Hey everyone, welcome to the 84th Emerging Growth Conference and day one of our two-day virtual investor conference. I'm Anna Barry. Just a few notes: today we're running until 4:30 P.M. Eastern. When we switch to the next company, you'll see a black screen for a moment. Don't go anywhere; that's just us moving to the next company. If you do experience downtime, refresh your browser. Everything usually works properly once you do that, and our platform does work best on Google Chrome. During each company's presentation today, you can submit questions through the webcast module, and the presenters will attempt to address as many of these questions at the end of their presentation. Remember, all of our conferences, they're uploaded to the Emerging Growth Conference YouTube channel, so please subscribe: youtube.com/emerginggrowthconference. Happy to begin with BitFuFu Inc., trades on the NASDAQ under the symbol FUFU.
It's a world-leading Bitcoin miner and mining services innovator who's committed to empowering the global Bitcoin network through its industry-leading cloud mining platform, rapidly scaling infrastructure, and innovative mining services. Happy to welcome the Vice President of Investor Relations, Charlie Brady. Nice to see you, Charlie. Welcome to the conference today.
Thanks, Anna. Appreciate it. Thanks for having us.
All right, the floor is yours. Call me back when you're ready for some questions.
Great, will do. Thanks. This is our typical disclaimer slide. I won't go through it all, but just be aware that we are making forward-looking statements on this call. So who we are, you know, as Anna said, we're a Bitcoin miner, a world-leading Bitcoin miner, depending on how you arrange it. Top five or six Bitcoin miners in the world globally. We also provide mining services. We're not just a self-miner, but we have a cloud mining business as well, where we offer cloud mining and the opportunity for customers to mine Bitcoin on their own without having to go through the complex infrastructure and build-out that you would as a self-miner. I'll get into that in just a few more minutes. As you said, one of the reasons the founding of the company was really to empower the global Bitcoin network.
The founding of Bitcoin really is based on decentralization and making it accessible to everyone. I think over time, as it's become more institutionalized, as the difficulty rate's gone up, it's become more and more difficult for individual people to self-mine Bitcoin themselves. The cost structure has really changed dramatically over the past few years, particularly with the last halving we had in April. Cloud mining really offers an opportunity to decentralize, democratize Bitcoin mining, and make it available to virtually anyone who wants to do so. We are the number one cloud mining service provider in the world. As of the end of June this year, we had over 623,000 registered users on our cloud mining website. These are users that have gone through the KYC, know your customer, background check, have inputted their own Bitcoin wallet into the system.
They don't necessarily mine Bitcoin 100% all the same time, but those are users that have been customers of ours and are registered on the cloud mining website. It's about a, we estimate, greater than 40% market share globally. Pretty substantial position in the cloud mining business. We also have a solid foundation that we do mine on. These are real locations, real machines, as of the end of June this year. We do put out monthly operational statistics to update this. In early August, you'll see the July statistics. As of the end of June, we managed 36.2 exahash of computing power, 728 MW of capacity, and that's on five continents. I'll get into that in more detail in just a couple of slides.
As of the end of June, on our balance sheet, we held 1,792 Bitcoin on the balance sheet, which I think puts us up or near the top 20 as far as public companies with Bitcoin on the balance sheet. We have two real basic businesses here and a couple of smaller sub-businesses. Self-mining and mining services are the two platforms. The self-mining is what you'd find with a typical self-mining Bitcoin miner. We have a number of rigs in various locations around the world, five continents, as I said, and we use this to mine Bitcoin for ourselves. These machines run essentially 24/7. We do have an internal software system, proprietary algorithms, proprietary IP that we use to monitor the system in real time and optimize all these machines running across the entire network. In fiscal 2024, we produced 2,537 Bitcoin on the self-mining side.
On the mining services side, which is where our cloud mining business is held, we produced 4,947 Bitcoin. Those are Bitcoin that went to customers. That is customer-mined Bitcoin on the mining services side, on the cloud side. We do have some software that we've put out in the past six months. We have our own mining pool now called BitFuFu Pool that customers on the cloud side can use to choose one of either BitFuFu Pool or two other pools when they do their cloud mining services. We also have our own operating system now to overclock, underclock, and really optimize mining rigs. If someone has, we do host third-party mining rigs. That's a small piece of our revenue, but we do make that software available to customers to allow them to tweak the mining rigs that they are running in some of our hosting centers.
This gives you a sense of the track record we've had over the past several years. The company was founded late 2020, in December 2020. The real first year of operations was 2021. You can see from 2021 through the end of calendar year 2024, we've had pretty substantial growth. Revenue growth is 65% CAGR. Our adjusted EBITDA growth is 173% CAGR. You can see how we're positioned over the past four years of growth relative to some of the other mining peers out there. We stand up pretty well. We're pretty proud of that fact. Also, on the lower right bottom of this corner here, you'll see in terms of the profitability of net income per year.
The company has been profitable from a net income standpoint every single year it's been in operation, which, in the mining industry, is not something you see from every public company out there. This is kind of a map and overview of where our data centers are located. You can see in the U.S., we've got 353 MW out of the total 728 we had at the end of June, so just a little under 1/2 there. Ethiopia is the next largest area with 229 MW, a little over 30% of capacity there. We're also in Finland, Oman, Indonesia, Brazil, and Paraguay. Of the 728 MW that we have globally on these five continents, a majority of this is leased capacity. We currently secure and own 164 MW of this capacity. The rest is leased capacity.
The company was founded on a 100% asset-light model where all of the capacity was leased. Last year, beginning in October, we started to pivot toward acquiring our own data centers. We acquired a data center in Ethiopia, 80 MW there, acquired Oklahoma, 51 MW, and then two long-term leases for two more sites in the U.S. for 33 MW. That's what we currently have under our control. The rest is leased. Part of the strategy here is to go and acquire more data centers to change that mix to a little more heavy-owned and less leased capacity for a number of reasons.
It allows us to lower our cost basis, gives us more control over the hosting contracts as far as longevity and stability of those contracts, and really allows us to have those data centers positioned if we, down the road, ever want to pivot into HPC or AI data centers, which you've probably seen some of the miners have started doing or considering doing. That's the strategy going forward as far as data centers. Our current pipeline of M&A that we're looking at to acquire data centers is somewhere in the neighborhood of 700- 800 MW of capacity globally. Most of that is brownfield. There is some greenfield where we would build a new site from scratch as opposed to acquiring an operating site, which is what we've done in the past. The majority of it's going to be concentrated probably in the U.S.
Although, clearly, given our international exposure, we're certainly looking at data centers outside the U.S., such as additional in Ethiopia as well as in the Middle East and Africa. What we've created here with the mining pool, the mining operating system, the cloud mining services, we also sell mining machines and lease mining machines for customers. We've got the self-mining business. It really is a one-stop mining service platform. Anything you want to do within mining Bitcoin, you can do from a BitFuFu platform. Buy a machine, lease a machine, do cloud mining, host your machine there, optimize your machine with the operating system, use our pool. It's right there, one stop. We really apply full transparency on the cloud mining side. I'll get into that in a minute.
That's one of the key differentiators that's out there as well, that these are real machines in real locations that can be physically identified. When we say cloud, it doesn't mean it's just out there in the cloud. It actually is at a physical location being mined. You can see on the cloud side in the bottom right here, we've had pretty substantial growth in the number of registered users, over 100% growth from Q4 2023 through June of this year. That number continues to grow. That's really with very little, quite frankly, advertising and marketing effort on our part. It really is more of a word of mouth. I would also mention that that's without selling cloud mining into the U.S. market. We currently don't offer cloud mining into the U.S. market due to regulatory controls. We are actively looking at doing that, getting into the U.S. market.
We just want to make sure as a publicly listed company that we follow all the correct regulations and rules and make sure we're doing it correctly and don't trip over ourselves on some state or federal regulations. We are actively involved in trying to move that into the U.S. market, but we don't do it at this time. I think at some point down the road, we probably will. That's a completely untapped market for us. All this growth has been without even tapping the U.S. market. Really into the cloud mining business and how does the cloud mining benefit a user, a customer versus buying off an exchange? What we have found over time is that cloud mining allows a customer to acquire more Bitcoin than if they were to acquire directly on an exchange.
Part of the reason for this is that when you're buying on the cloud mining side, there's two parts to the fee you're paying. There's the terahash fee per day cost, and there's a service fee per day per cost. The terahash fee gets paid 100% upfront. However, the service fee, customers are allowed to pay a minimum of 10 days of that service fee, depending on the length of the contract. What that allows them to do is to actually self-leverage themselves up and apply more of their dollars upfront to cloud mining to mine Bitcoin rather than having to pay the full service fee immediately up top. It's kind of built-in leverage on the cloud mining, which allows you to acquire more hash rate, to acquire more Bitcoin.
Therefore, over time, at the end of the contract, we've found typically customers derive anywhere from as little as 2% up to 20% or 30% more Bitcoin than if they just took their dollars immediately at the start and bought them off an exchange. Obviously, that dynamic moves differently throughout the cycle as to where difficulty and rate goes and hash rate difficulty rate goes and Bitcoin pricing goes. Over time, that's kind of the benefit to the customer of mining on the cloud. It also allows customers to really do it from anywhere, right? They don't need the complexity. They don't need to go buy rigs. They don't need to find a mining pool. We've kind of taken that and offloaded all of that from the customer having to do that. If you're starting a self-mining rig, you've got to buy the rigs.
You've got to find a place to locate them. You also have to find electricity costs. Electricity costs are the number one cost input to mining Bitcoin. I would say given where the difficulty rate is today and the hash rate is today, if you're not mining for $0.04 a kW hour or less, it's going to be hard to make a meaningful profit on those machines in any sort of reasonable payback time. Cloud mining really allows you to not have to be as concerned about that and making sure you're getting an electricity rate. Not to mention the fact that from a self-mining perspective, people have mined them in their homes. These machines tend to be extremely loud if you're in a small enclosed environment. Really not something you want to have probably running in your basement.
As these machines become more and more powerful, obviously they use more electricity, they become louder. That's not something that a hobbyist is going to want to be probably putting in their basement. It requires different electrical contract connectors as well. The cloud mining really takes all that away and allows a customer to go on the website, do the KYC background check, put a mining wallet in, choose a pool, choose how much terahash they want to buy for what certain period of time. We sell it generally in 30-day chunks, so 30, 60, 90, 180 days. Pick that time period, pick the terahash, pay for the contract, and you're mining Bitcoin in a relatively short period of time. It's a pretty easy process for just about anyone. Self-mining, I won't spend a lot on this. It's typical with self-mining works from any other self-mining company.
We have machines at locations that are mining Bitcoin. This is Bitcoin that we keep for ourselves for our own account as opposed to the customer cloud mining Bitcoin that gets mined. Our HODL policy, typically we have sold enough Bitcoin on a daily basis to cover our hosting costs and electricity costs, and we've HODLed the rest. We've had some deviation from that. If you go back and look at our monthly operational updates, I believe it was in May, we had sold a bit more than normal Bitcoin because we had paused selling Bitcoin for three or four months when Bitcoin price was low. When Bitcoin got into above SGD 100,000, I think around SGD 104,000, we sold some Bitcoin to kind of make it catch up on that from working capital purposes.
By and large, our policy really is to hold as much Bitcoin in the balance sheet as possible. We are long-term believers in the long-term value of Bitcoin and see it going substantially higher from where it is today. I mentioned earlier, we have our own proprietary software we call Aladdin. What this allows us to do is to take our entire hash rate and slice it in between and slide it between an allocation between cloud mining and self-mining. This is unique to BitFuFu. Because we have this cloud mining side and this allocation software, proprietary software, we can take a look at our entire hash rate and say, okay, how much do we want to allocate to cloud, how much to self, depending on where we are in the cycle, the Bitcoin mining cycle, depending on various cost parameters, depending on the demand for cloud mining.
We will move that slider back and forth to allocate hash rate between self and cloud mining, depending on a variety of those factors. It really allows us to reduce the volatility of the revenue stream because we're not tied just to the price of Bitcoin. When we sell a cloud mining contract, the majority, all the terahash cost is paid up front. We lock that in at that time. A portion of the service cost is paid up front and that's locked in over time. That volatility of being tied just to the price of Bitcoin really does get reduced. Cloud mining has been anywhere from 50% - 60% of our total revenue. It's a large percent of our revenue that is locked in over the life of those contracts.
Among other things, this also handles managing the entire system, optimizing the fleet of miners that we have out there, making sure that they're operating at proper levels. We do guarantee 95% uptime to the cloud mining customer. The way we can do that is because we're monitoring these machines in real time all the time. Because we have this fleet of networks on five continents, if we were to ever have a data center go down, which does not happen, but let's say that happened, we could simply slide some of our self-mining capacity over to the cloud side and the customer would essentially never see that blip. That's a reason we can guarantee 95% uptime for customers. We actually run a little bit higher than that. That's the formal presentation.
I have a couple of slides at the back just on the second of the first quarter operational highlights and our June operational highlights. I mentioned we do put operational updates out every single month with some of the production numbers. You can see that for the month of June, we mined 445 Bitcoin versus 400 in May. Of that, 58 of the Bitcoin was self-mined, which is the Bitcoin that BitFuFu keeps, and 387 Bitcoin was mined by customers through the cloud mining operations. At the end of June, we held 1,792 Bitcoin on the balance sheet. That's the end of the formal broadcast. I'll turn it back to Anna if there's any questions that I can address.
Great job, Charlie. Yes, we do have some questions. Let's see. I want to start with BitFuFu. Do you have a plan to enter the AI HPC area or just stick with mining Bitcoin?
Yeah, and that's a great question because you have seen a number of miners start to pivot that. Some have gone 100% in going AI and really left the Bitcoin mining scene. We don't do it at the moment, but we're open to looking at the idea. I think part of the issue for us is right now we've got to acquire more data centers. As I mentioned, we have 164 MW out of the 782 that we actually control. To do HPC AI, you have to own the data centers. You can't do it from a lease model as an operator. I think as we go out and acquire more data centers, and we've said we want to get to a gigawatt of additional controlled capacity by the end of 2026. As we get to the next 12-18 months, we get more critical mass on that.
I think that's something we're going to take a look at. Right now, the cost structure between HPC AI and Bitcoin mining data centers is a multitude of 8x-1 0x higher as far as cap cost goes. It's something we're going to evaluate in the capital cost and make sure that we would have customers that are lined up for 10+ year contracts with multi-renewals. I think we'd probably partner up with someone who has been doing it, is doing it. I don't think we'd start cold, build a Greenfield AI HPC site. I think at some point you could see us doing that once we get a little bit more critical mass on the data center side.
Our viewer, Olive, wants to know, where do you store your Bitcoin and where do your customers store them? What does security look like where the Bitcoin is stored?
Yeah, from a customer standpoint, it goes from the mining pool to the customer's wallet. It doesn't come through BitFuFu. Once we sell them the contract, they input their own personal Bitcoin wallet into the system. They pick one of three pools. The Bitcoin that gets mined on the cloud side for the customer goes from the pool, distributed to the customer's wallet. However they're going to secure their own Bitcoin really is up to them, how they want to do that. From the corporate side, we use several cold wallets. I don't want to get into specifics of how we do that or how that's set up, but we use extremely secure measures to store our own Bitcoin personally for the corporate side.
What percentage of your cloud mining revenue is institutional customer versus retail customers?
Yeah, great question, because I think when people think cloud mining and they look at this 623,000+ users, they would think it's all retail. From a user account standpoint, there are more retail accounts in that 623,000. From a revenue standpoint, the majority of our cloud mining revenue is institutional revenue. This would be hedge funds, family offices, high net worth. There are some private corporations in there that are mining Bitcoin because they want to be involved in that. You've seen companies start using it for Bitcoin as a treasury strategy, buying off exchanges. We've seen customers looking at cloud mining as a treasury strategy to acquire Bitcoin. It really is a mix within that institutional side. Largely, certainly over 1/2, significantly more than 1/2 of our revenue in cloud is institutional money.
Hunter asks if you can talk about customer retention.
Yeah, that's a good question too. We do have a very high retention rate. I think if you look at, depending on how you want to slice it, we don't release the granularity on this, but I think our net retention rate tends to be in the 70%-80%. If you look at our top customers, or if you look at our VIP customers, which are customers that are spending $10,000 or more, that retention rate tends to be closer to 70%.
Talk a little bit about BitFuFu's strategy to expand your operations.
Yeah, and that kind of goes into acquiring more of these data centers. We bought the first one in Ethiopia in October of last year. We bought the Oklahoma site in January or February of this year. We did long-term leases back in December of last year in the U.S. We've got that 700-800 MW pipeline that we're looking at. That expansion, really, it's global, but it's probably going to be mostly in the U.S. Those are the sites we're looking at. I think those sites could range anywhere from, you know, at the very lowest end, 20 MW. That tends to be on the lower side. We probably wouldn't do a site lower than that. It doesn't make economic sense. All the way up to 100+ MW sites. Some of these sites are also expandable beyond the initial gigawatt sites.
We do want to add that gigawatt of capacity by the end of 2026.
Perfect. When it comes to your policy with selling or holding the Bitcoin it mines, what is your treasury policy in regards to that?
Yeah, you know, we've typically sold enough on a daily basis to cover those hosting fees that we pay and to cover electricity costs for owned sites. By and large, we want to really HODL as much as possible. That's why you've seen us accumulating Bitcoin on the balance sheet. We're going to continue to do that. We have on occasion stepped away from that daily purchase price and paused it and then sold a large chunk later on, which we did last May. On a typical run rate for us is to sell just enough on a daily basis to cover the hosting costs and electricity costs to run the data centers.
Besides your own mining, what's the revenue model generated by customers? If you can talk about that?
In terms of the breakdown between the self-mining and the cloud mining, the self-mining has been, if you look at 2023 and the first nine months of 2024, it was about a 33% split on the self-mining business, low 60%, 62%, 63% on cloud mining. There was a small piece in there, 1.5%, 2.5% that was hosting and other. In Q4 of 2024, we did sell SGD 30 million of mining machines, which skewed the mix a little bit. In Q1, we sold SGD 7 million of mining machines. The mix didn't hold perfectly there, but by and large, I'd say greater than 50% of the revenues are coming out of the cloud mining side. 35%- 45% are coming out of the self-mining side.
The rest is a mix of that other, which includes hosting, miner sales, mining leases, and the BitFuFu Pool and BitFuFu operating system, which those are relatively new products that we came out with in the past six months or so. The revenue generation from those is pretty small at this point. It doesn't really move the needle, but over time we think it develops into something larger and really helps us provide a full circle one-stop shop for the customers, which is another reason that we introduce those services to the customers.
Perfect. Christian asks, how do you cure skepticism from investors who have seen scam cloud miners out there?
Yeah, that's a great question. I think that the term cloud mining has certainly gone a bad route because there have been Ponzi schemes out there and scams, as you say, that have kind of given a bit of a black eye to some of the typical term cloud mining. I think the difference is that we can identify that these are physical location sites located on the ground. These are physical machines that are operating. We are a publicly traded company, so we are under obviously SEC regulation for our compliance purposes, which some of these scam companies obviously have been small private companies that have been offshored in faraway places that were never audited and didn't have that sort of compliance overlay on top of them. That's something that we think helps with the transparency.
In terms of the cloud mining too, when you're a cloud mining customer, you can see the Bitcoin being mined in real time, either on your phone app or on the website itself. We really do try and provide clarity on that. The pricing upfront is very clear. Here's what you're going to pay for terahash, here's what you're going to pay for the service fee, and that's the cost of life of the contract. It doesn't change regardless of what the price of Bitcoin does. I think that you can look at the 623,000 users that we have on the website, the market share we have on the website over the number of years of cloud mining.
I think that, and the number of repeat customers that we have, I think really points to some of the clear integrity that the cloud mining business that BitFuFu has versus some of these earlier so-called cloud mining companies where there really wasn't any Bitcoin being mined.
Charlie, I want to give you an opportunity to answer this last question and feel free to close with this however you would like. For a potential customer who might know nothing about Bitcoin, how would they get started with you and why?
Yeah, I think I'll answer the why first. The why is because the cloud mining side really doesn't require you to have a lot of knowledge about Bitcoin or blockchain or how it actually works. Certainly, we encourage our customers to learn as much as we can. If they go to our website, there's a lot of educational material on Bitcoin and blockchain and cloud mining that they can dig into. The second point is why do it really is from a complexity and ease standpoint, right? You go on there, it takes, you know, you do the KYC background check, you have your own Bitcoin wallet that you put into, you choose one of, we have three different pools to choose from, including BitFuFu Pool. I think there's also Foundry is on there and Amp Pool, which are some of the larger pools out there.
Once you choose that pool and do the KYC and put your wallet in, pick how much terahash you want to purchase for what period of time, you're essentially mining Bitcoin. It's just that simple. You don't have to worry about electricity costs that are going to go away or regulatory concerns or buying the rigs or placing the rigs or maintaining the rigs. Rigs do break down, right? If a rig breaks down, it's offline, it's not mining Bitcoin. We've taken all of that complexity and really offloaded it from the customer and made it as simple and streamlined as it possibly can be for the customer.
Perfect. There you have it, BitFuFu on the NASDAQ FUFU. Charlie, thank you so much. A great presentation, and we look forward to seeing you again real soon.
Thanks very much.
All right, everyone, we'll be right back with our next presenter.