data. Technicians are contributing to that data. Then Dan’s sales team is able to take that information and use it to engage new customers.
That’s what I call a complete plan for managing data—and empowering every part of the business. It’s not easy, but it’s absolutely powerful. That’s the value of doing it right.
Stephen:
Stuart, can I ask you a question? This is something we’re hearing a lot: the concept of self-service. More and more people are saying, “I should have access to everything. I should be able to make changes without asking.” Dan, they’re saying, “I don’t need to bother you, I want to do it myself.”
Even though CyrusOne is this incredible company that manages everything so well, this mindset is shifting. How do you deal with that self-service mentality, especially when collaboration is still critical? Isn’t it becoming an issue as customers ask for more—and risk getting overwhelmed?
Stuart:
It’s a double-edged sword, Stephen. It’s definitely valuable—customers want more self-service. They’re just as hungry for transparency and data.
It’s an easy sell to say, “We have more self-service options.” But the challenge is distilling everything we do from a data perspective into something customers can understand and act on.
Dan mentioned it earlier: in a shared environment like a colocation facility, you have thousands of branch circuits, all rolling up to shared UPS systems, shared switchboards, shared batteries, shared transformers—you name it. Navigating that data stream across all infrastructure assets is a serious challenge.
But when done right, it gives customers meaningful self-service options. For example, automated failover reports, temperature tracking, or a 52-week maintenance forecast. That kind of transparency helps customers plan around busy seasons and freeze periods.
But again—it’s not just a data science problem, it’s a people problem too. You need tools that help decision-makers schedule and plan impacts. And to cut to the chase—it’s worth it. But you need a specific tool to do this effectively. Otherwise, you’ll drown in data.
If you’re still trying to manage this in email or Excel, it’s not going to scale. You need something smarter to pipeline that data in a way that lets customers self-serve based on their unique needs.
Stephen:
Good one, good one. Okay—I’m being told we’re running out of time. So I want to hit this last point. Let’s open it up: data as a resource.
What do people need to know when working with you? How do you turn data from a cost center into profitability? At the end of the day, you’re saying “data is king”—so how do I get there?
And I love how you brought up data scientists early on. How do we even find people with those skills? Are we resorting to kidnapping now?
Dan, over to you. Any way you want to respond.
Dan:
Thanks, Stephen. One of the first things I want to circle back to is how customer needs have evolved.
They want more self-service. They want smart hands without having to send emails or qualify personnel. They want to upgrade rack power or request a cross-connect—through a portal. That’s the expectation now.
So we’re leveraging existing staff not just as “wrench turners,” but as trained professionals who can support those self-service models. Think of it from a customer’s view: it’s 3 a.m. on a Saturday. They don’t have staff on-site. For small and mid-sized companies, they may not have 24/7 resources anywhere.
That’s where we come in. Trained staff can respond—check cross-connects, reboot servers, swap out drives—all kinds of services. From the customer’s perspective, that augments their staffing. If we’re doing regular tape swaps or escorting vendors, they don’t need to keep staff on-prem full-time.
So, going back to our original theme: this shifts the model. It’s not just a cost center. It becomes a profit center—maybe not literally, but it reduces Opex for the customer. They don’t need to hire as many people to manage routine or even emergency tasks. And companies like CyrusOne can provide those services efficiently.
Stephen:
Absolutely fantastic. You’re really talking about leveraging and augmenting across the board—and yes, saving someone money is a kind of profit. I love this. CyrusOne is clearly a partner to every kind of business.
Stuart, I’ve got to hand it over to you. Dan started with “evolution.” I’m not sure if this is an evolution or a revolution.
Stuart:
That’s fair—and Dan brings up a great example where it’s really both. If he can provide a service at a lower cost and help customers save money, while still delivering high value—that’s a win-win.
When you treat data as a resource, you unlock those opportunities.
I’ve been inspired by what I’ve seen. Yes, we have classically trained data scientists on our team—but I’ve also seen a 21-year-old high school graduate operate a three-phase UPS at 450 volts. Those people are already on your team. They’re your operators. Your maintenance mappers. Your floor techs.
Once you give them the tools, they become data scientists in their own right. They take pride in making the right decisions, saving customers money, avoiding issues before they happen.
You just have to treat data like any other critical asset—like a chiller. You care for it, monitor it, maintain it, manage it. From selection, to pipeline, to who it’s going to—manage it all the same way. And the people already in your organization often know how to do that. Just give them the tools.
Stephen:
“They’re already data scientists”—I love that. Just give them the tools. You’re right—I started out as a third-shift engineer and wish I had all those tools.
Let’s go to final thoughts—key takeaways. And yes, it’s okay to say: “Don’t do this alone. Come to CyrusOne. Come to MCIM.” Dan, final thoughts?
Dan:
Just to build on Stuart’s point—data, from a broader business view, allows us to make informed decisions. And that’s essential for any business to survive, grow, and prosper.
We’re diving deep now—working with data scientists and business intelligence tools. It’s fascinating. We’ve got data lakes, data pools… things I’d never even heard of a few years ago.
But it all comes from the growth of data—from every sensor, every piece of equipment. There are so many inputs. It’s exciting to work with people—especially those on Stuart’s team—who take a vision, slice and dice the data, and come back with real insight.
The key takeaway for me is this: data allows us to focus on our customers, create transparent operations, and ultimately improve experience and performance. That’s what it’s all about.
Stephen:
Fantastic. Stuart—final thoughts? And again, thanks to MCIM and Fulcrum. You’ve been incredible throughout this conversation.
Stuart:
Thank you, Stephen—and thank you, Dan.
I’ll end simply: this is the tip of the spear. Everyone is struggling with this. Data centers are exploding, and the projections for the next decade are mind-blowing.
Solving the data challenge is part of that future. Everyone’s dealing with it. Reach out—we’d love to talk about it. If you need colocation space and a smart partner, reach out to Dan. If you’re trying to solve the data problem, reach out to us. We’re going to be dealing with this for the next 10 years—and beyond.
Stephen:
And longer than that! Thank you both, Dan and Stuart. This has been incredibly valuable. What a great discussion—so much insight.
And to everyone watching—reach out to CyrusOne, reach out to MCIM. They’re here to help. They’re your partners.
I hope today helped you understand not just what you’re doing—but what you’re doing right… and what you might improve.
