Welcome, and thank you for joining us today. We are really excited to share a 30,000-foot view of our strategy for dramatically scaling our business over the next several years. I'm really excited to introduce you all to Carly, our CMO, Gabe, our Chief Product Officer, and Jeff, our COO. Of course, you know Bill, our CFO. You'll get to hear how they are running their functions and their connectivity to our longer term strategy. They will share details on how we intend to extend our leadership position in the IaaS and PaaS markets to create a durable and differentiated growth company in a rapidly growing $70 billion market opportunity, serving developers through small and medium-sized businesses. Before we get started, I'd like to touch on two quick items, our safe harbor statement and the question and answer session.
The safe harbor statement can be found in both our investor video and presentation. For the question and answer session, please utilize the Q&A tool in the webcast console to submit your questions. Following our presentation, we'll answer as many questions as we have time for, and if we don't get to them all, our investor relations team will follow up with you following the event. Thank you again, and welcome to our 2022 Investor Day. Welcome to our first ever Investor Day. We are excited to meet with you all today to share our vision for scaling the company to $1 billion in revenue in 2024, and importantly, that we will be growing 30% or better as we approach that milestone.
Before we begin, on behalf of DigitalOcean, I'd like to express our appreciation and gratitude to our customers and our investors who have helped us get to where we are today. A special thank you to the DigitalOcean team for your dedication and passion to serving our customers and each other. Our values start with community and end with love, and that is why we are able to realize this incredible opportunity in front of us. We have an exciting agenda to reveal our long-term plans to substantially grow our business over the next three years and further through specific initiatives that we are funding. Gabe Monroy, our Chief Product Officer, will explain our growth strategy as a function of new product offerings and multi-attach strategies that serve customers throughout their journey while helping us achieve $1 billion in revenue by 2024.
You'll hear from Carly Brantz, our Chief Marketing Officer, who will discuss the marketing enhancements to further improve our highly efficient customer acquisition engine at the top, middle, and bottom of the funnel. Jeff Guy, our Chief Operating Officer, will share our customer nurture investments, our nascent and growing outbound sales approach, and investments we are making into partnerships to increase our percentage of revenue that comes from beyond the self-serve funnel. You will also hear from some of our customers today as they explain why they chose to build on DigitalOcean, why they stayed on DigitalOcean, even as their businesses scaled substantially, and the value of DigitalOcean's simplicity, community, and support. Before taking your questions, our Chief Financial Officer, Bill Sorenson, will share the compelling financial profile that we intend to deliver over the next few years.
Before we get to our first speaker, I wanna briefly walk you all through our journey, which began in 2012, when our founders pioneered the developer cloud platform with the mission to simplify cloud computing and enable developers and developer teams to quickly deploy and scale applications, collaborate efficiently, and improve business performance. We've scaled to over $520 million in ARR, serving more than 620,000 customers in more than 185 countries. Uniquely, 69% of our revenue is from outside of the United States. The global appeal of our simple, easy, intuitive on-ramp to the internet. At DigitalOcean, we truly are enabling aspiring developers and entrepreneurs to learn and to grow, to test their ideas, and yes, to realize their dreams.
Our market is measured with more than 100 million SMBs growing 14 million each year, and 30 million developers going to nearly 50 million by the end of this decade. Collectively, they spend a massive $70 billion today, and that spend is projected to grow 27% annually. This is a massive market opportunity, and we have a purpose-built set of capabilities to capture our fair share. As I joined this special company in the middle of 2019, we had a great foundation in terms of brand, relationship with our customers, and strong differentiators. What we have changed is the financial profile, as well as put a finer focus on the underlying metrics that drive this business. The results couldn't be clearer. We've transformed this business and are now set up to realize the limitless potential of DigitalOcean.
To quickly recap our transformation scorecard, growth has increased from mid-20s% to greater than 30%. ARPU has moved from growth in the teens% to more than 25%. Net dollar retention has gone from just under 100% to the mid-teens%. Gross profit margin has increased to 60% on its way to 70%, and operating profit has increased from mid-single digits% to double digits%. CapEx has been halved from nearly 50% and has incredible opportunity for improvement to sub-20%. Free cash flow margins have moved from -25% to mid- to upper-single digits% today. Finally, to make a finer point, a few years ago on the rule of 40, revenue growth plus profitability, we were in the -25% area, and now we are 40%+.
We are committed to delivering efficient revenue growth as foundational to how we build this business over time. Our longer-term targets as we hit our first $1 billion of revenue in 2024 are for consistent 30% or better growth, with GAAP gross margins approaching 70%, adjusted operating income of 25% or better, and free cash flow of more than 20%. We're targeting ARPU growth of 20% or better, NDR of 115% or better, with total customer growth of 10%. However, more importantly, customers above $50 per month to grow at least 2x our long-term annual target for total customer growth. I'm proud of our team for the results we put on the board, but we are in the early stages of a limitless market opportunity.
Today, you are going to hear a lot more about how we plan to realize our fair share of it. Central to our mission today is to share our growth strategy and the leverage we have to attract learners and help them graduate into builders and scalers on our platform. Learners are categorized as those that spend less than $50 per month with DigitalOcean, while builders and scalers spend more. Customers typically start with us as learners who spend $15-$20 per month. These developers are testing ideas, looking for a breakthrough. They represent approximately 80% of our customer base and about 15% of our revenue. As some of these customers make a breakthrough and start to ramp, they become builders experiencing the early stages of rapid growth. During this phase, we see them spending between $50 and $500 per month.
It's during this phase that they are growing rapidly, adopting more of our product portfolio to enable them to support their growth. They represent a little more than 10% of our total customers and approximately 1/3 of our total revenue. Finally, customers that spend more than $500 per month are very clearly scaling their business on our platform. They are multi-product adopters, typically using most, if not all, of our product suite. Scalers represent slightly more than half of total company revenue, which is growing much faster than our overall growth rate. This is unique. We aren't the typical land and expand tech company. We land, nurture, and we drive expansion as our customers generate their own rapid growth, and then we are accelerating that expansion as they leverage the breadth of our product portfolio.
We will share more about our highly efficient self-serve model and our ecosystem of educational content that provides millions of monthly website visitors that we will continue to enhance and invest in so that we can continue to acquire tens of thousands of customers at various stages of their own journey. This is a vital go-to-market motion because the learners start small in terms of dollars and they move up at their own pace. A low-cost acquisition model is vital to making our business viable over the long term. We will detail how these self-serve customers are drawn to DigitalOcean due to our knowledge base and community, and upon initial sign-up, are inexpensive to serve, but create a customer pool of potential high-value customers.
We will provide examples of these high-value customers that have graduated into viable and fast-growing small businesses that remain on our platform for years and in time become ever larger customers of DO. We will also dive into our small and medium business model, where we are investing in direct outbound sales, farming sales, and partner channel sales with an intention to obtain and grow customers that begin their journey on our platform already with a viable and proven business model and are ripe for expansion of net dollar retention. Given the product portfolio and the highly performant IaaS capabilities, we are attracting more and more scaling companies day one.
That's why we're building out a sales effort and expect it to fuel our growth as this go-to-market motion expands from roughly 3% of revenue today to nearly 10% as we approach our first $1 billion of revenue in a couple of years. We will share some more data on what industries our customers are commonly in, what products they currently are using, why they remain with us, and how we are going to expand and extend our platform to meet our customers' evolving needs while maintaining simplicity as an advantage. Also unique about DigitalOcean is that we are free cash flow positive at $500 million of ARR with growth at 30% or better. We are committed to sustaining our high-growth model while delivering strong free cash flow.
We look to supplement our growth strategy with targeted M&A investments, increasingly larger in size, as we look to accelerate our product portfolio and go-to-market efforts with high-growth businesses that help us fill gaps and go faster on execution of our roadmaps. We have the balance sheet, the experience on M&A, and the appetite to aggressively move to realize this limitless opportunity to serve the world's budding entrepreneurs. We are committed to generating efficient growth. It's our philosophy to invest appropriately in the growth opportunity, but also to deliver earnings and cash flow growth. It provides discipline within the organization, and it's creating sustainable value for our shareholders. To that end, earlier this year, we announced a $300 million share buyback, and just last month, we announced a second $300 million buyback.
That was done with this philosophy in mind, and over time, we will not just drive efficient growth, but continue to be capital efficient to drive attractive returns for our shareholders. We're excited that you're here with us today to learn more about DigitalOcean, to interact with our broader executive team and hear about our plans to grow this incredible business. I'll now turn it over to Gabe Monroy, our Chief Product Officer, to walk you through more details on builders and the journey they take with DigitalOcean.
Today, I'm gonna share how our product has contributed to the accelerating growth we've already experienced, and how we plan to grow even faster with a product strategy that is grounded in our customers, their challenges, their opportunities, and their desire to create software that changes the world.
My name is Tim Osborne. I'm the CTO at Atom Learning. Atom Learning is an education technology platform. We help kids learn adaptively. What we mean by adaptive is, as the kids are doing well, it gets harder, and if they're not doing so well, it gets easier for them. It's entirely online cloud-based platform, which kids can access from all over the world. We've had huge growth since, particularly since the pandemic and since the kind of explosion of online learning, around the world. We were with a different provider, and we were with just doing virtual servers ourselves, which was a lot of work. Being a bootstrap company, we wanted something that we could get off the ground pretty quickly, but also that was within budget and importantly had predictable pricing.
We knew exactly how much we were gonna be spending each month. One of the main drivers to moving to DigitalOcean was to have a more managed solution. We moved to use the managed Kubernetes and the managed MySQL databases were the first products that we came to DigitalOcean for. Since then, though, we've pretty much adopted the entire product suite. The last one that we weren't using was the container registry, but we've actually just started using that as well. We now I think pretty much use every single DO product. Adopting DigitalOcean was really seamless. I was able to set up a Slack channel with some of the solution engineers. They were super helpful.
They gave us lots of pointers, and within about a month, we put together a proof of concept deployment, and then one very late night, I switched over everything to DigitalOcean, and we've been on there since. One of the earliest points that I realized that DigitalOcean was gonna work at scale for us was when we launched a new feature on our platform, and we went into the exam season, which is our busiest time of the year. We were able to just completely scale the Kubernetes clusters without any additional configuration or anything from our side, which really naturally just scaled, you know, a threefold increase in traffic overnight without causing any problems for us. We have quadrupled in our traffic since we joined DO, and we're nearing 100,000 users on our school platform.
What was so great about DigitalOcean was that me, as someone, I didn't know Kubernetes at all at the time, but I was able to spin up a managed Kubernetes cluster and get to work without having to worry about all of the different options and various different configurations that that some of the other platforms make you go through. The main way DO has helped us grow our business is allow us to to just run our business and not have to worry about the infrastructure side of things. We've had very few people involved in the in the DevOps side, but yet we've been able to scale our business as much as we have without needing a huge team of people to keep the systems running.
With some of the other platforms, you need a lot of information and lots of qualifications or proprietary knowledge. Whereas with DO, it's very vendor unlocked, so you can use whatever bits of kit you want, but you're not learning something that's specific to DO. You're learning an industry standard, which is then very portable. One of the great things about DO is it's always moving with you and we've seen a huge amount of pace, and there's been some things where we're like, we need a new feature, and we chat on Slack to the solution engineers, and it's a month away, and they then release the new patch, and we have that updated set. It definitely feels like we've grown with DigitalOcean, and they are really complementing our growth that we're experiencing.
We're able to add new data centers. We're adding disaster recovery, failover capacity. All these things are possible on DO, and we're not hitting anything that, you know, where we're struggling to do something. There are some things in our infrastructure that I can never see needing to move away from DO. DO will always kind of be the core of our infrastructure.
DigitalOcean started its journey with the Droplet, a cloud-hosted virtual machine designed for developers by developers. Developers love Droplets because they're simple, predictably priced, and easy to manage and scale. Thanks to Droplets, our Infrastructure as a Service business has seen great success in the past 10 years, growing to $hundreds of millions in revenue and hundreds of thousands of customers. Just 3 years ago, all of our revenue came from infrastructure solutions like Droplets. Since then, we have diversified our product offerings, adding managed databases, managed Kubernetes, App Platform, and our newly released serverless offering, DigitalOcean Functions. All of these products are designed to meet the needs of our evolving customer base. Today, roughly 10% of our revenue comes from non-infrastructure products, and we see that doubling as a percentage of revenue in the next few years.
With 60 million new developers entering the market by 2025 and over 500 million new digital applications projected to be deployed in the next 2 years, we are just getting started, and I wanna share a few insights into our journey ahead. Our customers are at the heart of our future product strategy. As we've developed a greater understanding of our customers, we've noticed a pattern of usage on DigitalOcean that we refer to as the developer journey. Now, the developer journey starts with learning. With so much new technology being made available today, developers are constantly in learning mode. There's always a new open source technology to test or a new programming style to master. Once armed with this new expertise, developers set out to build their next application as part of a startup, a nonprofit, or a more established business.
If the application is successful, they may need to scale rapidly to meet the demands of customers across the globe. Now, whether the application is successful or not, developers will always move on to their next application, sometimes learning new technology and new skill sets to help them innovate faster. That's the developer journey we see on DigitalOcean. It starts with learning. It continues with building. It culminates with scaling, and it's a helpful way to think about our customers. We call the first group of customers learners, and they make up about 85% of our customers. Empowering the developer community is at the core of DigitalOcean's values. Today, 10 million developers come to our website every month to learn about programming languages, operating systems, open source, and more. We recently acquired CSS-Tricks to reach even more developers with learning content.
We have always been committed to providing free resources to the global developer community through our technical tutorials, published Q&As, tech talks, events like Hacktoberfest, or our partnership with the GitHub Education Program, where students can get credits to use DigitalOcean products. Our recently announced $4 Droplet is tailored for this audience, ideal for learning, hobby projects, and proof of concepts. I'm confident our continued product and community investments will allow us to expand upon DigitalOcean's position as the go-to cloud provider for learning. We call the second group of customers builders. These are typically developers aiming to build innovative digital products using APIs and cloud native applications. They wanna move fast with cloud products that are simple and easy to use with affordable and predictable pricing.
Builders are the fastest-growing customer segment by revenue and include early-stage startups like those we sponsor through Hatch, our global startup program. Our standard $6 Droplet provides builders the confidence to host and scale their VM-based applications. This product alone represents just over 50% of our revenue today. Builders are extending these Droplets with managed databases like MySQL, PostgreSQL, and MongoDB, and over 200 one-click applications in our marketplace, WordPress, LAMP, or Plesk. Managed databases have increased revenue 95% year-over-year. Container technology is also extremely popular with builders, and we see that with our managed Kubernetes service, which is growing 47% year-over-year. For the builders writing modern serverless and Jamstack applications, App Platform has seen its user base grow 120% over the last year.
We just announced general availability of DigitalOcean Functions following our acquisition of Nimbella in September of last year, providing builders with modern serverless capabilities, enabling more cost-efficient compute and better developer productivity. We call the last group of customers scalers, growing businesses who spend over $500 a month on our platform. Customers like Servrist, a global video delivery service, or Kea, a voice-based AI platform for restaurants. Scalers use multiple products across our compute, networking, storage, and managed service offerings. Many of these scalers actually started their journey as learners, spending less than $15 a month with us. Today, scalers only represent a small percentage of our customer base, but they account for half our revenue. Penetrating deeper into this customer segment is central to our product plans. We already offer these companies dedicated Droplets, which allows businesses to find the optimal performance for their workload.
We offer production-ready object storage and highly available block storage designed to make storing data simple and make data-driven insights possible. Our load balancers and networking products enable auto-scaling of applications, allowing cloud spend to increase and decrease based on demand. These business-oriented capabilities aren't restricted to Droplets or infrastructure offerings. They extend into all our products, container products, managed Kubernetes, serverless, and managed databases. Now, these groups together, learners, builders, and scalers, represent a massive market opportunity. The global developer population is projected to grow from roughly 27 million in 2021 to approximately 43 million in 2025. Each of these developers will embark on a journey of their own, and we welcome them with open arms on DigitalOcean. If you ask developers why they were attracted to DigitalOcean, they always answer with two things: simplicity and pricing.
Forrester ran a total economic impact assessment on DigitalOcean to surface the same results. Customers are using DigitalOcean to save on hosting costs and to save thousands of engineering hours thanks to our easy-to-use products. One of the things I learned working at Azure is that it's incredibly challenging to defend simplicity at scale. With so many product teams running in different directions, speed takes priority over design. That's why using AWS or Azure feels complex and inconsistent, like a group of mildly cooperating startups who share a food court rather than a fully integrated platform. There's a reason customers still value DigitalOcean simplicity after 10 years. Simplicity is a core part of our culture, and we're committed to infusing our product development process with design thinking and customer feedback. We know that our customers don't want 17 ways to run a container.
They want a developer experience that is carefully crafted and tailored to the needs of learners, builders, and scalers. Whether it is simplifying security for Droplets, enabling easy replication of data to our managed databases, or delivering a best of breed cloud console experience, DigitalOcean has a proven record of delivering products that are thoughtfully designed. We are laser-focused on scaling simplicity to maintain this differentiation, even as we add new products to our portfolio. Pricing is a good example of how simplicity shines through our products. Hyperscalers are obsessed with micro billing, highly technical usage meters. The theory is that customers wanna pay for what they use, and that's true. But it's also true that customers wanna be able to predict their bill at the end of the month, and they can't do that if their cloud bill looks like the utility bill for a housing complex.
Our customers don't have large annual budgets. They need to manage costs carefully. They also need price for performance that meets the needs of small and medium-sized businesses. One of the most common refrains we hear is that DigitalOcean's competitive and predictable pricing gives customers confidence that they know what their bill will look like at the end of the month and that they got a good deal. As we grow, we will continue to lean on what customers perceive as our core differentiators, simplicity and pricing designed for SMBs. The 2021 Stack Overflow Developer Survey ranked DigitalOcean the fourth most loved and wanted cloud platform behind the big three clouds. The feeling is mutual. We love them right back because growing together is core to our strategy.
We grow when our customers grow, from learning to building to scaling, and ultimately, to making software and digital products that change the world. The applications of the future will be global, hosted seamlessly across cloud and edge with great performance from anywhere in the world. They will store large amounts of data to enable machine learning and artificial intelligence, leading to a new generation of digital products that get smarter the more that they are used. More usage leads to more data. More data leads to better AI insights. Better AI leads to better digital products and to more customers across the globe, and the cycle continues. We call this the developer flywheel, and it is central to our product strategy. When our customers successfully unlock the developer flywheel, they outperform and outcompete the companies that do not.
As they grow, they need more compute, storage, networking, and value-added services. The developer flywheel is what powers DigitalOcean. When our customers are spinning the flywheel at full speed, revenue growth and NDR follow. Today, nearly 70% of our revenue comes from customers outside the United States. In an increasingly global world, edge infrastructure is required to deliver competitive application performance to users around the world. That's why the edge computing market is projected to reach $274 billion by 2025. We plan to use inorganic investment to accelerate our edge capabilities and infrastructure. With those new capabilities, we will be poised to deliver a set of products that are deeply integrated across cloud and edge with DigitalOcean's trademark simplicity and predictable pricing. This DigitalOcean customer uses DigitalOcean to provide an advertising analytics solution that helps companies optimize conversion.
They are a heavy user of our storage products, and their NDR is at 175%. This is the concept of data gravity at work. The more data customers store in our platform, the stronger the pull of applications that need to be close to that data. More data, more applications. More applications, more product usage. More products, more revenue and higher NDR. To capitalize on this phenomenon, we are accelerating product innovation using both organic and inorganic levers to bring even more data-intensive workloads to DigitalOcean. For example, we are investing in lower latency block storage, higher performance object storage, and better compute instances for our managed databases. EasySize is a DigitalOcean customer that is an AI-powered clothing fit and size platform utilized by fashion shops and ecommerce stores worldwide.
They took advantage of DigitalOcean's managed Kubernetes service to build a scalable and reliable solution that helped them increase the brands on their platform by 10x in just 12 months. They're emblematic of the trend of AI-enhanced applications, which is supporting a 35% CAGR for AI platforms over the next 5 years. Given this growth, AI and ML are popular learning subjects for developers. We have a unique opportunity to serve developers on their learning journey with AI and ML, and ultimately, like with EasySize, help them scale their AI-enhanced products on DigitalOcean. Expect product innovation through inorganic means designed to help small and medium-sized businesses realize the benefits of artificial intelligence and machine learning. Multi-product attach is strongly correlated with revenue and NDR growth.
Customers who use four products on DigitalOcean spend $1,100 a month, compared to just $28 a month for customers who use only one product. Yet on every cloud platform, including our competitors, it takes most developers days or weeks to stitch together products for common application patterns. DigitalOcean has the opportunity to radically increase time to value for customers using common application patterns that are inspired by the developer flywheel. Our goal is to enable developers to start building their next application in seconds rather than days or weeks with built-in best practices, edge capabilities, data services, and AI/ML. Solving the complexity of using multiple cloud products together will require novel innovation paired with inorganic acceleration. This is something our team is already hard at work on as part of our future product strategy.
For now, I'll just say that there is no company better suited to tackling this challenge than DigitalOcean. Many of the applications developers run on our platform will never see significant usage because they're intended for learning or proof of concepts, which are a core part of the developer journey. As our customers grow, DigitalOcean grows with them. As more developers have positive experiences building digital products on our platform, we earn more chances with more developers to win their next application, and in the process, we get an option on their future success. With 16 million new developers entering the market by 2025 and over 500 million new digital applications projected to be deployed by 2023, DigitalOcean is uniquely positioned to help developers across the globe get growing and create software that changes the world.
Now, I'd like to introduce our CMO, Carly Brantz, who will talk more about how we attract learners, builders, and scalers to DigitalOcean.
It has been an incredible honor to lead marketing as DigitalOcean's Chief Marketing Officer. At DigitalOcean, we start with customer experiences and end with love. For the last ten years, we've been creating the cloud designed to be loved by builders of all kinds, from developers learning new skills to businesses that are quickly scaling. We serve a global community who are driven to build in the cloud, and through that, we enable dreamers to dream, developers to create, and businesses to thrive. Whatever our customers are building, we are here for it. We champion their growth, and our team delivers customized experiences that no other cloud service providers can match. When I joined January of 2020, there was incredible potential at DigitalOcean's brand and platform that wasn't yet being realized.
We had a simple, reliable platform that was valued by developers and a wealth of tutorials and events for them to consume and learn from, but we hadn't optimized our marketing and product efforts toward the complete customer experience. In order to make improvements and ultimately drive more revenue, we had to become hyper-focused on the needs of our customers, understand how they differ at each stage of the developer journey, and create experiences that serve those needs. Our marketing efforts touch developers in all stages of their journey. We attract and enable learners and builders through our self-service funnel and provide marketing support to our sales team efforts with scalers. Next, Jeff will speak to you about how we are bringing more of those high-value scalers to our platform.
First, I want to speak with you about the groups that make up the majority of our customers, learners and builders. These customers can and do grow into scalers over time, but they start out in our simple and efficient self-service funnel. 90% of DigitalOcean's revenue today comes through our self-service funnel, which depends on both an efficient marketing operation and a seamless product experience to be successful. Since I joined, we have made a number of enhancements within our marketing engine so that we can see, track, and optimize the complete self-service funnel in order to drive growth. Our self-service funnel guides customers all the way from their first touch on the DigitalOcean website to signing up for our platform and adding new products over time. Because of this, to see a meaningful change, we needed to focus on all stages of the funnel.
With the end-to-end ownership of the self-service motion, there is not one stage of the funnel that is most important. It is necessary to make changes in each distinct area to see meaningful impacts. With the new focus on how our marketing efforts work in conjunction with our targeted sales model and product strategy, the customers that come through our self-service funnel serve as a moat for future growth. By using data science and propensity modeling, we are able to grow these customers faster and identify when they will be a candidate for a sales touch. Today, we see that 94% of our customers spending more than $500 a month came through our self-service channel and grew over time. This highlights the importance of our continued focus on growing this self-service moat.
Now I'd like to walk you through the different stages of the self-service funnel, what we've optimized already, and where we want to go in the future. Let's start with the top of the funnel and how we attract learners and prospective customers to our website. We call these visitors UVEX. Unique visitors excluding customers. We have hit our stride here. You can see that just as I joined, we were at 3.8 million UVEX per month in 2020, and now we are at 9.4 million. That is 247% growth in two years, and we are confident that we can continue to grow these numbers. Long before I started, DigitalOcean had made significant investments into providing value for our community of users and learners who aren't yet DigitalOcean users.
Our tutorials are considered the gold standard in the industry, and our community provides answers to thousands of technical questions. We engage with the startup community through our global startup program, Hatch, and we're founder of Hacktoberfest, the most popular hackathon in the world. This kind of developer love and engagement has created an amazing following in our developer community that has grown over the past decade. In the past few years, we've continued our focus on accelerated organic community growth with thoughtful content acquisitions complemented with internal content creation. As part of this accelerated strategy, we recently acquired CSS-Tricks, one of the largest developer education sites in the world. CSS-Tricks will now be powered by DO, and the site's 7,000 tutorial articles, reference guides, and videos will continue to be freely available for all developers.
In addition, we have rebuilt our paid demand generation channel by moving our advertising efforts in-house, resulting in the most efficient spend we have seen with dramatic improvements to traffic coming from this channel. Our paid demand generation, in combination with our organic traffic driven by our content, contributes to a large and growing top-of-funnel segment. Next, we move to the middle of the funnel focused on conversions. Ultimately, we want to increase our customer count by solving pain points in the sign-up and onboarding process to improve conversion rates. Once developers have engaged with our website for learning purposes or our emails, and we've established our credibility as a source of knowledge on a wide range of topics, we make sure that the rest of the onboarding and sign-up process is as intuitive as possible through simple UX, UI, and transparent pricing.
When developers are conducting their comparative research on our site or even setting up their very first Droplet, we ensure they're receiving simple and up-to-date information on our products, pricing, and what flat pricing or monthly caps mean for them. This entire streamlined process results in developers who are happy because they can easily create an account on DigitalOcean themselves. This principle of simplicity doesn't just end at enrollment, but it is carried out through the rest of their experience on the platform and with the product. Once on the platform, we also want to prioritize developers' time. They can be up and running with a Droplet or VM within minutes, and our app marketplace offers a variety of one-click installs as well. All of this enables developers to reach their goal of building applications more quickly.
It's important that our marketing team is working in lockstep with product and customer support teams to ensure a holistic onboarding. Our customer is going to be best served when the experience of becoming a customer also matches their experience once they are a customer.
My name is Simon Westlake, and I'm the CEO of Sonar. We make software for internet service providers. We started as a bootstrap company. It was just me and one other guy named Devin, who built all the software from the ground up. Six years later, we're at 70 employees, about 800 customers. You know, I'd been in the industry for a while, and a lot of the software that was typically in this space had to be installed on-premises. The management of that becomes a full-time job for somebody or a whole team. Especially when you're a very small start-up, you really don't want to be sinking a lot of time into dealing with all these unforeseen problems. The move to doing it in the cloud for us was just simplicity. We kind of ran the gamut of looking at the different options there.
Really what drove me away from the big, well-known names was just the overall complexity and cost of them. We looked at some other providers that were doing a similar thing to DigitalOcean, but I think what really got us with DO and what got us to stick there was the support that we got and just the clarity around everything. You can tell that they really think about making things simple when they're deploying a new product, and every time we pick one up, it's very easy to use. We've been involved in some upfront discussions about some of the new products with them, and, you know, they really take the input they get from their customers to heart, and that's the one thing you can really tell. We're using their virtual machines for hosting a lot of customer services.
I think we have 400 or 500 running in there today. We've also started using their Kubernetes service for our developers, and all our developers are working directly on DO every day, and that's become a pretty big part of our business with them. The services are great. We really never have any issues with them. In comparison to some other vendors that we use, just the level of support we get and the ease of access is really the thing that keeps us coming back. You know, we meet with them fairly regularly just to talk about what's happening and what's moving forward, and there's been a lot of situations there where they've given us good feedback or input on what's coming soon that's helped us make decisions.
I was talking to our first customer success manager at DO when I came in here, and I mentioned to her, you know, we're on our second now in six years with DO. Some other providers we use, we get a different manager every three months, and you just never get to build that relationship. There's value there, I think, in those ongoing relationships that you have. The people you do business with every day, that you're working with every day are critical, and the partners that you're working with every day for any of the services you get are critical. I mean, without hosting, we don't have a product. That extends to most of the services that we use, and I think I've always had the philosophy in business that the product is kind of marginally important, right?
There's a barrier you have to cross for people to wanna use it or buy it. Once you get beyond that, it's really all about the people, and that's really what comes through for us with Dio as well, is that there's good people behind it and we can trust them. They're easy to work with. You're not gonna get surprised with pricing, and the products work well. Frankly, for, I would say, probably 90% if not more of use cases, they've got everything you need there. The vast majority of people building software today, I think Dio's got everything you need.
We have made incredible progress to drive new account creation by updating both our website and our community site to encourage those users to sign up for an account, and by making the onboarding process clean and simple. By being highly customer-centric, we are continuously implementing changes that meet the needs of our customer base. One example of this was adding payment methods preferred by our global customers such as Google Pay and Apple Pay. In 2021, customer surveys found that our payment experience with net new users presented some friction with our customers, with our global customer base. Many potential customers did not natively own credit cards and struggled with the payment verification process. In late 2021, we ran A/B tests that found that a Google Pay integration would result in a 1%-4% uplift in conversions.
We rolled that out to our entire customer population. Now, up to 7% of our net new customer acquisitions use Google Pay when they first set up an account with us. This is a testament to our commitment to better serving our customers wherever they are, and a powerful case study of our team's approach to thoughtfully and rigorously identify, test, iterate, and make changes to better serve our customers. Our goal is to deepen the engagement we have with our customers through various nurture programs, depending on their customer lifecycle milestone. For example, we reach out to customers who are registered and not yet activated, and once activated, send them a pre-usage series with tips for how to be successful on our platform.
Once they are on the platform, we send a post-engage series with various educational resources to troubleshoot common errors to ensure a good ongoing experience. The last piece of our self-serve funnel is the bottom of the funnel or the retention and expansion stage. Ultimately, we want our customers to be successful and happy. We want them to attach other products and increase ARPU, so that they are stickier and find the most value on our platform. This is another area where we made tremendous progress. You can see here that the quality of our customer base continues to increase over time, with average month 4 MRR increasing 50% over the past three years. Our 2021 cohort starting month 4 at the same product multi-attach rate it took the 2019 cohort 17 months to achieve.
Our marketing team sends targeted upsell and cross-sell emails and in-app messages to customers based on their product usage. These include suggestions on the next product to try based on existing product usage or alerts that are nearing capacity limits for their virtual machines, with suggestions on how to resize or upgrade their Droplets. Last year, 40% of all customers who added an additional product received a specific targeted message from us a month before they upgraded. On the other side of the equation, we have also worked to reduce churn. Our churn prediction model ensures that early indicators of churn can be addressed, and we avoid losing customers by focusing on our at-risk customers before they actually churn out. We are now reducing customer churn by hundreds of customers per month from 2020 levels.
Looking ahead, we wanna dig even deeper into our customer journey to accelerate growth. As I said before, the self-serve motion acts as a moat for our larger DigitalOcean customers as we bring in more than 30,000 customers each month who will continue to fuel our overall growth. Now, our focus has shifted from the operational improvements in self-service to how we can enable our sales team to leverage these lower-paying customers and fast-track them into the scalers of tomorrow. Some of the areas we intend to bring even more focus to in the next years to turbocharge sales efforts are getting use cases right to help us better meet customer needs earlier and to further accelerate customer growth on the platform alongside multi-product attach. We know that product attach is driven more by customer use case needs than any one intervention.
Examples we are already working on include targeted campaigns for blockchain customers and video streaming customers. Leveraging our interventions and interactions to their fullest potential across our base. Given that marketing, customer support, and sales interactions appear to be complementary nudges alongside customer asks for new product ads. Sourcing future SMB customers early, supporting them in building their businesses, and participating in their growth alongside them. As I said earlier, 94% of our customers over $500 started small and grew on the platform. Bringing and nurturing this group over time with a human touch is a huge opportunity. DigitalOcean has a special place in the cloud market. We are, and will always be, for developers. These developers could be learners, gathering information about how to use the cloud in their own individual capacity.
They could be builders, working as part of a technology team within a business. Or they could be scalers, growing their businesses rapidly. The way we engage with customers, both on the marketing and sales side, all the way through to our product experience, is dramatically different from how the hyperscalers treat customers, and that's why our customers love us. We are owning our unique opportunity to be the only cloud provider that unapologetically focuses on helping the growth of modern digital-first businesses. Our mission is clear: We simplify cloud computing so builders can spend more time creating software that changes the world.
My name is Morad Mordehay. I'm the CEO of a company called Camcloud. Camcloud is a cloud video API built for developers that want to develop websites and video applications at scale. In order to process a video file, it takes a lot of computing power to convert it from one codec to another codec. We looked at all the solutions on the market. We took our ideas and plugged them into every cloud provider that we could find so we can find the absolute best combination of resources to fit our needs, and DigitalOcean was the hands down winner. Most of the other hyperscalers on the market didn't give us the flexibility and the enterprise level tools that we needed, and DigitalOcean gave us a very simple interface to build very complex solutions, at an affordable price.
Sometimes we process 8K videos and very difficult codecs that require a lot of compute resources. Whatever idea came to our head, DigitalOcean was able to handle anything that we threw out at it. We need a lot of computing power, which we get from the Droplets that DigitalOcean provides. We're able to ingest those videos and use economies of scale and parallel processing to transcode videos very, very quickly, 10x faster than real time, even for some of the most complicated codecs and resolutions out there. We want to offer our customers enterprise-level tools that are easy to use at a cost-efficient price point. DigitalOcean provides that to us directly. Pretty much from the very beginning, DigitalOcean was there with us, and since then, we've grown tremendously, tens of thousands of customers. We're transcoding millions of videos every week.
DigitalOcean has proven that they have the offering to help us continue to scale as we expand our business into new products. So far, there hasn't been any product that has come to our mind where DigitalOcean wasn't able to provide a solution and help scale that solution as our customer base grew. We love seeing DigitalOcean expand into new products, like when they started doing Kubernetes. Now they're rolling out new products to help with even more efficient scaling. We're excited to use those as well and see ourselves as being part of the DigitalOcean ecosystem for a long time to come. We really feel like we have a partner with DigitalOcean. They've been with us since the very inception of our business.
With the help that they gave us during the Hatch program, it enabled our business to become a real business that grew and generated revenue and made a dent in the market that we're targeting. The help and support that they provided to us was basically what allowed us to build a business.
For everywhere our customers are growing and whatever they need, we are with them. Whether they are learning, building, or scaling, DigitalOcean helps our customers get growing. This is our rally cry. It's a call to action, an invitation to our customers. It's our commitment to focus on growth, and it's our promise to our customers and the world that with DigitalOcean you'll get growing, whatever growth means to you. Now, I'd like to introduce our COO, Jeffrey Guy. He is truly my partner in driving growth and serving our customers. He will talk more about the next phase of our go-to-market expansion.
Today, I'm gonna talk to you about our maturing sales and partnership programs as part of our go-to-market expansion beyond the self-serve funnel. How we are growing our efforts to better attract, retain, and grow high-value customers in the builder and scaler categories. In the decades since DigitalOcean launched, our direct sales have been largely buoyed by the self-serve funnel that Carly described, which has allowed us to attract learners and builders to our platform. This efficient motion has helped us grow the business to more than 620,000 paying customers, and has kept our sales and marketing costs low. It's important to recognize that the developers of today, who make up a bulk of our customers, are the business leaders, founders, and entrepreneurs of tomorrow.
This path that developers take means we have a unique moment in time to make sure DigitalOcean is the default cloud for these learners, builders, and ultimately small to medium-sized businesses who are scaling. Our platform is a catalyst to this growth, and in the coming years, we intend to capitalize on this opportunity through several mechanisms, including, first, expanding on a builder and SMB-focused go-to-market motion, which will have a significant return on investment and is amplified by our competitive advantages. Second, continuing to focus on our value propositions of simplicity, predictable and reliable pricing, and excellent customer service remains paramount to nurturing and growing our existing customers and also acquiring new ones. Third, developing and scaling a partner model, which over time can provide outsized growth beyond our outbound sales motions.
Now I'd like to go through these sales motions in more detail and describe specific use cases where we've already seen success. Expansion of go-to-market motion. Our value propositions of simplicity, price predictability, and a strong customer support model already make DigitalOcean a valued partner for small businesses who want solutions tailored to their specific needs. As we develop and expand a holistic and intentional go-to-market motion, we are taking the entire customer journey into account, starting with the marketing efforts you heard about and through statistical data science algorithms such as propensity models that identify builders and scalers for expansion opportunities. Beyond digital nurture campaigns, our customer care and success teams migrate opportunities to leads for our highly technical solution engineers and account executives. As a unified account team, including service and support professionals, they support the customer's desired outcomes as they progress along their business journey.
Most recently, we have defined processes and unified the entire structure of our customer-facing teams to account for this evolving customer journey continuum from onboarding to growth through nurture and scale. Specifically, we are targeting industries where we have proven product market fit and where customers will experience growth over time through their business journey. These include blockchain, streaming, gaming, ed tech, and many others. Once we identify a particular vertical that is growing amongst our customer base, we run targeted acquisition campaigns and incorporate A/B testing that is validated by data science to identify our strongest campaigns. This customer intent is further amplified by human interaction from our technical and commercial teams. We find the combination of digital nurture and human interaction to be complementary to the overall customer experience.
Expansion occurs via increased volume of compute resources as well as with multi-attach of additional PaaS products like load balancer as a service or database as a service. One customer who exemplifies this journey that started out with us in the sandbox environment and ultimately wanted a more curated experience on their cloud platform. This customer started out at $6,000 of MRR in Q2 of 2021 and has worked closely with our solution engineering, sales, and customer success teams to grow their business over 100 times that amount, currently trusting DigitalOcean with over $600,000 of MRR of cloud deployment per month in less than three quarters. This is just one example of many that showcases how builders remain the champions within our customers and value simplicity coupled with our expertise and advice.
One of our streaming customer's CEO said it best, quote, "With DigitalOcean, you can create real-time live streaming at scale coupled with performance and reliability you can trust and a nice price tag to match." This type of net new outbound motion has inflected the curve on ARPU as we have experienced a 180% year-over-year increase in average deal size via sales while growing the contribution of sales to revenue by 235% year-over-year. While the notion of pure outbound selling is new for DO, this is an intentional go-to-market motion that is highly targeted and focused. We have deployed several tools for data enrichment and have dedicated outbound sales personnel to advance this lever. Currently, 20% of our sales is net new, and we intend to increase this to over 50% over the next two years.
These customer examples and motions are an output of brand plus technology equaling a significant competitive advantage. This is evidenced by our greater than 50% win rate in competitive situations. With the vast opportunity in front of us and the successes we've seen, we are gonna be doubling our sales team this year, as we have seen high efficiencies in productivity with each ramped account executive scaling towards $2 million of incremental annual revenue. With the additional sales efforts, we will be increasing the quota-carrying headcount ratio of the company, and ultimately, touch sales will be more than 20% of our annual revenue as a company by 2025. In addition to developing and expanding our outbound sales motion, we are also focused on helping new and existing customers scale and achieve success through targeted data-backed nurture efforts.
Not only do our customers stay with DigitalOcean with an NDR of 117% and NPS that is best in class, but they grow with us. Another customer who demonstrates this fact started out with us through the sales acquisition funnel, where we overserved them from the onboarding stage as a net new outbound prospect. This customer grew from $4,000 of MRR in November last year to over $200,000 in April, and they are still growing with us with active deployments in all of our data centers. We achieve this by being intentional about nurturing this customer and the associated multi-attach to allow for prescriptive growth with our customer success, technical account manager, and service offerings. Soon, we'll have even more opportunities to accelerate adoption among our customers.
This month, we are standing up our customer care function to ensure every one of our 600,000+ customers has a designated contact as they scale on DigitalOcean. This motion will allow us to build a retention and expansion core competency with 90% or more of our revenue covered by our account teams. These teams will continue to leverage data science, triggers, and alerting to help predict growth and opportunities for cross-sell into our sizable base. Our usage model allows for low cost adoption with human influence, which allows us to drive best in class revenue per employee. For our customers, the ability to connect direct with a representative at DigitalOcean is a large differentiator from hyperscalers and other competitors. I actually had a meeting with a customer last week that has their environment on a competitive cloud who didn't even know their account team.
Here, they have access to our entire commercial leadership team, coupled with technical experts to help them realize value. Efficiency, simplicity, and cost drove them to explore DigitalOcean, and the human touch keeps them here.
My name is Simon Kemper, or in Germany, Simon Kemper. I am the CEO and co-founder of Datacake. Datacake is a multipurpose IoT platform. That means we are providing an infrastructure where customers can bring their physical devices to the digital world using sensors and connectivity. As we started with DigitalOcean, what we were looking for is an infrastructure that we could use, where we could scale on, that helps us to grow, the Datacake and concentrate on the actual application. As a small company, when we started, pricing was super important for us, and we had the problem that with other services, the pricing was not clearly stated on the website, and it was hard to find out where we would end up with the pricing.
Working with DigitalOcean gives us a good feeling about transparency, and they communicate what they are, for example, charging and prices, and we really feel that this is an open and transparent place. When we started, we originally used managed Kubernetes and also Spaces where we can store data uploads from our customers. When we grew, we switched to managed things like managed Redis, managed PostgreSQL databases and so on. On a day-to-day basis, we are using DigitalOcean mainly through the web interface, but also the developers are using the command line tool where we can orchestrate the things on DigitalOcean. One of the things that we like about DigitalOcean is that we can scale easily and auto-scale our services.
At the time when we switched from our previous solution to DigitalOcean, we had, like, 10 users sign up on the platform, and by that time, we also decided to turn Datacake into an SaaS software service platform. Right now we're having more than 10K users, so our daily business has increased. This means we're communicating with customers, we are helping them, solving their problems that they have with Datacake or their use cases. That means that for our daily business, simplicity is super important. DigitalOcean is helping us to grow our business in the way that we can really focus on the application. It's all managed by DigitalOcean, and we could just move ahead and grow.
We are super happy that we have, like, a customer support manager that is in direct contact with us, and this is also super important. If we have an issue or something like that, there's always someone at the DigitalOcean side to take over our conversation. For startups and other companies, or everybody that is interested on the cloud infrastructure, DigitalOcean is a great fit because compare it to different products that are hard to maintain and you need, yeah, some kind of guidance to really find what you're looking for, DigitalOcean makes it super easy to spin up a server, a service, or whatever you're looking for. For the future, Datacake is looking to grow even further, to extend to more countries and also to bring more customers, on the platform.
If we continue in this way, that we quickly scale and grow the company, we will continue with DigitalOcean, because of the contact that we already have with the customer support manager, but also the product, the quality and the of the product itself, and also the continuous development of new products that actually fit to what we were requesting and what we need from the market.
The combination of targeted outbound sales combined with nurture efforts that identify those customers who are in the builder stage and moving towards being scalers means that we can find and work closely with customers who are likely to grow to spend thousands of dollars per month and more. To further the customer experience, we have a series of support and service offerings on the horizon catered specifically to our SMB clients. Another way we are supporting these scalers is through sales and operations planning or capacity management, where data center expansions proactively support the demand shocks of our rapidly growing businesses, ultimately ensuring they have the resources required to scale. We work closely with this high spend, high usage group of customers to anticipate their product mix, infrastructure and platform needs, and can then adapt our own expansions to fit those needs.
These enhancements include opening additional data centers and/or points of presence in locations most requested by existing customers in Southeast Asia, South America, Middle East, and Africa. Another go-to-market motion, in addition to deploying outbound sales processes and nurturing existing customers through retention and expansion motions, is partnerships. We have laid the foundation for partnerships as the lever to drive revenue growth. DigitalOcean simplified products and price predictability mean we are ideally suited to serve small businesses of all types. By partnering with digital agencies and managed service providers who resell DigitalOcean to their customers, we reach new customers that don't have in-house developers to manage their own cloud solution. We know that the channel and partnership opportunity can be a powerful accelerator.
Identifying select partners will help us drive efficient pull-through and will create access to multiple new customer populations within specific verticals that were either inaccessible or under-penetrated in the past. Among our 100+ recently developed partners, there are three major areas. Managed hosting providers, platform or app builders, and digital agencies. Managed hosting providers layer on value-added services for less technical customers while platform app builders create abstract solutions on our platform where they directly sell to their customers. Lastly, agencies, of which there are over 200,000 in the world, cover one-off project work for clients. When it comes to successfully growing our partner network, there are four major pillars we anchor on in addition to the creation of our ideal partner profile. For all partners, we provide a single point of contact.
Partner development managers will act as the single point of contact to manage relationships and simplify mechanics. A dedicated marketing partnership. We have a dedicated channel marketing role to quarterback specific co-marketing efforts. Sales support. Active engagement from DO sales team will supplement partner sales teams to develop strategies, represent the solution, and close more deals. Next-level service. High-value partners and their customers will have a dedicated customer success manager to provide personalized product support. Partners are also not limited to channel partners. We have accelerated our managed database offerings to market with alliances across MySQL, MongoDB, and Aiven. When you add in the marketplace opportunities Gabe referenced earlier, we will become even more a prescriptive resource to our customers. In our path to multiple billions of revenue, we are targeting partnerships and channel influence revenue to double year-over-year for the foreseeable future.
In addition, we will deploy two more strategies to expand revenue growth. First, we will leverage monetization tactics to package and price our product offerings to provide simplicity to our customer base while maximizing revenue where propensity to pay exists. Our recent packaging changes to the Droplets better serve our learning, building, and scaling personas is an example of a monetization lever. Beyond build or partner options, inorganic expansion to our capabilities and/or products along the existing growth levers will also be utilized to stimulate growth. Product extensions and new product capabilities will have a build by partner lens with a strong bias towards velocity. It's important to remember that we are still at the very beginning of our efforts around nurturing through inside sales, outbound sales, partnerships, monetization, and inorganic expansion.
As we continue to hone these core competencies, we believe that we will accelerate persistent growth throughout the builder and scaler customers. Thank you for joining us today to learn more about our numerous go-to-market initiatives that we are investing in. We are excited about the opportunity ahead, and we look forward to updating you on our progress in the future.
I'm glad that you've had a chance to hear from our impressive leadership team on how we intend to achieve our growth objectives to scale this business with product development and sales and marketing enhancements. In many ways, we believe we're just getting started, and with the rich roadmap that we've shared, we hope that the investment community can see our path forward from here. Before outlining our plans for the future growth of DigitalOcean, I'd like to provide you with a bit of context as to where we've been, which creates the foundation for confidence in the journey ahead. I joined DigitalOcean in the summer of 2019, nearly three years ago. During that time, the company has been transformed from top to bottom, including our product breadth, our marketing acumen, and only recently, our ramping sales motion.
The financial profile of the company is no different and has markedly improved from over $250 million in bank debt and cash flow negative to a publicly traded company with a market cap in the billions, over $1.5 billion in cash on the balance sheet, and generating real free cash flow. Perhaps most importantly, we've seen growth accelerate meaningfully from the mid 20s to 30%+ . We're often asked how we were able to achieve this transformation. Please let me briefly walk you through the tactical improvements we've made. Our growth formula is derived from three primary sectors, NDR, ARPU, and customers. Beginning with net dollar retention or NDR, this is our foundation for durable growth, and we've improved this metric substantially from 100% to better than 115% in each of our last three quarters.
For a business that is comprised of learners, builders and scalers, this is an excellent indicator of long-term success. We've increased NDR through a deliberate focus on broadening our product portfolio, engaging high potential customers through direct outreach campaigns and improvements in customer support. At the same time, we have materially reduced churn to 10%, best in class for a small business-focused company. Aiding our NDR has been revenue per customer or ARPU. This is the second key driver of our top-line growth, and it has grown as we've introduced new products such as database offerings, App Platform, and Kubernetes, and most recently, serverless. We've been growing revenue per customer well above 25% for the past year, and we see this trajectory continuing and likely accelerate as we introduce new products and packages to the builders and scalers that Gabe referenced.
Finally, rounding out the growth algorithm is customer growth. As you've heard Carly share, we attract millions of visitors every month to our website, and tens of thousands sign up each month as customers. Today, you've heard about the enhancements we have and continue to make in order to attract new customers to DigitalOcean and then nurture them with our sales motion. Of particular importance is the focus we are putting on our higher-spending customers. These are the customers that spend more than $50 per month, and we are prioritizing investments to attract, convert, and retain these customers as they are the drivers of our business. These customers come from a variety of verticals, including gaming, streaming, blockchain, SaaS, and web agencies. As you can see from the testimonials today, they love DigitalOcean for our simplicity, support, and efficiency.
The opportunity to continue to add them to our platform is virtually limitless. With that background explained, I'd like to share our thinking with you as to how we are going to durably grow our business greater than 30% for the next several years by building upon and expanding on our growth investments that have helped us get to this exciting juncture. For us, it begins with the customers, and specifically the high-spend customers. These customers are growing 20% annually, and their revenue has increased significantly as a percentage of total revenue over the last few years, as it has been growing much faster than the total company revenue. We don't see this trend slowing down, and we are laser-focused on attracting more of these customers through the investments Carly outlined, and nurturing and targeting them through the motions Jeff described.
We aim to continue to grow the builder and scaler customer cohort well above 10% over the next several years in order to drive top-line revenue growth to $1 billion. These customers are critical to our success, and accordingly, going forward, we will report only this customer count rather than the total customer count, which includes so many learners. Annually, we will report total customers, but we want the external community to have a singular focus on the customers who are driving 84% of our revenue, up from 72% in 2018. The next factor for 30% plus durable growth is increasing revenue per customer. As Gabe shared, we have a rich roadmap ahead to supplement our already impressive product offerings. As you've heard us say today, just a few years ago, our revenue was entirely infrastructure-related.
As we have introduced managed databases, managed Kubernetes and App Platform, we've seen ARPU grow faster and faster, driven by higher multi-attach rates. Now, with the launch of serverless and other coming introductions, we see a long runway of continued growth in ARPU and see revenue from these products growing from roughly 11% today to 20% over the next several years. As Carly shared, the quality of our customer base has made significant improvement over the past several years. Specifically, the average spend of our customers has accelerated dramatically, and these cohorts demonstrate the improving ramp we have seen as we attracted more high-spend customers and broadened our product set. Our objective here is clear, to augment our self-serve funnel with sales investments that Jeff discussed.
We believe that we can significantly increase our percentage of revenue that comes from a traditional sales model, and that will be a powerful growth lever. These customers that come onto the platform from the sales motion have day one ARPU that is $5,000-$6,000, have NDR that is above the company average as they expand faster and churn less. As we add these customers onto the platform at an increasing rate, revenue growth will benefit. Before sharing our long-term financial projections, I'd like to share our thinking on improving efficiencies that will help us achieve these healthy targets. Starting with gross margin. While we've made great improvements here, we have room for further gains. Gross margins have been trending higher as we are improving the performance of our servers while concurrently managing our costs with an aim for more efficiency.
A few years ago, we were able to negotiate meaningful savings from our server manufacturers. That certainly helped drive gross margin higher, but that is only one piece of the puzzle. We've also been improving our server performance, which has improved our return on invested capital. We have a very healthy return on invested capital of over 700% on our servers, and we will continue to improve gross margins as we improve costs on other components of our cost of goods sold. Similar to improving our server costs, we see opportunities with our data center providers and chip manufacturers. From an operating margin standpoint, we see leverage opportunities primarily in G&A expense. We currently have G&A spend that is approximately 20% of revenue, which has room for improvement.
We're going to lower G&A expenses by controlling headcount growth in these areas, greater efficiency in our supporting technology spend, lower processing fees, and decreasing bad debt. Finally, capital expenditures is where we have made the most improvements in the business, and I believe there is more to come. While this is a capital-intensive business, we've managed this expense from more than a half of revenue to roughly a fifth. We see further opportunity to lower this spend by working with our providers as we become a larger customer and give them insights into our growth objectives 6-8 quarters in advance. Long term, we see CapEx being below 20% of revenue through lower server and component costs and greater revenue yield per server. From a long-term perspective, we believe DigitalOcean will have a very attractive financial profile.
Beginning with growth, we're targeting long-term revenue growth of greater than 30%. We plan for gross margins to approach 70%. We see research and development of approximately 20%. We anticipate sales and marketing to continue to be best in class at 10%-12% of revenue. Finally, we expect G&A expense to materially improve towards 10%. Altogether, the business model will enable us to drive an operating margin approaching 30% in the long term. The combination of improving margins and efficient capital investment will drive DO's free cash flow generating capability towards 20% of revenue. We have our sights set on continuing to build DigitalOcean as a financially sound, durable company and can't wait to continue the work necessary to live up to our long-term potential. As you can see, we're excited about the future for DigitalOcean.
The company has been transformed these past few years, and we think we can continue that transformation with enduring growth, improving efficiency, and ramping free cash flow. Thank you for your time today, and we look forward to discussing our progress with you in the near future.
Good afternoon, everyone, and thank you for joining us for our Investor Day. Hope that the presentation that we shared earlier was valuable and that you learned some good insights about our business. We've now received a number of questions from our analysts that cover us, and we're going to start answering the questions from our executive team here on my left. Beginning with the first question, which came from Raimo Lenschow of Barclays. His question is, your initial database offering of just one database grew very strongly in the first quarter. Does this strong momentum motivate you to look at more database options besides Managed MongoDB? Perhaps, Gabe, you could take that one.
Yes. Yes, it does. You know, look, it all starts for us with our customers and their jobs to be done. You know, as they look to build new digital products, they're always looking at new database engines, and we're excited to do more there.
Excellent. Excellent. The next question comes from Pinjalim Bora at JP Morgan, and he asks, this year, you are doing a sizable price increase, but it seems like you are keeping the $1 billion goal at the end of fiscal 2024. Are you baking in some conservatism from the current macro environment, giving you some degrees of freedom to drive upside if the macro headwinds don't pan out? Yancey?
Well, I mean, I think first off, we set expectations we intend to beat. We've been consistently beating the drum on that over time. Second, the marker of $1 billion in 2024, we said, you know, we don't expect it to be $1.0 billion on 31st December 2024. We think we're and hopefully, we've highlighted some things today to give you a sense of our focus on growing through $1 billion at 30% or better growth. You know, I think this price action we've taken, which will, you know, go into effect on 1st July , is but one of the many levers that we're pulling, whether it's new product, go-to-market motions, getting more efficient on infrastructure, better features, functionality, getting more efficient in serving customers, community, etc.
All of those go into the billion and beyond strategy.
Great. Thanks, Yancey. The next question comes from David Hynes from Canaccord, and he asked, "I would love to hear you talk a little bit about the channel MSP opportunity. What are you doing to cultivate that, and how and when might it be more meaningfully contributing to revenue?" Jeff?
Sure. We're being methodical about all our growth levers that are out there. Again, we started and built a business with the self-serve funnel. Following that, we've done a lot to drive NDR across the business and fine-tune our inside motions. We're now setting up a more aggressive outside sales motion, and partnership is yet another lever that we can drive. The current situation is that we are building the process tools and technology Q2 and Q3 of this year to have the infrastructure available for us to scale out the partnership network. There's several hundred thousand partners for us to attract, and what we're working on now is what is the ideal profile of that partner.
Once we have the ideal profile established, we'll be out there attacking those several hundred thousand partners.
Excellent. Thank you very much, Jeff. The next question comes from Pat Walravens at JMP, and he asks. Maybe this one's for Bill. What are the key drivers to achieving 20% free cash flow by 2024? How should we think about the trajectory of free cash flow margins over that timeframe?
Well, over the next 2+ years, we see the achieving 20% sort of on a progressive up and to the right fashion as we've basically achieved that over the past 24 months. The real drivers there continue to be driving the overall cost of the component elements and the servers that we buy lower by virtue of us being a bigger and bigger customer. More importantly perhaps is our efforts around driving incremental revenue through those servers, the work that our infrastructure engineering teams are doing to continue to increase the overall revenue through capacity utilization in terms of how we sell the availability that we have on our servers using new technologies to power more through those servers. Constantly focusing on driving more and more revenue out of the capacity capabilities that we have.
We still see more runway there, and you'll see it to be a progressive, a continued improvement towards 20% over the next 24-36 months.
Great. Thank you. The next question comes from Gabriela Borges from Goldman Sachs, and she asked this of Yancey: How is the near-term demand environment? Is there any risk of churn for the folks who started pet projects during COVID that are now back to work? And what about crypto? Some industry folks think that crypto price correction is a leading indicator for less developer activity. How do you think about this?
What we're seeing right now is, you know, business for net new customers, as we talked about last month in the earnings call, whether it's our sales motions or self-serve, was robust in early May. That continues. We're seeing new people attracted to our platform, which is a good indicator. You know, a couple years ago, we worked through a recession, a deep recession, with the onset of the pandemic, and we found that the customers are small, medium-sized businesses. Our customers are businesses that are scaling are pretty resilient.
Yeah, they may have some fluctuations in the business, but as you're seeing, and there's a lot of evidence, we just put out a report today about the resilience of these new people quitting jobs and starting businesses, and we fit right into that in terms of our cloud capability. When we gave guidance last quarter related to the year and the quarter, you know, we talked about the uncertainty in Ukraine and Russia. That obviously is playing out. We are seeing, you know, some slowing growth in Europe and Asia, but we remain comfortable with our outlook. What we're focused on is a great period of time, you know, when there's a lot of uncertainty, is to stay close to your customers.
It worked for us in the last recession, and we think that's a great formula. We're launching new products. We're launching a lot of new capabilities. We're engaging more with our customers. In my experience, that's the right formula to sort of work through a tough time is with your customers and give them innovation. Keep investing, give them confidence that, you know, we're here for the long haul, which we are.
Terrific. I've got a number of questions about pricing, so I think we should perhaps split them into a few different ones. The first one's, I think, a quick hitter for Bill, and it comes from Josh Baer at Morgan Stanley, and he asked how do the pricing changes impact your guidance for 2022?
Well, we provided the guidance for 2022, as you know, months ago. We reaffirmed those during our last earnings call. We believe that the price increase will be net accretive to us over the course of the year, but we've not changed our outlook for the year. Even though we indicated some softness in Q2 from the guidance that we provided, we still feel very comfortable relative to the targets we've put forward. Given the multiple things that we have basically put in place, the new products that Gabe was talking about, the work in sales that Jeff was talking about, the increased UX that Carly was talking about in video, all of these are showing positive traction.
We've seen those results already, so we're very, very comfortable for the numbers that we've provided for the year, and as always, we'll look to overperform.
Thank you. Wamsi Mohan at Bank of America also wanted to understand a little bit about the strategy and the thinking behind the price changes. Gabe, maybe you can take that one.
Sure. You know, we're a very different company today than we were, you know, 10 years ago. We cater to lots of different customers today, not just, you know, learners, you know, folks who are, hobbyist developers. You know, we're looking at a pricing strategy that is reflective of the diversity that we see in our customer base today. We have a lot of different levers that we look at. Monetization and sort of pricing changes are just, you know, one of the levers that we look at when, you know, adapting, you know, sort of to this new customer base. You know, that was really the strategy. You know, it was in the works for a while.
You know, this is something we've been looking at for quite a bit of time.
Yeah. Terrific. A number of people also asked about CapEx, and so, I can attribute it to Barclays as well as Canaccord. The question is, maybe perhaps for you, Jeff, you know, what levers do you have to get CapEx under 20% over time, and how should we think about growth in investments for growth or maintenance CapEx?
Sure. We're super proud of the progress we've made on capital efficiency. At one point in time, it was half of the revenue of the business. We've pushed that down to almost a fifth, and our long-term target is less than 20%. There's three tenets behind our capital efficiency program. The first is a strategic procurement effort. We have a strategic group working closely with our infrastructure. They're working on setting up not only one, but dual source and triple source, so we can drive additional cost savings out of our procurement effort. Strategic procurement's a huge pillar in driving down our capital costs. In addition to that, the capacity utilization.
We have an engineering team that's working hand in hand, looking at the workloads coming through the platform and how do we maximize the utilization and effectiveness of our platform. The last part is really our shift from IaaS products to PaaS products. As we have more PaaS products that come on the platform, they consume less CapEx. It's a combination of these three together and cross-functional teams that are driving our capital efficiency to our long-term goal of less than 20%.
You know, that's very helpful. Thank you, Jeff. Question that we got from Tim Horan at Oppenheimer, and this one's probably for you, Gabe. What percentage of revenue can serverless be in five years, and how difficult was it for us to roll out? In other words, you know, can our competitors launch this service very easily? Maybe you could talk a little bit about the serverless launch and, you know, how we see it attaching.
Sure, yeah. Yeah, we're really excited with what we're seeing so far. We've got 1,800 customers actively using the serverless products since GA, which is a you know really great stat. You know, shows the engagement and the interest. You know, we look for anything that we do in the space to contribute to you know 2-3 points of growth as we look out a few years. That's the case here. I won't get into specifics around you know what the revenue targets are. I will say I'm really proud of how you know we went about the acquisition and the integration. It's challenging to integrate companies.
You know, as we look to do more inorganic moves in the future, I think what we've been able to do with Nimbella, you know, acquired them in September of last year. We got a product out in GA pretty quickly thereafter. I think it is a good sign for what we're able to do, and I expect to do that going forward.
Awesome. Carly Brantz, perhaps you can take this question. It's another one from Wamsi Mohan at Bank of America. He asks, "What has driven the increase in UVEX, or unique visitors? And, the conversion of these into customers is slowing. What explains that?
Sure. A couple of things. I think this is on. The first one around what's driving the increase in UVEX, it's really two pieces to that. The first is we brought all of our paid demand gen efforts in-house, and once we got that in working order, we were able to scale that. We're seeing the most efficient return, driving a ton of traffic from that channel than we've ever seen before. In addition to that, we have started to do some thoughtful content acquisitions. I spoke in the video a little bit about CSS-Tricks, which doubles our content, and tutorials and other learning resources. That is another huge one, and we'll continue to pursue other content acquisitions like that in the future.
On the conversion piece, it might sound a little counterintuitive, but our primary goal is not conversion. A key differentiator for us at DigitalOcean with our community is creating value for all those learners who are coming, and we see that it's a long path sometimes to conversion. Developers are coming to our website, they're learning new skills. They might leave, and once they're ready to build their next idea, they'll come back to DigitalOcean, and that is why we have such strong love and loyalty among the developer community.
Awesome. Next question comes from Michael Turits at KeyBank, and he asks, "Can you discuss a little bit about the product roadmap around AI ML, and maybe how you envision those capabilities fitting into the broader product suite?" Gabe, I think that one's got your name all over it.
Yeah, sure. You know, I talked about this a bit in the video, but I think it's a really interesting trend with artificial intelligence and machine learning, this idea of digital products that get better the more that they're used, thanks to AI-enabled insights. We have customers like EasySize on the platform today who are a great example of customers on DO already who are emblematic of this trend. Yeah, I think one of the opportunities that we have in this space is that, you know, a lot of machine learning and AI technology today is really the domain of the tech elite, right? Big companies, large enterprises, they're able to get access to this technology, but small to medium-sized businesses don't, right? They're sort of cut out of that.
I think, you know, through simple products and the addition of AI capability, the addition of machine learning capability that doesn't require data scientists to be trained up on staff, there's a real big opportunity for us to take the trademark DO simplicity, apply it to this area, and help our customers be at the forefront of creating AI-enabled digital applications that will be defining the future.
Cool. Gabe, maybe have a sip of water because I got a three-part question for you. So this is from Jim Fish at Piper Sandler, and he asked, "Just please discuss the movement towards edge and edge spending getting to $274 billion. However, how will DigitalOcean, which is more centralized cloud, take advantage of that spend? And what percentage of that market do you address today versus five years from now? Wouldn't you need to spend more on CapEx as a result to keep up with some of the other edge vendors? And finally, how do you think about Akamai and Linode competitively, given the combination and announced roadmap that they shared recently?
Well, gosh, that is a mouthful. You know, just first taking the Akamai Linode question. There was recently an interview with the CEO of Akamai on, I think it was Protocol, where the CEO describes Akamai and Linode's combined future as being intertwined with enterprise. You know, it wasn't the first time they said that. There's really lots of examples of this idea of sort of Akamai and Linode and the history there sort of taking them away from this SMB direction that, you know, is DigitalOcean's core focus and will continue to be our core focus. I don't really see us overlapping in terms of capability, customer, you know, conflicts over time.
In terms of the broader question around edge, I do think this is a really interesting question. 70% of our customers today are already using CDN capabilities, so we know that there is, in fact, demand for edge solutions within our customer base today. How we go about bringing those capabilities to the platform, well, I mean, anytime we have that challenge, we look at a range of build, buy, partner options. You know, everything is on the table for us at this point. I do think we need to figure out how we can move quickly in this space, and you know, look at is there an opportunity for us to use inorganic growth to really accelerate our path there.
Thank you. Yancey, this one's for you, and it comes from JP Morgan. Yancey, since you joined a few years ago, you have put in motion a lot of changes across the board, from top of funnel to customer success to CapEx optimizations. We've seen the positive results flow through over the last several quarters. As you look back, would you say that there were one or two things that surprised you to the positive relative to what you had expected, and one or two things that there are still ahead?
Yeah. I don't know about surprises, but you know, I was taking a look at joining. One of the things or maybe the most attractive thing was the brand perception of DO in the marketplace. You know, as I was onboarding and meeting with customers, I'm sure the folks that work at Apple when they talk to their customers probably sense this, but our customers really love our company, and I think it's because of what we enable them to do. They're just so grateful that they can come to a platform that is simple, easy, intuitive, and you know, offers them support, et cetera.
I think I don't know that I was surprised by that, but it is very powerful, and I think some of the things that we've been able to do, and it's flowing through the results, both the financial results, but also our employees love working here, and I think there's a relationship to that. I think that was probably the one thing that sticks out as, you know, how have we been able to get the company turned, frankly, so quickly, and off on, you know, talking about $1 billion-plus over time is probably that. You know, I think Gabe talked a lot about the product strategy, and I think where we're coming now is the product innovation and the velocity of that innovation is the thing that we've been working on.
It's a lot of people in technology businesses work on that, and we're really turning the corner on that. I'm excited. We've shown. You know, we've gone from 0% of revenue just three years ago in PaaS to now 13% last quarter, which is a huge growth wave, and that's gonna continue to be a growth wave. We've shown that if we add relevant and the correct sort of product mix to the platform, we can, you know, grow share of wallet with customers. Now, we just need to do it more consistently with velocity. We're on track to do that. This next wave, I think, that's gonna propel us through $1 billion on the product side is really gonna help. We got the go-to-market working well, so they're ready to sell it.
I think that's the next thing that's to look for with our business, both organically and as we've shown through Nimbella and serverless, Nanobox and App Platform, our content strategy. It's not gonna be all built here. We'll leverage the balance sheet with M&A.
That's helpful. A number of people have asked about our sales investments, so there's too many to attribute just to one person. But a couple of the questions, maybe Jeff, you could start, are about, you know, is the sales investments gonna require, you know, or is it gonna be a step back in margin as a result? Also, how do we think about the lifetime value of a customer relative to the spend to get them with a touch sales motion versus a self-serve model?
Sure. Again, the business was built on a very strong and sophisticated self-serve motion. We've spent a lot of time retooling the support and success organizations to be able to complement the digital nurture that's occurring within marketing. We're seeing a lot of power between the human connection with customers, with solution engineers, and with account executives, along with the digital nurture campaigns. In doing so, we will strengthen our inside sales motion and have more human touch all the way through so that we can help accelerate the customer's velocity through their developer journey. At the same time, help them adopt products faster and have more multi-attach that's going through the process.
In addition, as we look at a lot of the scalers that have moved beyond $500, which is just a few percent of the volume of customers that we have, we see a lot of like use cases and personas and use cases. As we've identified and stratified these out, we now have a sales team that's in EMEA, in APAC, in the Americas, that can go find other like customers. Therefore, when they come onto the platform, they're coming on with a higher ARPU than an initial self-serve customer. Now, we have metrics wrapped all around this. If Carly can have paid marketing that has a higher return of LTV to CAC than I can have within a sales motion inside or outside, all of those are competing dollars of investment for us to drive revenue.
We will enable self-serve plus monetization, plus inside sales, plus outside sales, plus partnerships and products in inorganic are all the growth levers we have, and we will adjudicate between them to maximize our revenue growth.
That's really helpful. This one, Carly, I think, was perfect for you. It says, this is from David Hynes at Canaccord Genuity, and he said that when you look at the cohorts from 2014 through 2017, they have not grown as strongly as in the more recent years. Those older cohorts are still about 50% of revenue today. How much opportunity is there for expansion in the older cohorts? Could you catalyze growth in older ones to look more like newer ones?
That's a great question. I think there's incredible opportunity and growth in all of our customers across the 600,000. We're seeing in more recent years that we're bringing folks onto the platform, we're growing them more rapidly over time. I'd still think that yes, a lot of those customers from 2014- 2017, they don't know a lot about the offerings that we've had since they joined. We have done a lot of propensity modeling of figuring out which product combinations folks have and going back in to make thoughtful recommendations. We'll continue to do that with new products on the horizon as well. We can go back in, 600,000 is a ton of opportunity to cross-sell and expand.
Great. Bill, I think this one is for you. You said you won't give total customer count anymore. Why not continue to provide that and the higher paying cohort? Will you give us the ARPU associated with the learners or revenue from learners, or will there be a way to reconcile users times ARPU?
What we've provided today is a lot of information relative to how we look at over 600,000 customers. One of the things that we're always asked from investors is what's really driving growth. What we're trying to provide is deeper insight into indeed what is driving growth in the company, and also giving you an idea of how we're approaching that growth. By looking at the market, more specifically against the learners, builders, and scalers, we can target our initiatives in product, in marketing, and sales to increase the likelihood of success of selling to these customers, gaining a foothold with them, and keeping them over the course of their life.
Understanding their growth and the fact that they're growing faster at higher NDR, higher ARPU growth, we think that is the most important metric for you to be looking at in terms of where it grows going forward. Now, annually, we will continue to provide the total customer count, and you've heard Yancey talk and all of us talk about 10% growth, which is our target. We expect the $50 and above to grow faster than 10%, and everything we're doing is meant to do that. At this point in time, we provide NDR and ARPU and those metrics on the group overall. Down the road, we'll look at providing additional detail to help you further understand.
We think this is an important step in the right direction in terms of giving you insight in how we're approaching the business and give you the confidence that we're looking at the places where we think the product will continue to resonate and allow us to deliver north of 30% growth on a consistent basis.
Good. This next one might be perfect for Yancey, but or maybe others. This is from Raimo Lenschow again. There are differences in terms of the maturity of cloud adoption in different geographies. What are the opportunities in currently under-penetrated markets like the Middle East or Africa?
You know, I think the beauty of our company is when you look at the 620,000+ customers, they're spread around the world, you know, pretty evenly according to, you know, roughly according to GDP in those regions, Asia, U.S., Europe, Middle East, Africa, etc. I think that speaks to what we enable. We enable people to come to DigitalOcean and get their ideas on the internet, whether they're just testing them, they're learners figuring things out, or they're the future builders or scalers who are actually got an idea into customers' hands, and it's they're building a business or scaling a business. We enable that at a really low cost opportunity. You know, people, $15 a month, many of the learners come to the platform and then scale to really large levels.
I think, you know, our investment in community, you know, the millions of people that come from around the world every day to learn, to grow, they come to DigitalOcean to do that. We're enabling that. The products, as the products even continue to get better and more diversified, we give them longer runway to spend more of those dollars as they are growing, because we can capture more of their workflow as they grow employees. Our strategy, you know, it's an interesting business because when you look at revenue, you look at customers, we're already sort of nominally global. It's not like the typical company a year after the IPO that is 90% U.S., and they have to go build sales to build out a presence in the rest of the world.
We're already in the rest of the world. How we make decisions, for example, we just announced a data center that we're gonna launch later this year in Australia. We're already in Australia. We have lots of customers in Australia, but it's a high-growth tech market. We see opportunities to grow even faster, potentially, than that market. We're looking at other data centers across Middle East, Africa, South America, where we have presence, we have customers, but we think we can grow faster by adding those capabilities. Jeff mentioned adding sales capabilities in the three regions as we look at them. Those are gonna help us grow faster in some of these under-penetrated markets.
It's an interesting question because we're already global, so it raises the bar on marginal investment, because we gotta be judicious about choosing one versus the other because we have to have conviction if we're gonna be able to over-penetrate in the market because we're already there.
Gabe, I have another product question that just came across from Pinjalim Bora at JP Morgan. He said, "We've heard some focus on AI ML use cases. Should we expect a GPU offering from DigitalOcean at some point?
It's certainly one of the things that we're looking at. You know, GPU is definitely synonymous with AI ML. I think as the question points out, also synonymous with things like video streaming and, you know, encoding, decoding, which is another top use case that we see on the platform today. It also is a challenging effort because there is a level of capital expenditure required to do GPU right. It's just an area where, you know, we're constantly looking at the ROI and, you know, when we think the timing is right for something like that, but it's definitely, you know, part of our roadmap.
Great. This is another flavor of pricing question, but a little bit different. This is from Josh Baer at Morgan Stanley, and he says, "How will you maintain your pricing advantage as you ramp your sales processes? Historically, lower pricing was an extension of extremely efficient self-service. As the go-to-market shifts, how will the economics work to maintain lower pricing compared to the hyperscalers, offer better and differentiated customer support, and still yield similar margins even at smaller scale?
Do you wanna take that or want me to take it?
Yeah. Why don't you take that one, Yancey?
It's a great question. I think pricing for us. By the way, all the hyperscalers raised prices this year in their cloud service. You know, we're roughly 50%. We've seen when customers come to us from one of the three big hyperscalers, they generally save 50% of their bill. They have an incredibly better experience with the customized support, the documentation works for them, et cetera, and the product set is relevant for them. I think what happens so often with smaller businesses is you go to a big enterprise-focused company with deep tech, broad tech capabilities, but you're so early in the journey, you can't use but a small percentage of it.
Yet it's complex, it's expensive, the terms of service are difficult, and it's just not aligned with the experience you need to focus your efforts on your customers and building products that enable you to build your business. You know, we think even though we're adding a sales capability, I think Jeff outlined it pretty well, we're staying focused on the same. We think this is a $1 billion+ on the SMB opportunity, learners, builders, and scalers. That gets us to a $1 billion+. You know, we talked about some additional products. We haven't talked about a long line of products, a few areas. We keep our costs manageable because we're focused on a unique customer experience with a relevant set of products, and we use a lot of digital engagement.
Because so many of the customers come here through self-serve, even though we're investing in a sales capability, we're still gonna have a lot of leverage. This pricing action that we've taken is really reflecting a lot of the value proposition that we think we offer in creating real differentiation between Linode and Vultr and us. You know, we were essentially priced at parity to them, yet have a much broader set of assets and capabilities that we can offer our customers. We're still maintaining a pretty significant gap with the hyperscalers, yet we offer an experience that allows people to go from zero to people with an idea to $100 million dollar businesses being run on our platform.
We think that's gonna continue in the future, and we think that we can continue to invest across the board and still manage pricing, as a differentiator, just like simplicity, community support, and open source.
Great. Bill, this next one I think is perfect for you. How much visibility do you have into the next 12 months revenue? Is there a base case level of infrastructure activity that you can count on even if growth of customers slows, but assuming that they don't churn/turn off completely?
Well, with 10 years of cohorts and over 600,000 customers, we obviously have a long tail that continues to provide revenue for us going forward. It's paramount for us to make sure that we always have sufficient capacity to provide our current customers, but also to be ready as new customers come to the platform, as our new products launch. We're always looking at what our capacity is. At the same time, our overall CapEx spend is comprised of growth and maintenance as well as expansion. Jeff talked about Australia. We're constantly looking forward over the next 6-8 quarters at what our needs are. We move in those investments depending upon where need is, and if need is continuing to go high, maybe we're slowing up a little on maintenance.
If there's growth a little bit less than we thought, we can increase maintenance at any one point in time. It's critically important for us to continue to manage it in that way, and paramount among that is continuing to make progress at increasing cash flow and reducing CapEx as a percentage of revenue. All of those dynamics work together for us in terms of how we plan. If indeed there is a slowdown, let's say, we'll use the opportunity to use dollars to invest in infrastructure, and we'll be ready to capture growth if it comes up. We do have a very long tail, which is fabulous for us and does give us a lot of visibility in the future.
We look to continue to accelerate it, but we're always ensuring that we have the capacity there to meet the need, and also to try to make sure we're not overextended in any way, in other words, buying too much at any one point in time. We haven't had that problem as of yet.
Thanks. Probably have the time only for one last one, but a number of people asked it, and it's really about our use of capital. You have a lot of capital on the balance or a lot of cash on the balance sheet. How do you think about prioritizing uses of it? Maybe, Yancey, you can take that one.
Well, I think number one,two,three, for prioritization of capital is growth, both in terms of, you know, net new products, you know, building out our go-to-market capability, building out community, building out our team, and the overall strength of the organization. We're investing in the long-term opportunity. That's the priority for our balance sheet, both, you know, organically, through things like the data center. We can work with our vendors. You know, it's a challenging environment out there right now, still challenging from a supply chain. So the balance sheet helps us with our vendors, give them flexibility to keep us at the front of the line. That's a priority to enable to have a smooth growth path. M&A.
We've done, you know, community deals. We've done product deals with the, you know, as you've seen with serverless. You know, I think you could see us doing larger deals. We hope to be announcing, you know, one, two, three M&A deals a year. We've done one with CSS-Tricks this year. We did Nimbella last year. Hopefully, we can accelerate the pace of that, you know, as we get through this year and beyond. That's our priority is to drive 30% or better growth, you know, get to and then, you know, manage 20% or better free cash flow upon arrival at $1 billion of revenue. That's our core financial metrics.
At the same time, we think we can do that and manage the balance sheet so that investors are gonna see, you know, a high grower with free cash flow and a slimmed-down balance sheet that's gonna drive compelling returns. That's the strategy around buyback. It's not our first priority, but it is integral to the thinking of how do we leverage a dollar. You've heard a lot of talk here about how we are disciplined about, you know, our choices and where we invest, and the buyback is the same thing. We look at, you know, how many shares outstanding are required to drive a billion-dollar-plus business growing 30%, generating 20% free cash flow.
You know, the buyback is an element of use of cash, but the first priority and the first few items on the list are certainly to invest in this, what we see as a large, limitless growth opportunity, really.
That's a great note to end on. You know, thank you all for joining us today. I hope you learned a lot about our strategy over the next several years to achieve these goals. We know we're eager to check in with you routinely, as you have questions of, you know, these, leaders of the company. Thank you so much for joining us. If you have questions, from these materials or, from anything we've said today, please don't hesitate to reach out. We're very responsive to investor inquiries, and, we hope, you know, it's a good use of your time. Thank you very much.