Data warehouses may not grab the attention of young founders looking to build the next viral app or drone. But at startup Snowflake Computing, CEO Bob Muglia is proving there’s still huge money in the category.
Snowflake announced a new funding round on Thursday that will pump $263 million into San Mateo, Calif.-based company. Led by Sequoia Capital and previous investors ICONIQ Capital and Altimeter Capital, the funding values the company at $1.5 billion.
“Every CEO knows they need to be data-driven in the modern world, but if you go out and talk to them, they’d tell you they have a tremendous distance they could improve,” Muglia says. “Many of them are frankly in the dark ages of working with data.”
Snowflake has now raised nearly half a billion dollars – $473 million – since its founding in 2012. It’s been using the money to double headcount, from 160 in 2017 to about 330 now, and about 600 by the start of 2019. But according to new investor Sequoia, Snowflake is raising so much money in part because it’s simply selling so fast. “The biggest reason is that they are dramatically outperforming their own plan,” says investor Pat Grady, who says that Snowflake already had about 18 months of cash still in the bank but has more flexibility now.
Why Snowflake is finding demand in the market, according to Muglia, is that it’s helping big businesses make their data more readily available for use in the cloud. Companies upload their data to Snowflake’s data centers in hours (a whole system would take one or two days) and then plug that data layer into a business intelligence tool like Tableau to get insights into customers and sales. The customers pay on a subscription or pay-as-you-go basis, with contracts running into the hundreds of thousands of dollars for Snowflake.
At membership-based commerce site Rue La La, the core business still runs on legacy Oracle systems that haven’t changed much in several decades. Activity data like a customer’s shopping history, what marketing emails they engaged with or what they looked at on the company’s website weren’t easily available for use by Rue La La’s data teams in Tableau and other business intelligence tools. So Rue La La reproduced that data in Snowflake and retired a Hadoop-based big data lake it’d been using in recent years. “We didn’t have a 360 degree view of the customer,” says Erick Roesch, director and architect of the business intelligence and data warehouse program at Rue La La.
With Snowflake, Rue La La is able to study all of those touch points with a customer much faster, allowing it to build a recommendation system that filters different pieces of clothing to shoppers in real-time. Each member visiting Rue La La’s website sees a different page of items; every email is personalized, too. “We’ve moved data warehousing from a back office view to really powering our site,” says Roesch. “It’s mission critical and it’s hands free, and our customers are seeing the benefit.”
At Snowflake, Muglia says that he hopes customers with access to such data can change the way they do business to make more decisions more tied to what’s moving their bottom line. It seems to be working for Snowflake itself – the company says its tripled its sales in each of its three years in production, with the average customer quadrupling the amount of data they housed with Snowflake over the past year.
A former longtime Microsoft executive, Muglia says the startup he joined in 2014 won’t be easily copycatted by Amazon, Oracle or his former employer. “I get asked that a lot, but it’s really hard to. There are maybe 20 people who can build a relational database from scratch, and we have four or five of them on staff,” he says. Eventually he wants businesses to use Snowflake to share their own data with business partners in real-time, such as a parts manufacturer working with an airline on a new plane. “Data used to be a precious resource, and you wouldn’t let anyone touch it,” Muglia says. “Snowflake makes that go away – and that creates a massive cultural shift that is a journey just beginning.”
