Back to How Databricks Went $1M to $7B+ ARR in 10 Years | Ron Gabrisko, CRO, with Ron Gabrisko.
Why Did Databricks Stop Charging Per User?
Ron Gabrisko says the early Databricks deals were about 15,000 to 18,000 dollars a year, which looked small for enterprise sales. The product was adding a lot of value, but usage-based accounts were not using enough to match that value. The team tried to raise the price toward the value. They added a platform fee, then user fees, on top of cloud pricing tied to compute.
The surprise was what the user fees did. Once Databricks charged for users, buyers restricted who could use the product, and usage fell with the restriction. He took the user fees off and made every user free. Usage took off, because everyone could use it and get value from it.
He now treats user-based pricing as a thing of the past. It limits how much value most software can capture. Data, queries, and the number of people and agents asking questions all grow, so the price should follow usage. He told early people at Cursor the same thing: move from a price per user to a price on usage.
How Databricks Went $1M to $7B+ ARR in 10 Years | Ron Gabrisko, CRO ·
From the transcript
There's like 10 answers to that. I want all of them. That was actually a question from, actually one of my portfolio company founders asked me, she's like, I want to know this, so. You know, I think, so one thing that I noticed is like, when we started selling Databricks, we were doing like 15K, 18K deals, right? This is annual? Yeah, yeah.
That's pretty small. People say like, especially if you're doing enterprise sales, that's bad. No, exactly. Could be better. New rule, no deals less than my monthly Uber bill. So, you know, I think, yeah, but when I would talk to the customers, we're adding tons of value. But just a lot of the use cases, because we're usage-based, weren't driving a lot of use.