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

Okay. Well, good afternoon, everyone. Good morning to those on the West Coast. I'm Nick Del Deo with MoffettNathanson. And I'm thrilled to be joined by Jon Lin, the EVP and General Manager of Data Center Services at Equinix. This is the first chance I've had to meet with Jon. So I'm really looking forward to this discussion. Thanks for joining us.

Absolutely. Nice to meet you, Nick.

Yes. And if memory serves, you probably have some exciting legal news to share with us before we start.

That's right. 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.

All right. Great. Well, you heard it here first.

So to get started, Jon, you're an Equinix veteran, but it was only several months ago that you started this -- your current role, which previously didn't exist. So I guess to start, maybe can you give folks in the audience an overview of what your new position entails and why Charles and the rest of the leadership team thought it was important to create this role?

No, absolutely. In my new role, I'm focused on really driving the performance of the Equinix Data Center Services portfolio developing, leading execution on a comprehensive and coordinated strategy just across the business, right, really looking at it holistically end-to-end globally and making sure that we can continue to sustain growth and really enhance our operating leverage inside of those facilities. So I've got direct responsibility for the product management team and all of the associated software engineering teams that are supporting that portfolio and just partnering really closely with all of our global and regional leadership to just orchestrate all the resources across the company to continue our expansion, making sure we're continuing to extend our leadership, both geographically as well as with our critical customers and digital ecosystems. And also, we really look at how holistically we can drive scaling and efficiency programs in pursuit of our long-term margin expansion targets.

Okay. Great. A lot of key stuff that Charles has talked about in the past. I guess to dig in a bit, let's start with kind of the core Equinix colocation and interconnection business, that's the heart and soul of the company. The team has really emphasized the strength of demand for your offering as of late. To what do you attribute the strength of demand today? Is this kind of a post-pandemic pop as business kind of figure out what they need in this new future? Or is it something more durable and lasting than that?

It feels very durable, I would say. We certainly saw some amount of benefit during the pandemic around it, the need for remote work, et cetera. But the bulk of what we're solving for customers is this fundamental drive around digital transformation. And we're seeing that across enterprises across kind of every vertical, every segment. How do we use data more? How do we get in front of customers? How do we go ahead and build more efficiently? Like it really doesn't matter which vertical everyone knows that using data and kind of fundamentally transforming your business to be able to take advantage of that is critical. And our role in helping support that around helping customers get to the cloud, get to the multi-cloud helping customers get to more locations, so they have more ability to drive value off of the data sets that they need to and connect to more partners to be able to actually exchange that data efficiently. Like it's still like we're in the kind of no longer early stages. I'd say we're kind of like in the middle innings that what we're seeing is robust amounts of demand across all of the customer segments. And so our job is really, hey, how do we make sure we're focusing then our sales and marketing engine, but from a sensing perspective to find where the pockets of demand are. That's kind of most ready to consume that so we can kind of prioritize our cycles and also being able to drive and engage with those customers and then help them realize those results. So and you can kind of see that across the business as a whole. We've gotten, again, strong revenue growth across all 3 regions. APAC leading the pack there, but Americas again a double-digit growth and EMEA kind of snapping back nicely as we had expected there. We're seeing our same-store revenue growth picking up. We're seeing rising MRR per cab and really good mix of revenue and strong pricing trends there and churn coming down as well. So like all of the work that we've been doing around segmentation, targeting, propensity of customers and making sure we're going for the right use cases with the right set of clients is really kind of paying off.

Okay. Okay. And then Jon, you mentioned getting a better sense from a sales perspective in your answer. I guess can you talk a little bit about whether it's the tools or techniques you're developing to enable that?

Yes. I'd say it's across the spectrum. It's one, making sure we've enabled our sellers. It was a rigorous training program, rigorous set of just disciplined candidly around, hey, how do we make sure we're changing the pivot of the conversation from what we saw -- what is the customer need right in front of them now towards and how can we be a trusted partner for them and listen on like not just single quarter but a multi-quarter or multiyear basis, where are they in that transformation cycle so that we know exactly when to reengage along the way. I think a big part of that as well has been our expanded work with the channel, right? Our -- the relationships that we've been building with not just the client but then the clients trusted partners on the technology side, whether that's Microsoft or Amazon or Dell or HP or Cisco understanding where all of that kind of converges together and then talking about how we can help support that use case is incredibly important. And then on the kind of the tools and side, we certainly have been continuing to refine our data science around our clients, right, and continuing to build really a data science-driven propensity model so that we know this customer that's out there, this enterprise out in the field or they're in the locations that matter. They're using this set of vendors that we have affinity with. They're on this far along their transformation journey. This is kind of the proportion of spend that they're doing against digital transformation. We know them that they're kind of a warm target for us to go after. That's all part of the journey together. So our kind of conversion ratio then and kind of our cycle time can be optimized.

Okay. Okay. Great. Listen, one of the topics that I want to dig into that you brought up a few moments ago was hybrid multi-cloud because it seems like that's been a real key driver of demand. It's something that you've been positioning the company to capitalize on for many years now. It seems to finally be paying off in a big way. So I guess, where do we stand in terms of companies that can benefit from this sort of architecture, actually implementing it? How much runway do you see from here?

I'd say a lot of runway. I think it's -- while the conversation and awareness around digital transformation, it's like present at large. I think the actual implementation of that is still relatively early. And I mean even with larger companies, 40% of the Global 2000, nearly 60% of the Fortune 500 are customers of Equinix. We think all of them should be, right? And I think we will converge to that over time to be able to get that to kind of the 90-plus percent ratio there. I think enterprises at large or -- they're thinking through that strategy. As you said, we've been talking about it for years, but 4 or 5 years ago, there was still the question of is a customer just going to be on one cloud? And I think when we're talking to enterprises now broadly, that answers -- that question has been almost entirely resolved, and the answer is it has to be a multi-cloud strategy, whether that's for diversity of vendor, whether that's for just diversity of capability inside of those clouds. And then they're also using a lot of different SaaS applications, right? And so those SaaS providers as well are huge elements of our ecosystem. And so being able to connect to all of those providers via our Fabric is just a huge value add for them. So again, I think, again, relatively early stages, there's tons of opportunity, and we continue to see a backdrop of very strong demand and very strong [indiscernible]

Okay. Okay. Great. Is there a way that we should think about either the share of new demand or the share of deals or some other metric? Share of deals that Equinix is booking that are somehow related to hybrid multi-cloud.

Yes. I mean if you look at the Americas region, enterprise customers in general represented over half of the region's bookings in Q1. And the largest use cases we see within that enterprise segment and kind of what we lead with are -- is that conversation around hybrid multi-cloud deployments and then WAN optimization as part of that. And so as data is becoming more interconnected and distributed, customers are looking for the place to put that data proximate to the cloud, which is Equinix.

Okay. Okay. Now I think consistent with your answer, it seems like the Americas or the U.S. in particular, is really leading the way globally in terms of hybrid multi-cloud adoption. To the extent that, that's a real propellant of growth today, in the Americas, should we think that a couple of years down the road, we'll see a similar bump in EMEA and Asia Pacific?

Yes. I think, I mean we're seeing solid demand across all 3 regions. The U.S. typically leads the way on some of those adoption trends, especially with these large technology providers. But I think we're not seeing any hesitance in the other regions or the other client bases around the world around the same approach. So...

Okay. Okay. Now obviously, you would want every customer globally adopting a hybrid multi-cloud architecture to go to Equinix. Your share probably isn't 100%. To the extent that a customer doesn't choose Equinix, what's the reason? And I guess, as a follow-up question would be, what's the opportunity for you to improve your service to attract those customers?

That's a great question. I mean globally, we obviously have just exceptionally strong ecosystems and just like a leading share of the cloud on ramps. And so if a deal is global or kind of like multi-metro even in nature, we generally have a really, really high win rate because no one can match kind of the scale and reach of the platform. If a deal is focused on one specific metro though, we may or not win, right? There are certain markets where our competitors have strong ecosystems. It could be harder to compete against if they've got particular strength of density or strength of relationship inside of a particular market. But that being said, the customers that we're talking with and then also our entire focus from a marketing and sales strategy, again, is looking for customers that are going for that multi-metro regional or global deal. And that's why, again, we're spending our time predominantly focus that not for kind of a local business, local enterprise there. What we do instead is, we're enabling our channel partners with our story and then hopefully, we can capture that business via our channel.

Okay. And presumably, if a customer is going to you for a multi-metro or global-type deal, your offering is pretty unique. So your churn characteristic is going to be much, much lower. Your ARPU might be higher and so on.

Yes. That's right. Yes. I mean we've got better command and capabilities. And I think the partnership with the client allows us to then be able to leverage a lot of different of those attributes.

Okay. So you noted your share of cloud on-ramps that you host globally, I think we estimate it's in the 40% range. I think that's consistent what you guys share. That depth of on-ramps plus the foundation of networks that you host really what makes Equinix an appealing place for people to deploy interconnected infrastructure architectures. But at the same time, we see that the big cloud service providers are trying to -- as they should, trying to mitigate the risk of having too much power in your hands, concentrating too many of their on-ramps in the hands of one counterparty. So do you think the large CSPs are having any real success diversifying the facilities through which their customers access their platforms. Like we see them putting nodes in, say, a QTS or CyrusOne or some other player, but it's -- I wonder how much traction they're actually getting by doing so.

Well, I think there's 2 reasons they're doing now. I mean, one is -- they just need diversity from a kind of provider perspective or a facility perspective to make sure that their network is resilient, right? And I think like in any given metro that you're looking at, if I was the CSP as well, I want to make sure I was -- I had that level of diversity. And part of that might be for kind of control reasons. But I think a lot of that is just from a risk perspective, that's smart. But when you look at those other providers that you mentioned like QTS or CyrusOne, I think there's probably some amount of incremental traction that they're getting. But a lot of that business is very different than how we look at the market, right? We're really focused on driving as many customers as possible with kind of a very diverse but very optimized set of footprints into our facilities. So our campuses are -- any given campus, our customer accounted number in the hundreds or potentially thousands in some cases. When you look at those other providers, they're really focused on filling up their real estate as quickly as possible. So if they end up filling the building with 2 clients, that's actually a win for them. So it's a really different model, which is going to drive a little -- a very different density of customer to the CSP. So I think it's, again, they're probably getting -- satisfying the diversity requirements they want in terms of actually like net driving customer feedback and demand, I would just say we spend a lot of time with the CSPs and kind of their sales teams in the field together. We don't really generally see that kind of same engagement with some of those other providers.

Okay. Okay. Makes sense. Now again, you host, call it, roughly 40% of on-ramps globally. I would imagine that the share of, say, traffic going to the CSP platforms, the share of connections going through Equinix to those platforms is much higher than 40%. Is there a way to conceptualize, one, do you believe that's correct? And two, is there a way to conceptualize how much greater that is?

I think that's right. And it's hard to say because most of that connectivity go it's all via private connectivity. So you won't see that in Internet exchange traffic trends, et cetera. But I think a good way to think about it is look at all of the counterparties that will logically try and exchange traffic, right? So look at the density of carriers, look at the density of other CSPs with the density of other like large content drivers that are in each of those markets. And then that will give you kind of a proportional share of how much activity is going to be driven in that secondary facility because while they do want resiliency and diversity around that, they're not going to be passing the same amount of traffic across both of those, right? There's definitely going to be one primary node where all of that traffic goes and it will be -- it will follow kind of the logical course of what were all the carriers, et cetera. So...

Okay. Okay. Now I think to date, the big CSPs, the Amazons, the Microsoft's and a handful of others are probably driving the bulk of business being steered towards those on-ramps. Are you seeing, call it, a second tier of CSPs develop that customers are more interested in connecting to? And if so, how is that mix of interconnections to those sort of second-tier cloud providers changed versus the past?

Yes. We're actually seeing that develop quite a bit. I think for us, for example, we've got over 30 cloud and SaaS providers that are running workloads that are able to be accessed via Fabric. And our use cases that are growing fastest across our Fabric platform, it's actually -- we're seeing a much higher velocity and growth rate around connectivity to the non kind of like big 3, big 5 kind of hyperscalers and more towards that diverse base as well as like new use cases just around partner-to-partner connectivity around that. And so you're really starting to see that benefit of essentially having kind of an industrialized Internet, so to speak, you get all the security, you get all the performance characteristics of private networking. But you can access any of these clients that are part of the Equinix ecosystem anywhere around the world. There's just a tremendous amount of value that they're seeing in that. And so we're starting to see a lot of those use cases start to open up.

Okay. Okay. That's great. Listen, cloud is obviously one of the most important drivers of your business, some of the most important ecosystems that you host but it's not the only one. I guess how would you characterize the health and growth of some of your more mature, more tenured ecosystems like network-to-network or content digital media network or others?

On the network side, it was a solid quarter with good momentum across all 3 regions. I'd say what we're seeing more and really encourages us again is the networks are not looking at us just as kind of the provider of choice for them to connect like their infrastructure into customers. They're also looking at it as a partner to do resell activity, right? And so we're seeing a record amount of network resale activity where they're bundling essentially our colocation offering or some of our digital service offering into their sale as they're trying to expand from just being the carrier to be more of a whole solution provider for their enterprise customers. Kind of digital media, again, strong bookings. You see the amount of broadcast that's going on and that pivot and move now towards streaming, like, again, we're seeing a lot of upside and benefit to that as that content becomes more and more diversified, right? And more and more streaming platforms are arising out of the market. Well, the route to actually exchange that traffic efficiently generally comes to Equinix. And so we've got more opportunities to engage with a more diverse set of providers.

Okay. Okay. Are there any...

I think the other one that's been like, again, really exciting to watch over the last couple of years has been the financial ecosystem, right? When you think about the trading environment and kind of that move now towards not just digitization of that and potential cloudification of that but also just a new kind of opportunities there, whether that's digital currencies, whether that's like exchange, like using blockchain for internal transfer purposes, et cetera, like there's just a ton emerging there. And it's been really exciting to watch that development. We still see a robust amount of demand and expansion globally in the form of the financial ecosystem.

Okay. Okay. You mentioned blockchain. Any other kind of nascent ecosystems that we should keep an eye on in the coming years.

I think the whole idea around Internet of Things and like how do we move towards smart manufacturing, for instance, or connected vehicles, like that's definitely starting to emerge. I think the consistent trend that we're seeing around that is there's a ton of data being created. That data needs to be processed relatively proximate towards where the location of that generation is, and then it needs to come back to the core to actually end up being modeled or integrated into other business cases and use cases around that. So that we're keeping a close eye on definitely have a lot of business development and sales activity focused around pursuing that and making sure on the early stages of that landing that and then like all that traffic and all those use cases and workloads and flowing back into and through Equinix.

Okay. Okay. I guess more broadly, one of the ways that you've been trying to make your product more appealing and easier to use for customers is through software innovations, perhaps most notably through Equinix Fabric, which you mentioned a few moments ago. Your virtual connections have been growing at a pretty healthy rate. When I think about the drivers of growth in virtual connections, is that a function of the CSPs kind of steering customers that way because it's easier for them to manage? Are the customers making that choice on their own? If so, for what reasons? Any thoughts you can share on that topic?

Yes. I think they're really complementary, right? I think when you look at cross connects versus virtual connections, it's just how large is the amount of traffic? How repeatable is it if it's a high volume of traffic and very steady, then a cross-connect makes complete sense and is probably the cost-optimized solution to that and performance optimized solution? But if you've got a dynamic environment and the customers are moving workloads or moving traffic flow between those end points and then kind of like depending on those use cases, a virtual interconnection might be better suited there. So we think of them as it's not one way or the other. I think having both is incredibly important there, and we want to make sure both of those solutions are easy to digest and consume by customers. So in our portal and all our software development work, like want to make Fabric look as easy as cross-connects and vice versa. And so we've been doing a lot of work on simplifying the user experience.

Okay. Okay. And again, maybe it's a bit of an artificial distinction since as you said, they're complementary. But should we look for a day when virtual connect ads are greater than physical ads? And if so, how far in the future might that be?

Yes. Just because of the law of large numbers, Fabric is gaining share or total interconnections relative to physical cross-connects because it is growing faster. It's hard to say if and when like that virtual cross connection addition will surpass physical. But I think that when you think about the unit economics and value that we're providing are really attractive on both, right? So for us, we don't really care if the customer is using physical or virtual. But I think the opportunity because of virtual connection can be so friction-free and like so API-driven and automated as we start to like really engaging more and more with the developer community at large and creating more and more robust software interfaces for that. I could see that being one where you could see very, very rapid expansion into that on the virtual connection side. Again, the risk there potentially might be that up and down. It could be transitory, but I think on a net basis, though, that's all going to be exciting additions into the flow and stream of things. And again, the standup and tear down for us is essentially free on a virtual connection. So it's actually -- like that would be great, right? I think as you can imagine, like going out and mining a physical cross-connect, every time we do that, it involves labor and lots of work. The more we can push some of those use cases towards virtual connections, the better for our teams.

Yes. You said that, I think about [ Chip Ones ] telling me about how you would audit cross-connect making acquisitions. And I could say the virtual will be lower appealing on that front. I thought one of the other interesting opportunities that I always thought Fabric might offer you is the idea of dropping nodes into partner facilities as a way to expand your reach without necessarily having to own the facility. If I'm not mistaken, you did a trial like that in Belgium a couple of years ago. So I thought maybe you could expand on that topic and whether that -- what that experience has been like in Belgium and whether you see that as something you could do more broadly?

Yes. I'd say, still a little bit of early days there. The progress that we've seen on that has been relatively slow. And I think we're continuing to refine our process with kind of go-to-market activity, et cetera. So we expect to see it like slow and steady ramping up that. I think -- but our focus really is how can we drive and meet the demand for our own data centers like across our footprint, right, whether that's in EMEA or APAC and Amer. And I think the opportunity and when we want to put the bulk of our effort isn't really going out and finding that value in those. Like third-party centers, that can come probably later down the road. I think there's still incredible opportunity and really driving increased attach rates within our own customer base and finding new use cases for that, that creates less friction and ultimately actually drives more value to our own platform.

Okay. Okay. Makes sense. And kind of sticking with this theme, from the start of Equinix, there have always been kind of other solutions that could potentially substitute for cross connects and other circumstances like transit or Internet exchanges. Over the last few years, we've seen stand-alone operators of fabrics like Megaport crop up. We see network service providers offering cloud connect or similarly branded services. Other data center operators have embraced some of the products that you've developed and offer. I think a lot of times, these alternatives just bring traffic into Equinix at the end of the day. So it's could still be good for your business. But I guess, kind of big picture, have there been any broader implications for Equinix from kind of the increasing diversity of ways that people could access clouds or access other partners?

Yes. I think part of it is it has reinforced our belief in and also the value of being a neutral provider around that. And so as you say, while there are other providers offering those services like it's great for us to be able to see that and also offer our customers, that diversity of access, and we never want to be in a situation where we're telling clients like the only way to do these things is Fabric, we want to make sure we're building a solution, though, that offers more value for them, right, and more -- and ultimately, the customers will decide around that. But the SDN providers such as Megaport to us, they look like carriers, right? They're facilitating remote access to network and cloud-based services. They're connecting them to the end users. They generally act like telcos for us. They're aggregating that traffic inside of Equinix and we're happy to work with them around that. A lot of those services though would be ultimately rely entirely on physical interconnection, right, which lives inside of our data centers. So as you said that traffic is coming back and making sure that we can then help facilitate and increase the amount of flow into those cloud destinations, makes it easier for the cloud providers to know it like this is a location where we're going to continue to be a scale and drive traffic and not end up having like a highly diversified set of facilities to do that. I think that creates a bit of a flywheel for us in driving more engagement with more CSPs and more capability there. But like I said, at the end of the day, we love Fabric. It's -- I think we're seeing a tremendous amount of adoption on the enterprise side there. We've got all of our customers are able to access like all each other, right, across the entire platform now and be able to deliver that in a software-defined like easily instantiated way. And I think -- between that. I think the proximity of being within us gives long-term fundamental cost advantage for a lot of our customers, like a lot of these connections happen within the same data center campus, less network equipment on our side or we're not leasing bandwidth, and we have control of those fiber routes, we can go ahead and have opportunity there to drive better margins and improvement there. And because those CSP on-ramps are directly at our sites, we also are more resilient, right? And we actually have lower latency, more predictable performance around that. There aren't any bottlenecks around the network that some of these providers may experience because they're oversubscribing their network to some degree, and they're going to hit some of those congestion points like we've got basically direct clear and free access into the heart of all the CSP's networks.

Okay. Okay. Terrific. Listen, one thing I want to pivot a bit and spend some time talking about some of the new digital services that you're developing, like Metal and Network Edge. I thought it was interesting on the last earnings call, Charles noted that Metal, your bare Metal service, which used to be packet, is really started to gain some steam. So I guess to start, what's driving customers to adopt Metal? What are the use cases that are really pushing them in that direction?

Yes. I think it's a combination of a couple of things. One is you're seeing people want to increase their agility and like reduce their life cycle like on technology management there. And it's also a fair bit of concern around supply chain constraints, right? And so the ability to get servers and networking gear and not just routers, but even like top-of-rack switches in all these markets globally, it's just gotten very challenging, right, whether that's because of logistics, whether that's because of being able to even buy some of that equipment from vendors. And so our ability to be an ad-scale provider having secured forward commitments and have a steady flow of that capacity available globally is something that is definitely an alleviator. And I think it's just a new way to consume Equinix's core value prop. I think as our teams have gotten much more comfortable with relaying that to customers and customers are like, Oh, this is still actually a physically like dedicated set of infrastructure that I can go ahead and trust as it's completely private, but it has all of the interconnection capability that I would expect at being at Equinix. It's a great starting point and a great ability to accelerate some of those deployments in the new regions. So we're really excited about that. I think we're still, I guess, early stages around that. And I think we are seeing like rate uptake and interest across our customer base, and it's certainly a key focus for our sales teams in the organization at large around how can we make sure we're supporting this activity? How can we help drive further adoption around that? And we see this as a virtual cycle, too. It's not a one solution fits all. Like we see Metal customers start as Metal, start as Network Edge, then convert and add-on and increment into colocation. And we see our current colo customers also looking at this and says how can I use Metal and Network Edge as an augment for some amount of capacity or quick deployment into a new market rapidly followed by additional colocation services. So it's really, I think a great cycle of just added value we can make to consume us.

Okay. Okay. Now how are you going about it from a sales perspective? Are your colo sales people trained to sell Metal or do you have Metal specialists? Or is it kind of a hybrid? How does that work?

Yes. Across our kind of our field sales and all of our sellers, they're really trained to sell the entire portfolio, right? And it's not just about any particular service, but it's also about selling the global platform, the capabilities there, the entire ecosystem, all of the offerings across the board. We do have specialists though because some of these solutions are highly technical. So for Network Edge or Metal, we've got an overlay team then that will go ahead in front and handle some of those questions for customers.

Okay. Okay. Got it. Now I looked on your website a couple of days ago, it looks like Metal is currently in 24 markets. So it's quite widespread. It's not pervasive in the footprint. Are you planning to expand that further? Or is what you have today consistent with what you need to be successful?

Yes. We're still expanding the footprint. I'd say that combination of Fabric and Network Edge and Metal will be available in 30 metros by year-end. That's kind of what we're planning against. But I'd also say, Fabric, for instance, that connectivity portion for all of our customers everywhere being able to interconnect is incredibly important. I think as we think about Network Edge and Metal, we want to make sure that we're also either we have that same intrinsic value prop that we need to deliver, which I think in many cases, the answer is going to be yes. But in some cases, the answer is there's going to be partners or we're not going to see enough market demand to do it at scale, and we can kind of make trade-off some decisions based on that over time.

Okay. Okay. What can you tell us about the returns associated with Metal? And as part of that, maybe talk a little bit about how you think of the returns to the extent you deploy Metal in an owned facility, the returns you would have received from leasing that space out in more traditional way customer versus putting your own gear in that edge and leasing it out.

Yes. I think we outlined this on Analyst Day, but the ROIC on our investment for Metal is as attractive as ROIC we expect to see on our own [indiscernible] assets, right? And I think the exciting part is over time, that Metal, we think will continue to drive significantly higher cabinet yields as that multi-tenancy around that single cabinet level is kind of even further ability to [ extenuate ] building on top of what we have as kind of our retail colocation offering there. And we don't think of it as like competing in terms of -- with our core colocation offering. I think there's also the opportunity to like drive yield at our stabilized assets because in any given metro in any given data center, there's a little bit of [ Tetrus ] being played in as kind of customers come in or leave and there's always that like last 5% to 7% that we're trying to think about how to optimize around. And so the ability to actually like deploy some of these services into that as our own and being able to kind of maximize then our full yield and that is pretty exciting to be for us to be able to do.

Okay. That's an interesting dynamic that I hadn't thought about the ability to fill in that couple of percent that would otherwise be hard to fill. You mentioned a little while ago that some customers are turning to Metal because they're having trouble from a supply chain perspective, which is good. So you're capitalizing on that. I guess, in turn, what are you seeing from a supply chain perspective? Should we expect challenges on that front to slow metals growth at some point? Or do you feel pretty comfortable?

Yes. We feel pretty comfortable. I mean, we definitely leverage our full scale and purchasing power with our vendors to mitigate what we saw as increasing pressures on the supply chain and logistics delivery. So we definitely secured inventory and kind of like stock availability, and we've got forward commitments out there from ourselves to our vendors and vice versa that give us a really good degree of confidence, but we'll be able to hit what we need to for both from a revenue perspective and a growth perspective in that we can kind of continue to drive growth ahead.

Okay. Okay. Let's talk about Network Edge for a couple of minutes, some of the virtual service that you're offering. How are customers using Network Edge today? And what's been stimulating adoption there?

Yes. So it's a -- we had a strong quarter I think when you look at enterprises, leveraging the service. We have over 270 customers now, I think, using Network Edge. And there's a couple of different applications or use cases that are really coming through loud and clear. One is cloud-to-cloud routing, right? So the opportunity to move data from one cloud to another having a virtual router deployed between that using Network Edge can seamlessly route that with ultra-low latency between clouds without having that customer needing to deploy like a physical colocation offering inside of that market, which might be -- it might be their primary location might be with us, with the physical footprint in Ashburn. They're using cloud regions inside of EMEA. Well, now they can stitch those clouds together without also having to do that. So huge value there. We're seeing a lot of branch to cloud kind of SD-WAN connectivity as well. So deploying some of these SD-WAN devices, adding connections with Equinix Fabric to reach their cloud providers and also securely connect back to their branch locations with our network ecosystem via Fabric there. And I think IP transit, we're seeing like just deploying a virtual firewall, right, connecting their customer applications hosted in the cloud to their own IP transit provider maybe defray some of those like cloud egress costs or just have more control over what network pads are being taken for that traffic. And the last is Remote Access solutions. I think this idea of deploying virtual firewalls and VPNs to be able to add capacity, enabling workers from home and doing that and being able to like basically using software again and being able to just go in and kind of define and create an entire enterprise WAN infrastructure set and be able to build that very quickly is just -- it's been a great solution. I think we're seeing some use cases now around M&A and divestiture to like, hey, how can I go ahead and create an entirely new network to be able to carve off and split a company into without actually having to buy gear, which, in some cases, customers can't do right now, right? I think that's another interesting area for us.

Okay. Okay. That's great. I think and as we think out over the next several years, what's your aspiration for these businesses? How material do you think that could be? And how much do you think they could help growth?

Yes. We're very excited about the opportunity ahead for all of our digital infrastructure services. And we haven't broken out specific assumptions relating to them, but you look at a great case study for us, which is Equinix Fabric, it eclipse the $150 million annualized revenue run rate in Q4 of '21 versus $100 million in Q3 of 2020. And now we have like 1/3 of our customers using it. So as we continue to drive adoption and attachment, we're actually continuing to see acceleration there.

Okay. Okay. Well, let's shift gears a bit. I believe global expansion falls under your portfolio responsibilities. You've always been kind of pushing into new markets and expanding your reach as part of what makes you different. One of the new markets you entered recently is India, which seems particularly interesting, big population, big software industry, increasing digital consumption down the list. And it seems like it's largely white space from a data center perspective. So I guess, kind of big picture, what are your aspirations for India over the next decade? And how big of a contributor do you think it could be to the business over that sort of time frame?

Yes. I mean we're incredibly excited about India, like we're experiencing good momentum in India with the GPX assets performing well above plan right now. We're investing behind that momentum. We recently announced our Mumbai 3 facility where we recently announced we're purchasing land for development in Chennai. And so when you look at what we have in terms of our assets today of the 2 data centers in Mumbai, having a really network dense campus there with more than 350 international and local companies and 6 of the on-ramps for the cloud service providers, we've got a great kernel of the ecosystem, and we're going to continue to drive more, right? I think there's just a massive opportunity in the Indian market. I think it's as you mentioned, we see it as quite a bit of white space with a huge amount of digital transformation being enabled by multinationals and also by like Indian domestic players that are obviously very large enterprise themselves. And so we're really excited about the growth opportunity there.

So you've got Mumbai in operation today, Chennai on the horizon. How many markets do you need to be in to serve a country like India? And how much capital could that soak up to kind of to scale that unit?

Yes. I think we'll probably guide towards more as you continue down the process. It's going to be, I would just say more, right? I think when you look at the overall India market, as you know, the population is huge. Geographic diversity is important and also having kind of reliability around regional coverage is very important as well as you think about kind of the transport and infrastructure in the electrical infrastructure of the nation. So we see a robust opportunity there. Again, we've obviously announced Mumbai 3 and we've announced the Chennai, but more to come over the coming years.

Okay. How do you feel about your -- how's your confidence in that you're going to become the market leader from an interconnection perspective in India on a sustained basis like you are in many of your other key markets?

Yes. I think we feel very confident. I mean that's why we felt so strongly about the team and kind of the assets that we got with GPX. It's a combination of having a really, really strong management team that's also been focused on how do we help the interconnection scale for the betterment of like the company or for the country and the company together. And so like that purpose of how do we help really create a fundamentally stable digital infrastructure for a country was like part of GPX' ethos and certainly part of some of these ethos and with the MainOne acquisition. And you kind of see a bit of that through line in some of the acquisitions we've done where we're not just buying kind of a random set of data centers. It's companies that are really thinking about this problem space in this mission in the same way of just really fundamentally enabling digital infrastructure for further countries. And so we get really excited about that opportunity. And so that's our entire focus, right? That's why we go to a new metro. That's why we exist in our existing metros. It's not about kind of just kind of cash flowing at or stabilizing units. We're always trying to push ourselves to look ahead with around the corner, drive more value for the existing clients and make sure we're securing new clients and building that robust ecosystem.

Okay. Okay. You mentioned the MainOne acquisition, and MainOne serves West Africa. Your peer Digital Realty has done a lot in Africa. They're in the process of acquiring a majority stake in [ Teraco ] in South Africa, which is sort of the interconnection leader on the continent. They've got a lot of cables that emanate from their [ Marseille ] properties, about a couple of other smaller data centers in Africa. I guess do you see a path to catching up with Digital Realty over time? And if so, how would you characterize that challenge for Equinix?

Yes, I think it's -- when you think about catching up, I would just say like South Africa is one market, right? It's a gigantic continent with a highly distributed population set and also infrastructure needs and geopolitical opportunities and aspirations around that. And so it's the right way to think about Africa is like you land a big flag and now you've gotten -- you've won. I think there's just tremendous opportunity ahead. I know we just see a huge amount of growth in a lot of different areas there. And so we're going to compete very aggressively in that market. I mean, obviously, South Africa is key market in any Africa strategy, and we're evaluating a lot of different entry strategies that we can go in and execute against. We see -- obviously, M&A is a part of our ongoing platform strategy, and we also think about different ways for us to go ahead and enter organically as well. So there's bid on both sides there. We want to make sure at the end of the day, we're being good stewards of capital, though. And so we've got quite a bit of discipline and rigor in our M&A process and our integration process. And I think one of the big differentiators that we always drive into all of our customers and the opportunity is when we go into these markets, we operate in the same way, right? We go ahead and standardize that portfolio. We standardized our operational elements around that. And customers really appreciate that. That's -- again, I talked earlier about we're focused around multinationals that want to be in these markets and want to do that work. I mean that's really important for their experience. And I'd just say our integration capabilities over the year is an incredibly well-honed engine and it's something that we take great pride in, in delivering that consistent experience is something that I'd say regardless of who the operator is we've put ourselves up on -- from a consistency of approach and operations perspective against anybody out there.

Okay. Okay. Are there opportunities to partner with some of the big cloud service providers that might be looking to grow in Africa, kind of use them as anchors for new facilities for yours?

Yes. I think whenever we're entering a new market, we've got a couple of different tools that we can use. It could be like as we did in some cases, M&A can be like new market expansion just through our own organic work like we did in Osaka, for instance, or in Melbourne. I think there's opportunity to really work with our customers and partners and say, hey, this might be an interesting area. We did that in Bordeaux, in Genoa. It's like landing an anchor with those cable landing. So we bring the entire bag of tools that we can. And now we've got xScale in our kind of bag as well, right? And so the opportunity to say, "Hey, is that something where we've got a kernel. We've got like a good team. We've got the construction capability now and we can go ahead and do them a larger deployment of capital for maybe on behalf of a specific hyperscaler customer." And there's a lot of potential and opportunities there for us to run with.

Okay. Okay. I think we have about 5 minutes left. And I think the topic that I want to focus on in the time we have left is edge. That's a potential driver of your footprint choices in the coming years. You may be the single company of size that has most benefited from the trend of distributed IT. And you seem to have taken more of a cautious or restrained view on edge compared to some other folks in the space or even more broadly, if you think about how some of the towers might be talking about the opportunity. So I guess, what signals are you looking for to decide whether footprint adjustments or other changes are needed to capitalize on a more distributed edge opportunity? And what can you do to prepare ahead of time?

No, it's a great question. I'd say what we've done is we've really continued to focus on our geographic expansion. I'd say when we think -- when we talk to our customers about Edge, we are kind of the best use case and best manifestation of Edge right now for most of our customers, right? We're in 70 markets with low latency performance for the vast majority of the population, serving all those use cases. And so that geographic coverage means like 80% of the population of North America, Western Europe and mostly Asian metros was within a 10 millisecond round trip of our facilities, which is great coverage there. And that ability to service the edge means like we've got things like Fabric and we've got Network Edge, and we've got Metal to be able to create diversity for our customers to go into all of these markets. I think that area around that Far Edge, right, in like, do we need to be in 200 markets in the U.S., for instance, we continue to have conversations every day, every month, every quarter with the large CSPs, the carriers, the enterprise customers to see if that's really emerging is something that customers want. And I'd say, so far, the demand of that feels relatively low. I think there's been opportunities here and there that we have looked at around, "Hey, does it make sense to do that deployment?" And part of it for us is the promise that we want to deliver to a customer when we start an effort is that they can expect that throughout, right? And so we don't want to set an expectation and say like, okay, we're going to build and do this one-off kind of deployment. Well, if they want to do 100 more of those, and we still don't actually see demand like manifesting in the market, like we don't want to disappoint that customer, right? And again, I'd just say market dynamics so far don't look like that a really compelling growth use case that we've seen. I know a lot of players have been out in the market for quite a number of years now, trying to find and test that demand and we have as well, and we've been like obviously actively listening and trying to work with customers to see what the right solution might be there. So it's not that we're not sensing there. It's not that we're blind to it. We're looking as aggressively as probably anybody out there around that but we are still waiting to make sure like, hey, that's a good use of both our capital and also good experience that our customers can rely on that we'll be able to take more broadly.

Okay. Okay. Got it. Recognizing that you're not seeing anything substantive today, are there use cases or applications that you view as most likely to stimulate that sort of far Edge demand or too soon to say still?

I think the one that resonates with me is this idea around smart manufacturing and IoT, I think kind of the requirements on latency to be able to be kind of at scale and then doing real-time manufacturing the latency characteristics are already extremely tight, right? Anyone who watch how it's made can tell you how quickly those machines are moving. And so being able to do a lot like of fault detection, et cetera, around that, it's a very tight latency window there. And that's interesting. But like when we've had customer conversations with manufacturing, for instance, that latency requirement is so tight that it actually needs to be in the factory floor. And so that may end up being what kind of the far Edge looks like, which is a deployment of some degree of infrastructure into that building and then tied back into one of our hubs.

Okay. Okay. I think we have time for one last question. I thought it'd kind of be an open-ended one. You've been at the company for over a decade. So you bring a pretty long perspective. I guess big picture, would you describe your advantage in the marketplace today is kind of being stable, expanding or moderating versus the past? And how do you -- how would you characterize your level of confidence in the business in the coming years?

Yes. I think the value prop we're offering our customers and the relevance and advantages are expanding. As the world's digital infrastructure provider, like, we're enabling our customers the ability to consume all of our value prop either on a physical basis or increasingly on a virtual basis with our digital service offerings. And so when you think about what that can mean for our clients for their users to be able to like engage with the developer community and really help transform from just having native software now in Equinix as completely relevant and understood all the way through the chain, I think that there's a ton of value that we can continue to expand on there. So really excited about the opportunity there. And I think the consistency of approach that we've done over the years of really being a neutral player to help enable all of this work around us and help connect all of these players and drive value is something that can't be understated. And I think the fact that we're there, we're a large public trusted player. They know who the provider is going to be 2 years down the road, it's going to be Equinix. We're not selling off our facilities. What's really important for us for them to know like, the crown jewels of their infrastructure assets are going to be with us. There's obviously a lot of activity happening out in the market. You hardly know who's going to be running your stuff in a couple of years, like they know it will be us, right? And I think that, that relevance and kind of advantage is going to continue to play well.

Okay. That's great. Listen, Jon, I thought this was a super informative discussion, and I really appreciate you taking the time to speak with us today. So thank you.

Absolutely. It's a great deal to talk to and just really excited about continuing the work in the year ahead.

Okay. Great. Thanks, again, Jon. Talk soon.

Cheers.