- DigitalOcean's new CEO opened up to Fierce about its strategy for growth and AI
- It is currently looking to woo customers away from hyperscalers and grow its number of big spenders
- On AI, it is looking beyond infrastructure to make it easy for customers to build applicaitons and agents
DigitalOcean is ramping efforts to woo smaller spenders away from hyperscale clouds to create a profitable army of customers looking to scale their cloud and AI deployments. Its plans, as outlined by CEO Paddy Srinivasan, have set it on a collision course with Amazon Web Services, Microsoft and Google Cloud, which are currently sitting on a hoard of customers spending less than $1,000 per month.
Srinivasan joined DigitalOcean in February, and since that time has reevaluated the company’s overarching strategy. His conclusion? There’s a “tremendous opportunity ahead of occupying the space just below the hyperscalers and servicing the customers who are moving their workloads to the cloud and also want an option that is simpler than the hyperscalers but at the same time as scalable.”
DigitalOcean just posted Q3 2024 results in which revenue climbed 12% year on year to $198.5 million and net income which jumped from $19.2 million to $32.9 million. DigitalOcean now has a $798 million annual run rate business with 638,000 paying customers.
Those customers are broken into three buckets: learners, builders and scalers. These are based on monthly spending levels, with learners the lowest tier, builders in the $50-$500 per month range and scalers anyone spending $500 or more per month.
Srinivasan told Fierce that builders and scalers account for about 88% of DigitalOcean’s revenue today. He added it has 18,000 scalers which spend an average of $25,000 or more per month on its services. That figure is growing almost 19% year on year, he said.
“That’s a number I pay a lot of attention to, to see how fast our big customers are growing on our platform because a lot of the innovation that we are pumping out now is aimed at solving their needs so that we can accelerate some of the migration from hyperscalers,” the CEO told Fierce.
Srinivasan added that it doesn’t plan to ignore its learners or builders, but it is making a concerted effort to release features aimed at building its base of top-tier spenders.
Looking at data about the customer base for hyperscale cloud players, it appears DigitalOcean has a huge land and expand opportunity ahead of it. According to HGInsights, nine out of ten Google Cloud’s nearly 1 million customers spend less than $1,000 per month on its services. For AWS, that spending category accounts for 86% of its nearly 2.4 million strong customer base. Microsoft has the smallest portion of “tiny spenders” at 78%.
Asked whether the hoped-for migration from hyperscale clouds has begun, Srinivasan said yes.
“We’re seeing that as a big trend. We have several examples of customers that have done that for a number of reasons,” he said.
End game
DigitalOcean’s COO Jeff Guy told Fierce 18 months ago that the company was aiming to build itself into a $2 billion annual run rate business. Asked whether this was still its target, Srinivasan said he didn’t want to put a cap on its growth ambitions by stating a specific number. Instead, he said simply it’s aiming to cash in on an opportunity for “tremendous value creation” over the next five years and wants to secure its “fair share” of the market.
The company has already released a slew of new features in the past 90 days targeting scalers, including GPU droplets for those with artificial intelligence (AI) ambitions, virtual private cloud capabilities for those with security concerns and a global load balancer.
For its next move, Srinivasan said it’s looking to move up the stack. And, like everyone else, DigitalOcean has its eye on the AI opportunity.
“Everyone in the market is now focused on the infrastructure and perhaps a little bit on the platform,” including DigitalOcean, he said. “But our long-term strategy is to move up stack to provide building blocks on the platform layer and also enable application development or agentic development to be super simple on the DigitalOcean AI platform.”
“That’s our overall direction over the next two to three years,” Srinivasan continued. “Our AI strategy is to abstract out all of this complex and expensive infrastructure, make it easy, make it affordable, make it predictable for our customers so that they can innovation building the right applications and agents so that they can get business benefits out of it. I feel like all of these GPUs and LLMs are a means to an end, they’re not an end unto themselves.”