Databricks’ success is hard to overstate. They had the largest raise of 2024, $10 billion at a $62 billion valuation, making them the fourth most highly-valued US-based startup. They’ve achieved 60% YoY growth in recent quarters and crossed a $3 billion revenue run-rate in January 2025. So, on the GTM side, how did they get there?
At Scale’s recent GTM Summit, Databricks CMO Rick Schultz shared the secrets behind this trajectory. Here’s a peek into the pivotal decisions they made to drive their success—and let them create and own their category.
Founder involvement as a competitive advantage
Databricks has seven founders, with five still involved in day-to-day operations. Crucially, they’ve remained deeply engaged in marketing, a practice many companies lose as they scale. It’s not because they want to be, or because Schultz wants them to be, he says. But, there are two critical decisions founders need to be involved in, especially in the early stages.
First is on the people side. The founders were intensely involved in selecting marketing leadership. Schultz spent a lot of time with the team prior to getting the job offer. His first interview with co-founder Ali was a four hour dinner, diving deep into thinking and approach.
Second, they were heavily involved in positioning and messaging.
“What’s more important than how you position your company when you’re trying to be heard of, stand out, create brand awareness, and drive pipeline, preference, and interest in your company?” Schultz says.
A big piece of this is the founders understanding how integrated marketing is to all other functions. Schultz points out, for example, that Ali Ghodsi, Databricks’ CEO, rejected the term product marketing because it suggested building a product and then “throwing it over the wall” to marketing. Instead, Ghodsi viewed product marketing, positioning, and product strategy as unified—understanding the market, building solutions that uniquely win that market, and crafting cohesive messaging around those solutions.
Audience focus that drives results
Databricks made an early strategic choice that many companies struggle with: extreme audience specificity. In the early days, it was critical to be targeted. A line of business is not one audience but a bucket of audiences, and that’s too many for an early startup. Instead, they developed a simple four-box matrix: data science teams and data engineering teams, addressing both practitioners (data scientists and engineers) and their direct management (directors, VPs of data engineering, chief data scientists). Interestingly, this targeting strategy also aligned with their product vision of breaking down traditional silos between these teams.
This strategy worked because they had specific use cases, and the individuals making adoption decisions around those use cases were not in the C-suite. Databricks didn’t start targeting Chief Data Officers or CIOs until several years in, after reaching hundreds of millions in revenue.
The messaging marathon
Databricks invested Schultz’s entire first quarter solely on messaging. This was foundational work that they revisited if the founders felt like they weren’t nailing how they described who Databricks is and what they do.
Their process combined hiring an outside agency to run a rigorous process with deep internal involvement, particularly from founders. They tested messaging with existing customers who understood their value, then moved to prospects.
Schultz described a critical moment when flying to New York with Ghodsi for focus groups. When prospects didn’t immediately understand their message, Ghodsi nearly “jumped through the two-way glass” in frustration. Yet this real-time feedback proved invaluable in refining their approach.
Once the messaging was settled, Databricks maintained what Schultz calls “message stamina”—extraordinary discipline in consistency over time. Too many companies constantly tweak messaging, never giving it enough time to penetrate. In reality, prospects need to hear messages 7-9 times before internalizing them, which means hammering on it quarter after quarter.
Creating a category with the Data Lakehouse
Their creation of the Data Lakehouse category transformed Databricks’s market position. Schultz insists not every company needs to create a category. For them, however, it worked in a big way.
Rick said they knew that they were creating a category years before actually introducing the term. “You have to condition the market around the pain,” he said. Once everyone agrees that they have this pain and that the other solutions don’t solve it, the key is convincing the market the other way is better, which they’d done with their Unified Data Analytics — a much clunkier term than Data Lakehouse.
When they unveiled Data Lakehouse (a portmanteau of data lake and data warehouse), it immediately resonated. Customers got it right away, and the market followed. What accelerated adoption wasn’t just the terminology but their provocative positioning.
That’s not to say that everyone was on board right away. When Gartner analysts and even Bill Inmon, the “father of data warehousing,” initially dismissed the concept, they engaged critics directly, explaining their vision. Two years later, Inmon authored the definitive guidebook on Data Lakehouses. They were able to do this because they stuck with it. They set a solid foundation, with consistent messaging to prime the market for the new category. Then, despite some initial backlash, they didn’t leave the message, the name, or the category, instead educating their customers and the market on how to think and talk about it.
The tipping point came when major customers like Capital One began publicly discussing their Data Lakehouse strategies. Initially hesitant to use terminology that might sound unfamiliar or upset existing vendors, these endorsements created momentum that made the Data Lakehouse the industry standard.
Shared outcomes between sales and marketing
Schultz underscores the importance of the CMO CRO partnership. When he joined, Databricks’s CRO, Ron Gabrisko didn’t waste energy on pipeline attribution battles, recognizing that source attribution is inherently imperfect.
Instead, they focused on shared accountability on agreed upon numbers: determining overall business targets, working backward to identify needed pipeline, and holding both teams jointly responsible. They all failed or succeeded together.
For demand generation, they stayed ruthlessly focused on effective channels. While allocating some budget for awareness through PR, the vast majority went to direct demand generation. They were highly selective about conferences, typically only attending major events that were essential for their audience.
Instead, they invested heavily in reaching their customers where they are via digital marketing. Work happens online, so that’s where they needed to be.
Adapting to the AI revolution
When ChatGPT changed the game, Databricks faced an interesting challenge. While they’d been talking about AI for five or six years in the classic sense (machine learning and numerical applications rather than generative content), they weren’t automatically associated with this new wave of AI innovation.
Their response included creating a second category—the Data Intelligence Platform—positioned as an evolution of the Data Lakehouse. They also made strategic acquisitions, including Mosaic ML, renamed as Mosaic AI, preserving its distinct identity within their portfolio.
