Skip to content

Search the site

Databricks just raised a fresh $500 million from investors. Why?

"We haven’t disclosed a specific timeline to reaching profitability, and we are continuing to invest aggressively"

Databricks just added a cool half-billion to its war chest – raising $500 million in a funding round that valued the company at $43 billion.

That's more than double what some analysts had viewed its valuation as being, just months earlier; and approximately a 28x revenue multiple.

NVIDIA and Capital One Ventures joined existing investors in the round – which follows a $1.6 billion Series H round in 2021 that valued it at $38 billion.

Such huge raises had arguably seemed like a bit of a “zero interest rate phenomenon” with many technology companies moving from boasting about how big their valuation was to trying to eke out a small profit. (A startling number of software companies continue to be loss-making.)

Databricks has yet to achieve profitability, but with a revenue run rate of $1.5 billion, 10,000 global customers and 50% revenue growth year-on-year, growth is strong and coin landing in coffers. Why tap up investors anew?

Databricks Series I: Collaboration and Acquisitions?

Databricks valuation hits $42 billion in Series I

CEO Ali Ghodsi told The Stack that his company was in “excellent financial standing” but that “collaboration has always been incredibly important… from our origins in academia and the open source community, to working more closely with partners like NVIDIA to build our potential in AI."

Databricks is best known for a unified analytics platform built on Apache Spark for big data processing and machine learning tasks that were traditionally CPU-centric and traditionally supervised. But the world is changing fast at the data and the hardware layer, with GPU acceleration and neural networks requiring a different approach to working with data.

Databricks already supports clusters accelerated with GPUs, e.g. customers can use Databricks Container Services on clusters with GPUs to create portable deep learning environments with customised libraries, but there is more to do on this front, critics say. Ensuring that Databricks is a well-rounded partner for those training and fine-tuning generative AI is requiring some something of a strategic rethink – and, its CEO hinted, potentially more acquisitions in the wake of its recent MosaicML deal.

Ghodsi added in an emailed comment: “We are continuing to invest heavily in our data and AI platform to ensure that generative AI is accessible for every organisation... we plan to further develop our offering to allow customers to continue to build their own generative AI solutions faster. We also plan to evaluate additional data and AI M&A investments.”

(“Enterprise data is a goldmine for generative AI," said NVIDIA CEO Jensen Huang on September 14: “Databricks is doing incredible work with NVIDIA technology to accelerate data processing and generative AI models.”)

What about profitability, meanwhile?

“We haven’t disclosed a specific timeline to reaching profitability, and we are continuing to invest aggressively to go after the data and AI market” Databricks CEO Ghodsi told The Stack in response to this question.

He added: “We’ve improved our margins significantly in the most recent quarter while still driving over 50% year-on-year growth, including our strongest ever quarter for net new revenue.”

Join peers following The Stack on LinkedIn