Kevin McVeigh, part of the UBS research effort here. I think, Amar, this is the second year we've had you here, so we're thrilled to have Amar Maletira and the broader Rackspace team. For anybody in the audience, if you have any questions, there's going to be a mic in the room. But anybody in the audience, if you want to email me, Kevin.McVeigh, M-C-V-E-I-G-H at UBS dot com.
I'll definitely have the questions answered. I'm here to introduce Amar. I think he's going to read us Safe Harbor real quick, and then we're going to get right into the Q&A.
OK. Thank you. Thank you very much, Kevin. So let me do the most interesting part of this presentation here. So I should have brought my glasses.
That's all right.
Rackspace may make certain comments today that may be forward-looking. These statements involve risk and uncertainties, which could cause actual results to differ. A discussion of these risks and uncertainties is included in Rackspace SEC filings. Rackspace assumes no obligations to update the information discussed today, except as required by law.
That was fantastic.
Great. OK.
So Amar, and I've done this with a handful of the companies today. I want to go back and help just frame the Rackspace story a little bit. But I even want to go back a little bit further from the time you were public the first time, went private, and then came back out. Because I think the evolution of the company, you've gone through kind of a couple of strategic reviews.
And I think since you and Mark have really come over, you've seen a credible amount of stabilization and execution that's been culminated by the debt restructuring, which has really helped stabilize the go-to-market motion.
But I think it's important for the market and anyone online or any audience to kind of help understand where the initial initiatives were, why the resets occurred, and why we think for sure you're on a path to success going forward.
But I think it'll help frame just a little bit how we got to kind of where we are and some of the decisions you made before we kind of talk about kind of the private versus the public cloud, the more current business mix. So, we spent a couple of minutes there.
I think it's important because I think it really helps crystallize why you made what are clearly the right decisions for the business longer term. What drove that decision?
Yeah, well, thank you very much for having us here, Kevin. So I took over as CEO, as you know, in late 2022. I've been with the company for I was a CFO prior to that. I was not there as a CFO when the company was taken public in June of 2020, roughly. And the company has been in business for 25 years. Right?
25 years, whenever I go to customers, I don't have to explain who Rackspace is. They know who Rackspace is. Right? But they don't know how we evolved. And if you just think about the company, 25 years of very rich history of innovation, one of the first companies to put customers who didn't have a data center in data center in 1998, OK?
One of the first companies to coin the term Fanatical Customer Experience, which was very important, that stays with us as a DNA today, and that's one of our key differentiators. One of the first companies to host VMware on private cloud, VMware, right, and today, we are one of the largest hosted VCF providers in the world, and we are the largest CSP for VMware. One of the first companies to launch OpenStack, right?
Today, OpenStack is in demand again, but OpenStack, working with NASA, and we were actually growing even faster than Hyperscalers between 2012 to 2014. A nd then today, we have evolved into what I call as a hybrid multi-cloud and AI solutions company, so we do only two things very well. We do cloud, which is public and private, and we do AI, so we are in very good markets.
I couldn't have asked for a company in a better market. Right? And the market is growing because workloads are moving from customer data centers. They have to land either in public cloud. And now many customers have realized that not all workloads are suited for public cloud.
And those workloads in regulated industries, and we'll talk about some of the vertical bets we have made, like in health care and BFSI, et cetera, and workloads that cannot be refactored need to find a home.
And that's hosted private cloud. And so we catch the workloads on both public side as well as private side, as well as at the edge. And AI is a net new workload that didn't exist before. And so that is a massive TAM expansion for us. Right? So that is what Rackspace is all about.
So let me tell you how we have transformed the company and how in the process of transforming. So two years ago, when I joined as a CEO, I had three priorities. And I talk about it in every earnings call. Number one, priority number one is turning around, doing an operational turnaround of the company. Priority number two is being a technology company, catching the next big wave in technology, which is AI.
And number three is right-sizing our capital structure. So, on the first one, on the turnaround, we laid the foundation in 2023. We set a clear vision, clear strategy, clear direction for the company. We restructured the company, created two business units and operationalized the two business units across public and private, refreshed the talent both at the leadership level as well as various levels in the organization, including sales talent.
About 50% of the sales organization is new. And it was a very deliberate process in late 2023. We invested in new offerings, so to speak. And we said, in public cloud, we want to ride the secular growth wave with these hyperscalers. And we want to lead with services.
That was a hard pivot we made. Leading with services and attaching infrastructure to it, as opposed to in prior years, it was infrastructure-led motion, assuming that once you land with infrastructure, you can expand into services.
We pivoted that completely. Right? So that is number one. And number two, in private cloud, we said, listen, "we don't want to go after all the workloads". Right? Because workloads will go to public cloud. We want to go after workloads that are suited for custom cloud, which I call as a private cloud.
Those workloads are in regulated industries like Health care, BFSI, workloads that cannot be refactored, workloads that need security and compliance. And so, we made some bets on verticals. We made a bet on health care vertical. We made a bet on BFSI, banking, financial services, and insurance. And we've made a bet on Sovereign. And that bet has worked out because we've been executing against that strategy for the last 18-24 months.
So for example, in Public Cloud, as we refresh the sales organization and as services-led motion, the last 3/4 , which is bookings is a leading indicator, bookings have been growing sequentially as well as year on year the last three quarters. Because remember, we were making all the changes in 2023. And now the sales force is now hitting the productivity curve. Right? So we saw a good jump in bookings.
We signed a major agreement, a strategic agreement with AWS for the next three years that gives us a lot of funding for MAP, which is Migration Assistance Programs, and so on and so forth, including AI. We also, for the first time in many quarters, we saw our services revenue flatten from Q2 to Q3, first time in many quarters. So these are some leading indicators, so to speak.
Our services attach has gone up significantly. Right? Roughly 65%-70% of our bookings are now services. Out of the 28 large deals that we signed last quarter, 23 had services attached. Right? We believe that we can attach 20%-30% of services for every dollar of infrastructure at the point of sale. And that should grow to $5 or $6 for the dollar of infrastructure over the life of that infrastructure.
And this is validated even by AWS. So, the opportunity to lead with services and drive value even on the infrastructure side is humongous. And all hyperscalers love it because now you're driving more volumes to the infrastructure. OK? So Public Cloud is on track. And we are very pleased. And we have people with digital services transformation in the organization.
On the Private Cloud side, we made some really good bets. We invested in new products. We launched our SDDC product line, complete product line, refreshed it. And we also made bets in Health care. Health care, we made bets in provider. And within provider, we said we want to go host and manage Epic workloads on our health care cloud. And that is, I'm very happy to report that we have made tremendous progress. Right?
That business now, which is roughly 10%-11% of our Private Cloud revenue today, which was less than 5%, is now 10%-11% in the last 12 months, is growing 30% year on year. Right? It's a substantial base. And I expect it to continue growing double digit in fiscal 2025 because we have won some large deals. The deal that we talked about, Kevin, if you recall, during the earnings call that slipped into Q4, we closed it last quarter.
We are making immense progress in Sovereign. We won the opportunity, the contract to design and build a Sovereign Cloud for the Kingdom of Saudi Arabia. Right? We are doing that with UK Sovereign. And that business, Sovereign Cloud, is also growing 50% year on year. Right? So, AI is another thing that we can talk about. So long story short, great.
We are starting to see early signs of turnaround. OK? When it comes to the second priority on AI, we launched FAIR, Foundry for AI by Rackspace, in June of 2023. We have won 60 plus new engagements, 250 new opportunities, tip of the spear services offerings. Now we want to monetize it either on Public Cloud, as they want to run the workloads, or on P rivate, especially on inferencing. And we can talk about that later.
So good traction there. And on the debt side, which you brought up, we were able to restructure the debt in the last 12 months. And we reduced the debt by net $900 million. That gives us more liquidity. There's more cash infusion by the debt holders. So I would like to thank the debt holders who trusted the company's story.
And that gives us liquidity to go win more business and make sure that the return on capital for the debt holders and all the capital holders is very high. So pretty pleased with the progress. And what you're seeing right now is Rackspace 2.0.
Sure. And listen, I think one thing that's important too, and it was super helpful, was they contributed an incremental $250 million of capital to kind of bridge the cash flow, which we thought was super important too. Because not only the willingness to recapitalize you folks, but also contributed an incremental $250 million. And just the CPA in me won't allow me to kind of talk to that without saying there was no going concern opinion.
It was very efficient in the public markets without any concern. And I think one thing is we think about the value proposition, and maybe we talked to this a little bit, is there's been some shifts in the revenue. But as you were going through that, I mean, the core of what you do in a lot of different ways is an outsource function, a pretty important function.
And to me, I think you overperformed on that as you're going through that restructuring. Because anyone that would have been concerned about how that process could have potentially turned out as opposed to how it did, it didn't feel like there was a pause in the business at all. And I think that's a function of the product you bring to bear as well as the senior management shift.
So maybe talk to that a little bit to the extent you can. And we see it as a potential tailwind to the business now that there's not that concern about liquidity. And I think that probably affords you the opportunity to maybe even be more effective on the go-to-market.
Oh, absolutely. I think there's no concern about liquidity. We have enough of liquidity. My CFO is here. He keeps track on cash every single day. And that's not an issue. We have enough of liquidity on the books, as well as we have not even tapped into the revolver. Right?
And even on cash CapEx side, we are making very informed decisions on whether we want to do a financing lease because the cost of capital might be high, or using our cash CapEx cash to go invest, or even talking to the customers of investing upfront on some of the buildouts. Right?
So very efficiently using it, it has helped our go-to-market, to your point, the refinancing and the restructuring of the debt and the liquidity. So that's not a growing concern at all.
Otherwise, why would large health care providers actually sign a multi-year deal with us? We are bringing the EMR, EHR application into our data center. It's massive. They are running the whole hospital on the workloads that we are managing and operating. We just migrated for AdventHealth, and it's public information because we put a press release out there.
One of the largest health care provider systems in the U.S., 56 hospitals, 38,000 concurrent users, and we managed them in record time, and that helped them to even plan for the hurricane that was approaching at the same time that we were doing this migration, so long story short, performance has improved. We offer five nines, high availability for these customers, and we've been managing mission-critical workloads for 25 years, so we know what we're doing here.
In a moment, maybe shifting gears a little bit, talk about the revenue mix today, Private versus Public cloud. And on the private side, I think one of the opportunities we see too is you've got a fair amount of data center capacity. So as some of that revenue starts to scale, maybe talk about the margin nuances of the business, the rates of private versus public cloud. And there's a lot here, but I think it's important.
And you alluded to this earlier, but if I heard you right, I think it was about 25%-30% of Professional Services signed at point of contract, and that scales to five or six. Is that ratable? Is that kind of $5-$6 over the life of the contract? Because again, it seems like you're seeing some outsized success on the Professional Services side.
That's something that historically was a little bit of a struggle for you.
Yeah. Yeah. So let me start with the last one, and then I'll give you the mix of public, private, as well as what we think of from a gross margin perspective. I think this is a very pertinent question, so on the services side, I think what we have seen is we are starting to expand into mid-market and enterprise. OK?
Mid-market customers are customers that don't get the love from the large service providers because they don't have the coverage model. OK? They don't have the coverage model. The large SIs go for larger deals and larger enterprise customers.
Mid-market customers have the wallet, and the competition is less, but coverage becomes a challenge. And that's why we basically go with the hyperscalers and then lead with services, bring in infrastructure, and help them. OK? So we are seeing expansion in mid-market.
I'll give you a little color on the services side too. On the enterprise, we said we pick top 25 enterprise. We sign MSAs, Master Service Agreements, with those enterprises. We have signed about now 12 or 13 MSAs out of the 25 that we picked. That gives us a license to hunt for services on those large enterprises. These are massive enterprises. Right? Now, how do we go differentiate?
Why would they buy from Rackspace Cloud Services? Why not go to some large service providers? Three reasons. Number one, they know Rackspace DNA is all about customer service. Our Fanatical Experience gives us very high NPS scores. Number two, we always innovate. It's an open innovation is what we believe in. We bring the right solution for the customers, working with different providers, OK? Partners. That gives the flexibility.
And number three, which is very important, is most of these service providers. It's a labor plus model. OK? More the labor, more the revenue. That's why they announce their attrition rates, their hiring rates, because that's a leading indicator. We are a labor minus model. By that, I mean when we go to the customer and have a discussion, hey, your data migration project, I'm not going to put 100 resources.
I'm going to bring my IP and tool and automation to move that from whatever source onto Azure or AWS and do it very effectively. That labor minus model is absolutely that's where we get entry into customers. Right? Large customers. Number two is Elastic Engineering. Right? You have Managed Services on one side, which is outsourcing services, very sticky. Then you have professional services, very transactional. We created Elastic Engineering right in the middle.
Customers don't have to sign a three-year contract. But they take a part of resources that we have. That's an Uber kind of stuff. And then they use it for a number of years. Right? So, that innovative way is how we are winning business. OK? Now, let me talk about public and private. See, I think we are still in the transformation story. And I think we have made a lot of progress.
Ultimately, we want to grow both public and private cloud. OK? Public Cloud, if you look at the hyperscale market, it's a 15%-20% growth. Pick a number. Right? And we believe the infra side, we might grow mid-single digit. That is purposeful because we want to win the right type of business so that we can generate the right kind of margins and EBITDA dollars. On the services side, we'll grow double digits.
OK? So, we'll perform at or above market on services. And given all the MSAs we are winning, I think I feel that that's doable. On the Private Cloud side, the Private Cloud market is a mid-single-digit grower. OK? And I believe that we will stabilize in 2025, maybe flourish, just like decline in Private Cloud. And then we start growing in 2026.
And I think early signs of year-on-year growth we might see in the second half of 2025 on Private Cloud. And that is a big deal because now we are weaning off and running off the old generation products and customers. And now we are bringing a new backlog of customers that might be very sticky for the next five to 10 years. So we are really building a backlog of recurring revenue for this business.
And so, it takes time for a recurring business to fall off the cliff. And it takes time to basically bring it on. So, I think we have actually made good progress with making bets in Health care, Sovereign, BFSI, and other areas.
Importantly, at a much different margin profile.
And different margin profile. So, a good point that you make. So, I think Private Cloud, mid-30% to high 30%. Right? That's where we've been operating. Now, can we go drive 45%-50%? Yes. Because you've mentioned something very important. We have fixed cost in the model. We have data center capacity. As you pour revenue, that might come in at 45%-50% drop rate. But we are going to use some of it to actually win more business. OK?
So I want to maintain the margins between mid-30% to high 30%. Can we do better? Potentially. But I want to have that capacity to go win more business. In case of Public Cloud, public cloud gross margins because of the infra mix of 70% is roughly about 11%. I believe that 11% is going to expand too because our services mix will go up.
And we will also start winning good business on infra side and walk away from business which doesn't make sense from an ROIC perspective. So, margin expansion overall in EBITDA and EBIT, driven by OPEX efficiency. And slight gross margin expansion, mainly driven by the Public Cloud side of the house.
On the public side of multi, as you think about what's the optimal mix here in terms of professional services versus infrastructure over time, if there's a way to think about that?
So, I think we have a lot of infra in the business. Right? Because it is sticky and it's like long term. So, you can't change the mix overnight, Kevin. Right? So right now, it's 70%, 30%. As we continue to grow double digit our services, I think you should expect that to be sort of services going to 40%, 45%, and infra going down a little bit. Right?
So, I think we'll change the mix over a period of time. But we'll also make sure that infra revenue mix is a good quality revenue.
Sure. And I think one of the things you've had some success even on the margin on the infrastructure side because you come to realize their clients couldn't deliver the value, which is why maybe talk to that because I think you've been a little bit more rigid in terms of renewal and saying, hey, there's just this certain rate of return we're going to absorb. And if not, we move on. So maybe talk to that.
And so, we walked away from a few deals because there are two ways of approaching this. Right? That's why services-led motion is very important. Right? So, I'll give you an example of a deal where we were able to win the renewals with higher margins. So, there were deals where we went in. I told the customer, called the CIO, and said, I don't want this deal. You can take it back.
Then we went and had a discussion. I said, "hey, if you attach services to it and give me a better margin on infra, I can sign another five-year deal. We want that deal." OK? We won one more deal in Europe, similar one. A good case in point that you can go and talk to the customer. And the customer knows you're driving value. Now, the question is, you have to pay up for the value.
Right? But there are customers who won't agree for that. Right? They have other plans. They want to go directly to the hyperscalers. That's okay. If I don't make money, I'm okay with you going to the hyperscaler as long as my hyperscaler partner can retain that business because my relationship with the hyperscaler is very strong.
So those deals we walk away. And you have started seeing infrastructure business and revenue going down is because we are walking away from those deals. And we are very selective on the type of deals we win also in the market.
There's still a lot to talk about, but I think one of the things that is important is the GenAI opportunity, right, particularly as your clients shift their workloads. And you've been front-footed on that in terms of introducing new initiatives. Maybe talk to that a little bit. And is there a way to think about how that would sit across the organization just directionally?
Because the good thing is you're very healthy end markets as you're kind of transitioning the business. So that gives you an opportunity. And then maybe weaving into the GenAI, I think one of the debates that people have gotten comfortable with over time is why the hyperscalers wouldn't come in to kind of do what you do. And I think there's a lot of reasons for that, but maybe just for the broader audience, talk to that a little bit.
So, I think we will play across the entire AI lifecycle. OK? Starting with helping customers identify use cases. And we have a lot of customers. We have 60 plus engagements, one that we have won, where we help customers identify the use cases, help them pick the large language models, and help them train those large language models. Right? And then finally, once it's trained, the models are trained.
Now you have to go run the inferencing on it. And then you land it either on public cloud or on private cloud. And so you can monetize across the entire AI journey. And that's what we are doing. With FAIR, we said, "listen, currently enterprises are just dabbling still." OK? Most of the money is spent by the hyperscalers, by OpenAI, by Meta's of the world in training Large Language Models.
That's why you see CoreWeave, Lambda, the system providers, the chip providers like NVIDIA, Broadcom, et cetera, all making money. Right? Enterprise wave has not started yet. Right? The training wave is just getting started. Where we want to focus is training for sure a little bit, but mainly on inferencing. Inferencing is the day two plus workload. That is the longest workload.
And you want to manage the inferencing workload, manage and operate it like any other workload. Now, why will Private AI be suitable? Two reasons. Number one is there are a lot of customers who don't want to move their data, right, move their data to public cloud. They want to have AI come to where the data is as opposed to data going where AI is run.
So in those cases, that's where Rackspace comes in, right, with the Private AI offering.
We have a Private AI Anywhere offering. So, we have expanded our offering. That offering may not sit in my data center. I can have that offering sit the stack set in customer's data center. I can connect it back to a data center that might sit in my storage in my data center and manage and operate it for them. That's why it's called AI Anywhere. It can be anywhere.
But it can be in customer prem or it can be in a hosted private environment like ours. So in the entire lifecycle, we are going to monetize it. We are early cycle. So the reason FAIR has given us thought leadership that will lead to Mind Share and ultimately lead to Wallet Share. And we will go after this Wallet Share. But I think the real monetization is going to be in inferencing and fine-tuning.
And that's what we are going to focus on. For example, we also launched GPU- as- a-S ervice, by the way, on a Spot platform. You should go look it up. Very innovative. We have used Kubernetes platform and containers to launch GPU-as-a-Service for a developer community because now we are learning what do they need. Right?
And we are going to use that learning as we move into enterprise with large enterprise customers. So we have a large Japanese conglomerate who is actually running a POC with us on Private AI and running the inferencing workload with us.
Intriguing. I think the other thing too, Amar, you talked about this a little bit. But maybe the amount of workloads that are still on prem that can shift over. I mean, there's just a lot of runway in the broader model.
Absolutely. I think if you tell the hyperscalers, they'll say 80% of the workloads is still on prem. Right? So, pick a number, 60, 70, 80. There's a large number of workloads. Right? Customers in mid-market, even in enterprise, enterprise customers want to consolidate their data centers. Nobody wants to be in the data center business. We have 34 data centers. We know how challenging it is to manage data centers. Right?
And that's become our key strength here too. So that workload will have to move. 60%, 70% of the workload will have to move either to Public Cloud, to Private, which I call is a Custom Cloud, or to Edge. And we want to catch the workloads everywhere. On the Public Cloud side, we'll catch it on a services offering working with our hyperscalers. Private Cloud, we will be the infrastructure service provider.
In Edge, we'll have our own edge offering.
Maybe you're close to time, Amar. But I don't think we could end without talking about kind of your consultants, right, and kind of the maniacal client service focus and the IP you've created over time, which I think is one of the subtle points of the model. So maybe talk to that because I think there's been some stabilization on the consultant front too.
Maybe talk to that because there's been multiple decades of just IP and investment that help create just the culture and go-to-market motion around Rackspace.
Yeah. So, I think when you think about Rackspace, it's a very heavy engineering culture. Right? Everything is about solving customers' problems, ultimately driving business outcome through a solution driven by engineering. Right? And that culture, no one will take it away. It's been there for 25 years. It's embedded in the company. Right? And that culture we have continued to nurture. I'll give you an example.
See, OpenStack is coming back. OK? We have some Boomerang Rackers who have come back to the company on OpenStack because we invented OpenStack. And OpenStack, by the way, because we had the talent in OpenStack and VMware, et cetera, we were able to win the Kingdom of Saudi Arabia deal, knocking out almost half a dozen large providers. OK?
So the talent has stabilized. I think people are looking forward. We have now become more of a technology company.
If you go back into 2021, 2022, late 2022, we were mainly an infrastructure resale provider for hyperscalers. That's where all the growth was coming. So, when you have infrastructure resale motion and become just a reseller, you lose engineering talent.
So, in the last 18, 20 months, we have made investments in not just products and services but nurturing the engineering culture within the company. So it's existed for the last 25 years, and we want it to grow and thrive for the next 25 years.
And can you just remind us? I guess it's closing up. But I forget my train of thought. But anything we didn't ask? I mean, it's always terrific to have you. But just anything in the closing moments here you wanted to highlight?
No, I think we covered everything. I will just say, listen, I think this is, thank you very much for having me here.
It's a pleasure.
We are a very unique company. We're a category of one. Whenever I go to customers, as I say, I don't have to convince customers who Rackspace is. But customers are looking for help. Customers are looking for help from partners like us who will go with an unbiased but an opinionated view. Why do I say unbiased? Because we have talent across all three cloud providers as well as private, as well as on-prem.
So customers are looking for that kind of help. And customers also realize that not all workloads are suitable for ublic cloud because of cost, complexity reasons, and lack of talent. Right? And so I think they are looking for alternative. And I think you will see us continue to thrive in that space. I'm happy with the progress. More to do because there's a huge market out there.
But we have put all the right foundations in place to go drive it from here on. So the focus is all about execution. And plan conservatively, execute aggressively.
Without a doubt. Anything else? I keep saying last question, but as you've shifted the go-to-market motion a little bit from a competitive perspective, anyone to call out in terms of where you're seeing folks more or less in terms of?
No, I think so. Listen, I think we picked Epic as an example. Right? I think we were very focused on what we should not do. That is one thing that we did early on the last two years. Pick stuff where we can actually win. And there's enough of market. We are not a $20 billion, $30 billion, $40 billion company.
We're a $3 billion company. I need a market that is $5 billion or $10 billion. Good enough. If I get 20%, 30%, 40% market share, I'm growing. So we picked Epic. And the reason I'm bringing up Epic is when you think about Epic, 40-plus% market share in EMR/EHR, Cerner, and MEDITECH, other competitors there. Epic is growing here in the U.S. as well as internationally.
Epic, the only three ways people can do it, either Epic host Epic applications themselves or third-party providers like us, or how they do in their on-prem. As I said, no customers want to do on-prem because it's becoming tough to manage data centers. So, for us, we have picked spots where we can differentiate and become sole source. OK? And that is the one point I want. Other than that, in services business, there's a lot of competition.
There's a lot of competition. But we pick markets like mid-market where you don't see the big providers. So when you are an international company going into mid-market and customers have a global presence, you start differentiating there itself. Right? So we have good key differentiators. But competition is the same. I think the thing is how do we go outflank the competition. And that's what we are doing by picking.
That's a strategy that becomes very important and executing instead.
Terrific. I think we'll end it there. All right.
Thank you very much.
Really, really helpful.
For sure.
Yeah.
Thank you.