Equinix, Inc. (EQIX) Earnings Call Transcript & Summary

Okay. Well, good afternoon, everyone, and welcome to the 8th Annual MoffettNathanson Media & Communications Summit, our final session of the day. I'm Nick Del Deo, and joining me is Karl Strohmeyer, the Chief Customer and Revenue Officer for Equinix. Karl, thanks so much for taking the time to join us.

Yes. Thanks, Nick. My pleasure.

Yes. And Karl, I think you have some interesting news to share before we get started?

Yes. No. I'll read my disclosure statement first, which is some of what I will talk about today contains forward-looking statements. Please read our SEC filings for more information about factors that could affect these statements. So thanks for that, Nick. And we've had some fun with the Internet challenges here. So hopefully, this will work.

Yes. Yes. I was impressed with how you handled it on the fly. So...

Well, listen. Karl, I think it was a little over 2 years ago that you took on your current role.

Yes.

I don't think previously existed Equinix. So maybe can you talk a bit why it was important to consolidate sales, marketing, customer care and the other responsibilities you have globally under one leader, and the sorts of changes you've implemented and what the impact is that they've had?

Yes. No, thanks, Nick. I -- we -- as many of you know, not unlike many data center companies, we started from a geography-based position where we had very much of an org chart designed around geographies around the world. And we do P&Ls by country, by regions. And over the past, I will say, 7 years, has been my duration of Equinix, 7 years of my professional career. But that said, I'd say we tried very hard to accommodate the global nature of what our customers were looking for to get consistency of operations, consistency of how we went to market, consistency of brand and messaging and product deployment, et cetera. And it got hard. At the scale we were running, it became hard when we're sitting at 229 data centers around the world. We realize that a functional org chart may make that easier and it has. And so we moved to a fully functional org chart where the regions, all of the care, sales, the commercial decision-making, how we market our customers all roll up to me, and it's been great. I mean, it's actually been really helpful also to accelerate some of the transformation things that we knew we needed to do back then, and we've made a tremendous amount of progress on the last couple of years. And those -- think of those as getting consistent and better at the solution selling capability set that our customers are demanding us -- from us. We run an integrated sales play, meaning our sales teams out there represent the full portfolio of our capabilities, both from a geography standpoint and a service offering standpoint. And making sure they're enabled, they're equipped, they're focused on the right sets of customers, the right sets of new logos, with the overlay resources from a technical standpoint that we need in order to solve some of those business problems on behalf of customers. And so it's actually been excellent. It's -- we count our blessings because going through this pandemic, all of a sudden, overnight, having to stop the traveling and being home, I'm really pleased that we got through all of the org dynamics and alignment before then. And it made it much easier for us to talk consistently about how we are going to respond to customers during the pandemic with this new org design just as one example.

Okay. Okay. That's great. Kind of shifting gears a bit. I thought that one of the most encouraging things that you guys discussed on the Q1 earnings call was the strength in the Americas region, which you used to directly manage. Organic revenue growth expected to be in the 6% range or better over the remainder of the year. I guess, drill down a little bit there. To what degree is that being driven by gross bookings versus churn improvements? And has anything changed versus the past couple of years from a sales or bookings perspective to enable this uptick?

Yes. I mean, look, I would love to say the primary driver is just the fundamental changes that we've made around talent and alignment and clarity of who we're going after and the opportunity that's out there for us. And I think all of that played into it. And you can see that on the gross bookings line as that's continued to grow quarter after quarter coming out of the Americas, which we're incredibly excited about. But also, as you well know, we've been predicting that the Verizon churn, when we purchased those 29 data centers, we knew that Verizon had optimization to do as a customer on that platform, and we gave them space in the contract to do that. And of course, that showed up in churn and optimization that we help them with because the opportunity cost of using that space for more retail, interconnected-centric workloads is goodness for us. But that does put pressure on growth rates. And so I'd say it's a combination. We -- Jon Lin is the President of the Americas. He replaced me in that role. He's just done a tremendous job with his leadership team, making sure that they're focused on prosecuting the strategy. And we like what we're seeing out of that. So I think it's a combination of factors, but I like what I'm seeing, too.

Okay. Now I know you have to be careful when talking about projections. But do you think -- do you view that kind of level of growth, 6% or in that range, as sustainable over some reasonable period of time? Or should we expect that to kind of fall back just given the size of the business?

Yes. I mean, as you can imagine, we have our Analyst Day coming up here on June 23. So not to put a PR plug out there for everybody that's watching, but that is our chance to do a 5-year look of what are some of those underlying trends in the business, how will those translate into revenue growth and what we think is almost as important, if not, FFO per share growth. So we'll give you a better indication of that. But we like the trajectory. The things we've done are sustainable and repeatable, and so we're actually taking some of the lessons learned out of the go-to-market traction that we've seen in the Americas and applying them to EMEA and AP. One of the benefits of the new org chart is that we can get those best practices and rinse and repeat them far more effectively in this org model than we could before.

Okay. Okay. That's great. And just kind of tied into that point. The leadership team at Equinix has always highlighted the volume of deals that get exported from the Americas into other regions. Yes, how is that looking today versus the past? And what [indiscernible] are you seeing deals from EMEA or APAC flowing into the Americas?

Yes. I mean, no question, and you'll continue to hear us emphasize that. And back to my comment that we have an integrated sales force where every sales professional, all 650 of them around the world, represent the full portfolio. And then we have STAR, which we've talked about before, which is our targeting methodology around propensity. And so we've got our salespeople focused at customers that value our global platform, value the interconnection inherent on the platform and value the quality that is delivered once you're on the platform. And so you combine those and we're getting much better at selling the full portfolio across all 3 regions. And so the Americas, I think, has the benefit of American-headquartered companies have traditionally been far better at integrating services themselves around the world versus some of the other regions they've leveraged, SIs and some of the PTT partners, to help them when they go out of region. And so some of that dynamic is changing. And of course, our channel model is in service to that, and you'll see that growth through channel. But we're also seeing our ability to position the full portfolio in EMEA improving significantly quarter-over-quarter. So we will see more exports come out of EMEA and AP over time, I'm convinced.

Okay. Okay. That's great to hear. Kind of a bigger -- bigger picture question I get on the growth front is just, why is Equinix' organic revenue growth decelerated so much over the last several years? And kind of why is it coming in at maybe a shade below what management talked about at the 2018 Analyst Day? So I guess, in your view, are there specific aspects of the growth outlook that may have been, call it, harder to achieve than expected? And how do they inform your outlook and your priorities from here?

Yes. No. So big question. I think sometimes the law of large numbers is missed on people, right? Because we've been growing the business significantly, both organically and inorganically. And so maintaining a growth rate year-over-year when you get larger is always a challenge. But I would say the thing that's really key to understand is we could accelerate growth dramatically. But we would do so at the expense of returns significantly. And so balancing, making sure we're going after the right workloads that are durable and value that interconnection value prop is important. Making sure that they're priced effectively to -- in exchange for that value is important. And of course, the opportunity cost. And we've seen this in the past of going for growth and selling large footprint on the balance sheet in retail is great in the short term but can be bad in the long term because then it requires a lot of CapEx. And then, of course, the opportunity cost of not having that capacity to support the retail fill rate is a negative. And so trying to get that equation balanced so that we have long-term durable growth with customers that value the piece of the -- what we provide is important. And so there are aspects of the growth that I think could be accelerated over time, and we're working on that. Channel is a big one, as we've talked about many, many times. Good 30% of our bookings are, what we call, partner-attributed. It's very much of a cell with motion where we're out in the market with a Microsoft, we're out in the market with a Verizon. We're positioning a joint value prop to an enterprise buyer, which is really sticky, and we like over the long term. But what we're really working on is getting to a place where we can combine those offers, integrate them and make them consumable in a more automated way so that the sales forces of our partners can be a bit more self-sufficient than they are today. And that work is something that we are continuing to focus on. And we think that unleashes opportunity because, again, we had a 615 quota-bearing headcount, 145 solution architects. We don't have thousands of people on the street, but through the channel, it's a force multiplier. And if we can continue to enable them with those integrated value props, that will be goodness from a growth trajectory. And obviously, that's one vector we're particularly focused on as we have been over the past couple of years. And I think there's growth there that we need to untap.

Okay. Okay. And that's a topic I want to return to a little bit, too. From a pricing perspective, we've seen MRR per cab kind of hold steady or expand. But again, probably not at the historical rates that we've seen several years ago, particularly in the Americas. Should we interpret that as meaning your pricing is now closer to the, call it, the frontier of what's achievable than it used to be? Or should we think of other factors being at work, like mix -- customer mix or strategic decisions to drive share over pricing, things like that?

Yes. I mean, it's a really great question. And of course, all these answers are so multi-variable in nature, as you know. But look, we're expecting firm yield and we've been consistent with that, and that's what you've seen from the business. There are factors that change it from quarter-over-quarter, timing of churn, time of that installs, et cetera, which you know well, and that will continue. I think the largest -- the thing to really understand is you know our stat where I think we're up to 62% of our revenues across all 3 regions. 88% is across multi-markets. And so any one customer we sell, our expectation is we're going to upsell them incremental geographies and incremental products over time. I mean, almost up to 90% of our bookings in any 1 quarter is to existing customers. And so if we didn't have that mindset, we could mine for price. We could mine for margin at a particular data center site. But that would not -- that would be in conflict with adding that global value of the footprint for our customers. And again, we're going after customers that value that global footprint. And so yes, there are definitely -- to your point, I think you used strategic trade-offs. Those are definitely trade-offs to make sure that we're driving the value for our customers that is durable. And so there are puts and takes, and you'll see changes in any 1 quarter, but I would consider it to be flat going forward.

Okay. Okay. Okay. That makes sense. On your earnings call a couple of weeks ago, I thought Charles' comment that you snagged, I think '21 cloud on-ramp in the quarter was pretty interesting. I spoke with Chip afterwards, and he noted that, let's see, it's 40 metros, we now host at least 1 on-ramp from one of the big 6 cloud service providers, 31 markets with multiple on-ramps and 6 -- with all 6 of them present. How crucial is hosting more and more of these on-ramps to executing your enterprise strategy?

Yes. I mean, look, given interconnection is the focus of our strategy and always will be, and it has been since the beginning of time, we love ecosystems. And the largest one that obviously we're all interested in right now is the enterprise cloud ecosystem. And those ecosystems require having the sell side and the buy side, both, on the platform. And so yes, it's important to us. And as we -- as the clouds start to extend into more markets around the world, and we're in those markets, that will continue. We'll continue to get those and attract those nodes from an on-ramp perspective. Both because our enterprise customers are demanding it from the clouds and the clouds know that they can pick up the enterprise customers at Platform Equinix. And so they're very important. And so it's a focus with us and with the cloud providers. And obviously, it's a focus in the metros in which we're expanding.

If we look out several years, how pervasive will these on-ramps be throughout your footprint? Is it reasonable to think that at some point, you're going to have on-ramps from the major players in all your major markets? Or is that -- or would that be a bit much in the context of alternatives like using Fabric or what have you?

Yes. I mean, it's a -- the last part of the question was where I was going to go with the answer. It really -- we're seeing interesting use cases where because of Fabric and because of the software-defined nature of it, customers can take advantage of ecosystem density in a market that they're approximate to but they're not physically located in. And so that's goodness for both us and the cloud providers. And over time, when you see those trends, it helps actually the cloud providers localize their interconnection when they see a certain percent going between markets. And we share all that data with them from a planning standpoint. So I -- look, I think the net is it will continue to expand across the platform, the number of sites. And today, we're also adding new destinations. So in the last year, I think we've added 80 new cloud and SaaS destinations on the platform, which is a goodness, which means that the diversity of interconnection is growing. And the requirement to not just have 1 or 2 clouds, but to go to the other clouds, whether it's Google or Oracle or SAP or to go to the SaaS providers like ServiceNow, et cetera, we're seeing that trend across the Fabric, which I think is good diversification of interconnection.

Okay. That's great to hear. It's kind of a broadening of the base.

Yes.

I think if you look at the top cloud providers, I think you host something like 40% of our ramps globally. Is it your sense that you support more than 40% of cloud connections? Or do you feel like the CSPs have been good at kind of diversifying how the customers get to their platform?

Yes. I mean, I'll answer it in 2 ways, which is the best asked of them, of course, which is an obvious. And they do have diverse ways. They can -- carriers can directly interconnect. They can directly interconnect over us. And of course, a large lot of loads go over the public Internet as it will continue. I think that as we -- if you think about the value proposition that we have for any one of the large cloud -- large cloud providers, it's not just as an on-ramp and it's not, of course, xScale. We can talk about that as well. But it's really around this notion that we do a go-to-market alignment with each one of them in every markets we operate. And what's -- why they would do that with us? Why would a Microsoft sales team in a particular market want to partner with our sales team to go after an enterprise buyer, is because they know when their customers access their Azure or 365 through Fabric, they sell a lot more subscription revenue than if they go through an alternative means. And so that motion is acknowledged by them and by us, which is why we [indiscernible] we have this go-to-market relationship at a local level. And so I think that's just a nice proof point to say. They know that when customers access them via our Fabric or via our platform, it's goodness for them. And as long as that continues, I think the symbiotic dynamic will continue as well.

Okay. Obviously, that's -- hearing that you work together with the cloud providers to get customer is good to hear. On the flip side, they're known to be tough negotiators. They're kind of 800-pound gorillas in the market. If it ends up that a handful of large cloud service providers end up serving as the magnets to draw in a lot of your enterprises over the coming years, how is that going to influence, call it, your negotiating position and return profile of the business in a holistic sense?

Yes. I mean, to be worried about that, you'd have to have a strong belief that the diversification of destinations is not in the future, and we do not have that belief. So not to use a negative to answer the question, but we're seeing that diversification. We're seeing that the other cloud providers that weren't first movers like the first 2 big ones are gaining share, are gaining customers, are gaining applications, whether it's analytics or database or whatever is inherent with their value prop. And so that diversification will be goodness for the relationship because they're going to be want -- customers are going to want to be deployed where they can access those multiple cloud environments and the clouds are going to want to have access to those customers that are there. But there's no question. They are powerful negotiators, for sure.

Okay. Okay. Maybe switching gears a little bit. If I think about the broader interconnection landscape, I think Digital Realty has made a push to get deeper into interconnection over the last several years with TelX, Interxion, a few other transactions. It seems like they're taking very deliberate steps to build a global platform. The strategy they're pursuing is not the same as yours, but there's certainly more overlap today between your models than there used to be. So I guess, have you observed any change in how customers, channel partners are thinking about deploying the way they're working with you versus Digital or the perceived relevance of your offering versus theirs?

The very short answer is no. Interxion was a competitor in the region before, and they're a competitor in the region now. I think there's probably a tendency to go after larger footprints, probably as a mix, just given the ownership from Digital, which is understandable. But I think the underlying notion of the enterprise buyer, the ones that we're going after and we think have high propensity for our value, they want a global answer. They want a consistent operating answer, all the way back to the first question you asked, which is why did we reorganize. That's really important. I mean, the world is complicated enough than to have all multiple providers with different nuances in their offering across the world. And I think having that 229 data centers around 63 metros is a huge value prop to our customers because they know what to expect. They know the ecosystem density. They know they can get to their clouds. They know the performance environment. They know that Fabric will exist there. We've deployed Fabric, I think, now in 50 markets, growing to 56. And I think it's at 170 of our data centers around the world. We do that because of demand from customers, and that consistency is very important. The wrapper we put around that from the size of our sales force, the size of our presale technical capabilities, the size of our channel partnership relationships, the size of the customer care model that we support our customers, is hard to replicate. And so if you have a complicated enterprise architecture and you want to distribute your workloads towards the edge to gain that performance, there's not a lot of places to go to get that.

Okay. That's terrific. And I think that the benefits of consistency are probably -- it's probably an underappreciated nuance of the model. It was based on conversations I have. Yes. I guess, if I think about the coming years, one of the biggest opportunities for Equinix is probably pulling in a long tail of enterprises into your facilities. In the past, you've shown some big TAM numbers out there. I suggest you'll update those at the coming Analyst Day.

On the 23rd.

Yes. What are you seeing as like the primary use cases or architectures that are prompting enterprises to deploy in Equinix as opposed to other -- facilities operated by others?

Yes. I mean, it's the ones that you've heard about that have a lot of legs. It's network optimization. How do they go from their currently deployed network that's very hierarchy in nature, meaning they have a couple of destinations and they route all their traffic through those data centers for firewalls and then back out to the users. That's just not fit for purpose, for latency, for the number of applications, for cloud usage, et cetera. And so they come to us and say, help me localize my architectures in the geographies where I have users and where data is being created and consumed. And that's very much of an IT network optimization play. And we love that, hybrid multi-cloud. Exactly. I've got Cloudify. I know these apps are destined to this cloud. I think these should be to this cloud. I've got a bunch of workloads that are legacy that I'm not quite sure what to do with. But it's not enough for me to run my own data center. So help me do both. Help me consume cloud, localize the traffic and understand which workloads I need to retain in a private environment. It's distributed security, right? I've got -- work from home was a perfect example that really hit everybody. And then, of course, that hit network optimization. But I now have thousands of endpoints, maybe hundreds of thousands of endpoints instead of 10s to 20. And so how do I manage the ingress and egress of traffic in an effective way from a security standpoint? Distributed data, very similar. So those are kind of the applications that are generally we're seeing. And what's really powerful is back to something I said before, which is 90% of our bookings come from existing customers. We do a land and expand, and you've heard us say that forever, which is we start with these 5 nodes with an enterprise buyer and maybe the entry point is cloud or maybe the entry point is network optimization. And then we add capabilities over time, both in the form of geography and service offerings. And then we say -- then we get to a place where we're like, well, look, you have an ecosystem of customers and suppliers. Who are they? They're already on the platform, high probability. If they're not, we'll go attract them to the platform through Fabric. You can directly interconnect them and manage your own ecosystem associated with those partners and providers. And so we advance and mature the dialogue over time depending upon the entry point, which is ultimately your question about what are the applications.

Okay. That's terrific color. As fast as you've grown the enterprise piece of the business, a common question I get is, why hasn't it grown even faster?

Yes. I get that, too, sometimes.

Well, I guess, is it -- has that part of the business been growing consistent with your expectations such that the view of outsiders is a bit mismatched? Or have there been challenges to getting enterprises on board at the pace you expected, call it, few years ago?

Yes. No, I think the underlying nature of it is that it is a demand creation story, meaning we -- there's -- we have to position the value of a distributed IT architecture and then what you can get as a result of the interconnection counterparties that exist. And that takes time. It's not something that you just -- people intuitively know. Now we're seeing more and more companies realize that they need help and that we're the largest provider in the area across all 3 regions. And so they'll come to us with more open-ended questions. But for the longest time, we were creating the bulk of our demand out there through education and positioning. And then simultaneously trying to educate our channel partners to be as verse at positioning that value as well, and that took time. And I think those things are all coming together pretty nicely right now, where people understand the value of a distributed architecture. They understand the value of not owning their own data center. The cloud is either happening to them or they're intentionally gravitating to cloud and they need help. And the first thing they go to, is my infrastructure ready to support the complexity that's happening to it? And they come to us and ask for help. So I think the trends are culminating to accelerate at least the adoption and notion of the value, but it took a long time for us to educate.

Okay. Okay. That makes sense. Earlier, you talked a bit about channel partners and how they're becoming more important. It seems like you're making some investments in the business to support that growth, those investments currently weighing on margins. Can you dig in a little bit about some of the more specific changes you're making to facilitate working with channel partners, how long that's going to take, and how you think about the benefits coming out of that [indiscernible]?

Sure. Yes. No there's kind of 3 vectors of focus of the business as a whole. And believe me, this does play into the channel, and I'll bridge the 2. We -- and our kind of -- our business of space, power and interconnection, right, the thing we all know and love. We believe that there's opportunity to automate it. We believe there's opportunity to make it easier to be consumed. And we think that has a benefit both on the growth vector as well as the margin vector, right, to be more efficient in the delivery of every unit that we sell for our customers. On top of that, we also are building a set of digital services, and I'm sure we'll talk about that. But those are primarily there to abstract the need to physically deploy but still get the benefit of our interconnection geographically dispersed value prop. And so all of those are key components that are in service to the channel effort, meaning our channel partners today want to be able to API into our OSS stack and be able to quote our services, configure our services, order our services and then have those services delivered and project managed and statused for their end customer. We're working on being able to do that. And the area we're particularly focused on outside of our new digital services is our dedicated cabs. A dedicated cab that a customer deploys into our data centers around the world that is then interconnected to Fabric or cross-connects is the #1 volume of what I would call colocation services we sell today. But it's very manual. And so we're working on a way to automate that from end -- from the beginning to the end. And if we do that, then our channel partners are more capable of self-serving. And if they can self-serve, they can educate, deliver and do that without a lot of help from us, which means we can cover more opportunity and a larger list of targets over time. So you can see how automating those interfaces could drive scale. And so that's a key part that we're focused on to enable the channel. Our new sets of services, we think, are very channel-friendly. Because a lot of channel customers out there understand the cloudification of service offerings. And if you can go to a portal in minutes and spin up a server in a market and load your OS and applications, and try the performance benefits of being localized in that market, interconnected through Fabric, that's goodness. And so we're working on learning it ourselves from a direct standpoint and then making a channel partner-friendly as well. So we think both of those areas are important investments for channel partners.

Okay. Okay. That's great. And I'm glad you brought up the new services because that's just what I want to dig into next. That's been a key initiative for the company in recent years, Fabric, Metal, Network Edge and so on. Maybe starting with Fabric since it seems to be the most impactful and developed. You talked about cloud on -- accessing cloud on-ramps out of region, perhaps bringing on enterprise suppliers or what have you on to Fabric. What are some of the more important adoption or usage trends that you've seen with that service?

Yes. Great question. It's amazing how customers can teach you a lot based on usage and trends that we're seeing. We're seeing a lot of customers use it to connect to themselves. We're seeing customers use it for business continuity. We're seeing customers use it to interconnect, not just to cloud but to SaaS, but also, as you articulated, other ecosystem participants that are important to them, right, other suppliers or customers that are important to them. And so it's -- again, I think it's 170 -- I think I said that we have 170 data centers across the platform that it's available in. It is over-indexing in the interconnection growth across those BCs. I think we're over 31,000 BCs across the platform, and it's continuing to grow. And so I think we're learning from our customers those use cases. Our job is to make sure it's available. It's easy to consume. More and more of our partners are actually writing APIs so that their software-defined infrastructure can actually automate through Fabric to turn up capacity for multiple locations. I think that accelerates growth over time. And so we're -- it is the platform where we see the majority of interconnection occurring in the long term from us because it's just easy to consume.

Yes. Along those lines, do you see certain counterparties like big CSPs trying to steer their customers in the direction of virtual because it's easier to manage? Or are they still kind of letting customers decide how they want to connect to them?

It's a great question. When we get to -- back to activating the channel, when we go-to-market with one of those CSPs in a geography, if their end customer connects through Fabric, they're going to sell more subscription service. They're starting to push that. They wouldn't activate their sales force in Germany with ours, for example, to go after joint sets of customers if they didn't see that value prop. And so I think that's important. So if you could call that steering, maybe that is from a behavioral standpoint. Obviously, they'll take the traffic from any interconnection mode that the customer would want as we would as well and as we do support, but we like the trends we're seeing.

Okay. Okay. Let's turn to Metal. Yes. I think that's now in 15 markets or so. So you've been rolling that at a pretty rapid pace. You only bought Packet a year or so ago. Again, kind of recognizing that it's early, what types of customers are using Metal? And what are they using it for?

Yes, it's a great question. Again, we're learning. We -- I would say the most exciting thing for me, the person that's in charge of the go-to-market machine, is that we are very well positioned with infrastructure buyers, as you would expect, across the service provider community, across the enterprise community. And the good news is those infrastructure buyers, the value prop of Metal is resonating with them completely. So it's not -- and by the way, it is also the DevOps team that they have as well that it's resonating with. But what I -- I really love the fact that if our main goal is to help customers with their digital transformation, and we strongly believe deploying as close to the edge as possible is a huge part of enabling that speed for them, that if they physically deploy the core on -- using colocation and interconnection, they have now with Metal the ability to spin up an edge far faster and easier without -- with less friction to see whether or not the performance benefits at an application or a user meet the criteria to maintain that offering and so -- for that deployment. And so we're seeing customers try that and explore spinning up virtual instances, and we'll talk about that on Analyst Day with a couple of interesting examples of existing customers that you've grown here and love that we've advertised for many years. I think it's important to see how they're taking the new product portfolio. But there's also cloud-native customers that would never go to colo, right? They just wouldn't even think about physically deploying or ordering a server and signing a colocation contract. They're just not at that scale, or that's just not how they want to consume IT infrastructure. And so those users can just buy it without it being sold to them. And so we're learning how to manage those early adopters that spin up instances. We try to understand and care for them, manage their adoption and, over time, turn them into a committed customer, if appropriate. So it's -- yes, it's a new set of customers, it's new persona with existing customers, and it's existing persona within existing customers. Those use cases are across the whole -- that whole trajectory, which is why we're excited about it.

Okay. Okay. Do you see, call it, tangible examples of the synergy by virtue of having acquired Packet and integrated into your platform versus having, let's say, as a partner arrangement? Are you seeing that kind of flow through in [indiscernible]?

Yes. And we'll talk a lot about this because I'm sure you're going to want to get to the unit economics of some of our new service offerings, and we'll talk obviously a lot about that during Analyst Day. But by having our own space and power that's deployed in the right locations, that inherently has in the interconnection that you would want, that enables the servers, the Metal to be that much more powerful is a huge advantage to them. And then deploying those servers on our own infrastructure to enable that is huge. And I think some of the companies have struggled because of the cost of having to build that infrastructure out themselves while they're trying to be good at the abstraction, that software enablement layer, and trying to be both an infrastructure company and a software enablement company has been hard for some of those smaller companies. Hence why Packet -- hence why we bought Packet. And so the combination of our footprint and capability and interconnection with their software development prowess and the ability for them to orchestrate applications across the servers is pretty powerful. And so we're learning a ton. They're teaching us a ton. Hopefully, we're teaching them a ton. But yes, no, I think the synergies are pretty exciting.

Got it. And since you mentioned it, what can you tell us about the economics? Or do we have to wait until...

You're going to have to wait. Yes, you're going to have to wait. I will say that they're attractive, right? They're attractive enough for us to be excited about wanting to deploy it in many markets. They're attractive enough for us to be excited to have our integrated sales force position them now with our infrastructure buyers.

Okay. Well, I look forward to learning more. I think earlier in our conversation, you mentioned that the sales force sells all your products.

That's right.

Which is kind of interesting because key services, I guess, in some ways, different services, maybe they're different buyers or counterparties that your customers work with them. Have you had any challenges on that front? Or has it -- it hasn't been pretty smooth going so far?

No. There's always challenges, right? I mean, I think the adoption curve to have a portfolio that is becoming as broad as the one we have, for any one sales professional to be incredibly adept at knowing how to position all of those effectively is hard. The good news is they have a lot of help. And we've always been a huge fan and we invested early and many, many moons ago on the GSA or the Global Solution Architect. And I think there's 145 of them around the world. They are incredible weapons that the salespeople can then deploy to help go deeper around the technical applications and how a customer can use and leverage the capabilities over time. And of course, we've got some of the talent that we've acquired with these products like Metal that are helping with the positioning of that value for our larger customers. But there's always a journey on educating the broad sales force to understand how to be that solution provider, what sets of questions to ask and then how to apply our value at the appropriate time for -- based on the readiness of the buyer. And so we're on a maturing curve just like everybody else. I'm proud of the -- of where we are on that curve, but we've got more to do, for sure.

Okay. Okay. Let's talk about xScale, another major initiative for Equinix. Customer demand appears to have been strong, given some of the stats you've thrown out there. Obviously, large footprint deployments tend to be a lot more competitive and at lower returns in the core retail business. That's part of why you're doing off balance sheet. I guess in light of that competitive backdrop, how do you feel about the returns you're able to achieve versus others catering to large footprint deployments? Do you feel like the proximity to your interconnection points drives a little bit of a better return than, call it, average for that niche -- for that segment of the market?

The -- you almost answered this question with the qualification, which is because it's off balance sheet, right, it's -- yes, obviously, we want to outsized returns for both ourselves and the JV partner. But the real point in doing this is to have an answer for some of our most important customers around the world, and to have that answer, to your point, be accretive to our interconnection campus that exists today without having the dilutive effects from a return standpoint and having to have to compete with capital that we need for the retail deployment. And I think we've struck a really nice balance, and it's taken us a while to really get the machine moving, but you can see the momentum in just the announcements we've made. I mean, I think since our call, we've got 2 JVs we'll develop over 290 megawatts worth of -- just over the last several months, we preleased the London 11 asset. The Dublin 5 campus is nearly 40 megawatts of capacity, and that's fully committed. So we're starting to see the momentum of it occur, and that the returns are good enough that our partner is excited to continue to underwrite even more. And of course, we've got our fee structure, which I'll let Keith explain how that all attributes to the P&L or not. But it's most of them -- the thing to remember is that it's a high-quality built, purposely built for the hyperscalers, proximate to our interconnection campus, off balance sheet, and it's accelerating, which I think is greatness. So that tells you that the parties involved are excited about the proposition.

Okay. Now I think your initial plan with xScale, if I'm not mistaken, was to focus on, say, a dozen customers or so. Is that still the plan? Or are you contemplating logging that to the customer base?

That's still the plan. Yes. That's still the plan.

Okay. Let's talk about India for a minute. The GPX deal should close pretty soon. I think a lot of customers wanted you to enter India, which kind of got you there. It's soon to be the most populous country in the world, but it's still pretty early in its development cycle, maybe earlier than any other country in your footprint. I guess, as we kind of put all that together, how should we think about the contribution of India over time? Do we think of it as being a material contributor in the next 5 years or so? Or is this kind of a longer-term journey for you guys?

It's a great question. I mean, I can't -- I have to preface by saying my heart goes out to everybody in India right now with the struggles. I mean, it's tough. I'll balance that by saying, if I could find more GPX in more emerging markets, I would jump at it. It's an incredibly well positioned asset for our value prop around interconnection, as you know. It does come with just 2 sites in Mumbai, and I don't want to just say just 2 because they're important sites. It's got 130 ISPs and 30 network service providers in the sites. And the demand has been the most exciting for something we've yet to close. In fact, we're even funneling deals to them as we speak, and we'll continue to do that. But I believe that strongly, that both either inorganically or organically, it's a stepping stone to far more investment in the country. You're right. India is early days from a maturity around data center and infrastructure, but we think it will be a large contributor to our growth over time. What that window is, it's a good question, I don't know. It really is going to depend on our ability to expand and add capacity at a rapid rate, given what we're seeing from pipeline and demand right now. But I think it's -- again, I wish I could find more of them.

Well, that's great to hear. Bodes well for the future. I guess, maybe, let's talk about EMEA for a moment. I think it was a couple of years ago, you guys harmonized and raised prices for cross-connects in that market. You're almost through reaping the benefits, almost on reaping the benefits of those changes. I don't think you have any immediate plans to do that again. But the monetization rates for interconnection there seems well below that of your other regions. I think part of that is because of the IXs and how that's influenced the market historically. How is the -- I guess, how is the tension between, we'll call it, using IXs versus using cross-connects, one-to-one cross-connects evolved over the last several years? And do you see the marginal interconnection more likely to go over a cross-connect today more than it used to?

It's a good question. I would answer it this way, which is we're seeing growth across all of our interconnection platforms in EMEA, whether it's Fabric; whether it's the IX, we operate the largest IX, as you know, in the world from a counterparty and traffic standpoint, IXs, I should say; or its cross-connects. And I think that continues. I think various segments will prefer different platforms for interconnection, depending upon what they're trying to accomplish. Our job is to make sure we've got the right interconnection mechanism for all the various use cases and applications. So I think we're focused on all of them. I am not close enough to know whether one is over-indexing in the near term over the other. They're all continuing to grow, which is goodness. And I think our job is to make sure that we've got the right platform for that interconnection to occur.

Okay. Got it. I thought we're almost at time. I thought I'd throw in kind of one random question before we close out. If I look at the top data center markets in the U.S., the only one where you don't have a presence is Phoenix. Obviously, it's been more of a hyperscale market historically. I'd imagine a lot of the associated interconnection is maybe in California or Texas. But Phoenix has been -- a lot of people seem to think that's going to be like the Ashburn of the West [indiscernible] years. Is there an opportunity to develop another key U.S. market for you guys in Phoenix or the dynamics there are such that it's not likely to play out?

It's a good question. I -- we will never say never because we're always going to make sure that we're going to be where most of our customers are going to demand our value prop. That has not been a market where most of our customers have demanded our value prop. That doesn't mean it won't. It doesn't mean the growth dynamics associated with power, the cost of real estate, et cetera, isn't attractive enough to create incremental density there outside of hyperscaler, outside of wholesale and/or landing stations in Hillsboro. And so we'll keep an eye on it. I -- what I would argue is there are so many other markets in the world that have a growth rate that probably is a bit more attractive. And if you think about the markets in Asia where we're not, whether it's the Philippines, whether it's Vietnam, albeit small markets, they're -- the ecosystems are being created around them because of some of the constraints perhaps in Singapore. And so our focus is to make sure we're listening to our customers. They're telling us predominantly where they need to go. That so far hasn't been on the top of the list, but I would never say that it wouldn't be.

Okay. Okay. That makes sense. Yes. Maybe as a final question, I'll kind of give you a big picture qualitative one. I think you've been part of the leadership team at Equinix for a better part of a decade. How do you feel...

Call me old.

How are you feeling about the position of the business today, the level at which you're executing and the opportunities you see ahead versus years past?

Yes. It's hard for me to articulate this on a video call, but I'm more excited today than I was 7 years ago when I said yes. I think that the broad demand trends are in our favor. I think our ability to execute around the initiatives that we're talking about around scale and growth, we're more -- we're better situated to do so than ever. I think our balance sheet with investment-grade and the ability to even raise more debt if we wanted to, from a growth standpoint, gives us ammunition that we haven't had in the past. And I'd say we're in the best position, I think, we've ever been. And we've got the ability to continue to scale. And I love the go-to-market scale that I'm seeing. I love the productivity on a per head basis I'm seeing. I like the underlying trends, both with our existing products and the promise of our new products. So I think we're in a really good place right now.

Okay. Listen, I think that's a great note I want you to close.

Thank you.

Karl, again, thank you so much for taking the time to speak with us. I thought it was a very informative discussion.

Thank you. Appreciate the time.