2jenn - Fotolia

Snowflake's Slootman predicts future of cloud analytics

Cloud data platforms that effectively eliminate data silos and reduce the need for multiple tools will emerge over the next few years, according to Snowflake's CEO.

Frank Slootman has a vision of the future.

Slootman is the CEO of Snowflake, a cloud data warehouse vendor founded in 2012 that in August broke records with its initial public stock offering.

Priced at $120 per share when it hit the market -- well above an expected price range of between $75 and $100 per share -- Snowflake raised $3.4 billion with its IPO, more than tripling the previous record for tech companies held by VMware, which raised a little under $1 billion in 2007.

Now, less than four months later, the vendor's stock is trading at more than $350 per share.

Big improvements coming to analytics

Speaking on Wednesday with ThoughtSpot co-founder and executive chairman Ajeet Singh during the keynote address of Beyond.2020, ThoughtSpot's virtual user conference, Slootman said he thinks analytics will become far more efficient over the next few years than it is now.

In particular, he said, there will eventually be a data cloud where data silos are eliminated and a single copy of an organization's data will live and be accessed, analyzed and shared. In addition, he predicted that there will be cloud data platforms that execute all the data management and analysis tasks it now takes a multitude of different platforms to accomplish.

With respect to the data cloud, Slootman said that while organizations store much of their data in cloud data warehouses such as Snowflake, Amazon Redshift, Google BigQuery and Microsoft Azure, they also have data stored in a scattered array of SaaS applications and on premises.

ThoughtSpot co-founder and executive chairman Ajeet Singh (left) and Snowflake CEO Frank Slootman discuss the cloud in a fireside chat during ThoughtSpot's virtual user conference.
Snowflake CEO Frank Slootman (right) speaks with ThoughtSpot co-founder and executive chairman Ajeet Singh during ThoughtSpot's virtual user conference.

Data cloud to end data disconnect

As a result, data teams spend a significant amount of time trying to locate and join data from disparate sources rather than building the models that can help their organizations make data-driven decisions. A data cloud where those disparate sources are broken down and eliminated would have a huge impact on efficiency.

"As an industry, we have not yet been capable of creating the notion of a data cloud," Slootman said. "Our goal is to bring about a data universe where we can go and get and plug in and bring about relationships with other data providers.

"We can make data available; we can consume data from other sources, and we can do that without the heavy lifting of copying and replicating and so on," he continued.

Snowflake data cloud

Snowflake introduced its own data cloud in June, and at its virtual user conference in November added capabilities including Snowpark, a new way for users to program data in Snowflake, and a business intelligence platform called Snowsight.

Our goal is to bring about a data universe where we can go and get and plug in and bring about relationships with other data providers.
Frank SlootmanCEO, Snowflake

"A data cloud has not existed in the history of computing, and it's certainly our goal to make a huge dent in that and bring that about," Slootman declared. "We think it will unleash powerful data network effects because people will realize they might have a good workload execution platform, but the reasons they're [in the cloud data warehouse] are because of the data they need to access."

Regarding the development cloud data platforms that do the work it currently takes a multitude of platforms to complete, Slootman pointed out that data consumers now need platforms to extract, transform and load their data, store their data, wrangle their data, curate their data, visualize their data and analyze their data.

Eventually, Slootman predicted, a single cloud platform will be able to do each task that currently requires a platform of its own.

"What that means is I don't need to have 10 different solutions to handle the range of workloads I need to address," he said. "This is really important because that fragmentation that has existed has been the bane of our existence."

Data management and better analytics

He added that much of Snowflake's work has been to improve the data management process.

"Then customers have a fighting chance of bringing all their data management and data operations onto a single platform," Slootman said. "And the word platform is important because it implies that it's broadly multipurpose as opposed to a tool, which tends to be restricted in its purposes. That would be a huge thing because historically we've not been able to get there."

Meanwhile, Slootman pointed out that analytics is perhaps more important now amid the COVID-19 pandemic than ever.

As a result, as healthcare organizations try to stop the spread of the virus and treat those who have been infected, and enterprises attempt to survive deep losses in demand, efficiency is critical. That need for data has led many organizations that previously relied on anecdotal information to drive their decisions with analytics and has subsequently driven analytics vendors to innovate faster to meet the needs of customers.

"It activated our data culture, our data awareness, our data appreciation," Slootman said. "Now data is driving operations instead of just informing people."

Next Steps

ThoughtSpot unveils support for Snowflake developer platform

Speed, ease of use guide Snowflake approach to analytics

Time is now for enterprises to deploy analytics in the cloud

Future of analytics will close gap between data, decisions

Snowflake aims at financial services with data cloud

Dig Deeper on Data science and analytics

Data Management
Content Management