Sramana Mitra: Can you talk about the business model from that time? What were you charging? What were the deal sizes and so forth?
Jonathan Ellis: When we were first starting the company, we had a potential $80,000 deal. I told Matt, “You know if we can get a few deals like this, we might not have to raise funding.” That was kind of naïve for someone who has not done it before.
Sramana Mitra: Were you able to get the deal?
Jonathan Ellis: We did eventually sign a deal with them but it wasn’t for another year or so. So that was another introduction to enterprise sales where sale cycles can take longer than we think.
Sramana Mitra: How did 2011 evolve? At that point, you did have some funding. How many of these deals were you able to get on board?
Jonathan Ellis: We signed Netflix in the beginning of 2011. We definitely started to get more traction. We had the 0.7 release in October of 2010. In a lot of ways, that was a quantum leap in terms of ease of use from earlier Cassandra releases. It definitely helped a lot in building our traction.
Another major thing we didn’t know when we started was just how much manpower it takes to build an enterprise software company. Matt said we hired way ahead of plan in terms of both engineering and support. But I think we hired mostly according to plan in sales.
The unplanned growth in head count was mostly in engineering. But the cruel fact is that we were saying, “Hey, we’ll raise this $2.7 million that will fund us for a year and a half.” I don’t remember how long we were planning to go, but it wasn’t going to last that long because we’d increased our head count. So we ended up having to raise a Series B earlier than we planned as well, so some of the mistakes of inexperience did have a very real impact on the business side of things.
Sramana Mitra: But then before you went into raising Series B you had a bunch of serious paying customers?
Jonathan Ellis: We had a bunch of serious customers. I think the main thing that hurt us was that we didn’t have a real product. We had OpsCenter which is management and monitoring. The investors, however, want the core infrastructure as your product. We didn’t have that when we were raising Series B funds. The investors were saying, “For Series A, we will write you a check based on hope and dreams. But for Series B, we want to see that you’ve actually built what you said you were going to build and deliver to customers. We see that you have a customer list but they’re mostly paying you through support. Let’s be honest here. You’re still selling a dream that you will be able to build a real enterprise product on top of Cassandra. But you haven’t done that yet.” I think that was the biggest error.
Sramana Mitra: How did you navigate that kind of catch-22 in the Series B raise?
Jonathan Ellis: We did find a partner who was willing to buy the dream. One of the things that helped was that our VC was Crosslink and the incredible amount of footwork they did on our deal. They had people call just about every one of our customers as well as Cassandra users who were not customers. The feedback they got was, “They don’t have a product yet, but their technology is amazing and you should take a chance on them.”
Sramana Mitra: Basically, you had investors who could see the vision and who could see the caliber that you had on the team and believed that you would be able to bring the product to market. It’s just a matter of time.
Jonathan Ellis: Yes. I think fundamentally that’s the summary.
Sramana Mitra: When you go through a series of funding, the Series A is generally a larger round. When we talk about your $2.7 million A round, it does not seem like a lot of money. We prefer entrepreneurs raise a larger seed round than you did, and also a larger Series A round than you did. Here, you got caught in between milestones.
Jonathan Ellis: Yes. If I was going to do it over again, I definitely would have tried to raise a larger series A and gotten that extra runway.
Sramana Mitra: Yes. Without enough validation, people don’t give you a large series A’s either. So you need runway in the seed as well.
Jonathan Ellis: Yes. It’s quite possible that you’re right.
Sramana Mitra: How much did you end up raising in Series B then from CrossLink?
Jonathan Ellis: About $11 million.
Sramana Mitra: Okay. That’s a sizeable amount.