One of the more interesting companies to enter the capital markets space in recent years—in my opinion, anyway—has been Snowflake. For proof, just look at some of its recent wins.

Last year, it was announced that Snowflake would power BlackRock’s Aladdin Data Cloud, State Street’s Alpha Data Platform, FactSet’s outsourced data mapping platform Concordance, Citi Securities’ post-trade revamp, and Northern Trust’s data-sharing initiative for asset servicing. Among others, it counts clients such as the New York Stock Exchange, S&P Global, MSCI, Broadridge, Rimes, Crux Informatics, EY, Deloitte, Xignite and ISS ESG. It has partnerships with Amazon Web Services (AWS), Google Cloud (GCP), and Microsoft Azure.

The financials look pretty good, too. According to Seeking Alpha, for fiscal year 2022, which ended in March, product sales increased 106% year-on-year to over $1.1 billion, and the company expects that number to grow to $1.9 billion for FY2023. It’s also on track to hit $10 billion in annual sales by FY2029. That’s the good news. The bad news is that its stock (NYSE: SNOW) hit $400 per share in November—at the time of this writing, it was hovering around $180. (But optimists say the company’s stock will rebound). 

What’s more interesting to me is how the vendor is looking to proposition itself to the buy side. Most articles that discuss Snowflake will talk about how the company is looking to help asset managers better utilize their data. The thing is, when I talk with sources on the buy side, they don’t have much to say about Snowflake—but they do talk about how they use Google BigQuery and Amazon Redshift, and there’s Cloudera, IBM Db2, and Databricks’ new Lakehouse offering. It’s a competitive space.

That could very well be a biased view—I might just not be talking to the right buy-side sources. So back in February, Nyela Graham and I spoke with Matthew Glickman, who heads Snowflake’s financial services vertical. We wanted to better understand the company’s positioning.

The impression that I get—and Glickman doesn’t quite see it this way, so make of this what you will—is that Snowflake is the behind-the-scenes company powering cloud offerings from major platforms used by asset managers, hedge funds and insurance firms. It doesn’t have a one-on-one, hand-in-glove relationship with these buy-side shops. Glickman says Snowflake does not “force ourselves in the middle of,” Aladdin’s relationship with a buy-side client, for example. But if a hedge fund hypothetically wanted to have their own Snowflake deployment “to do things that overlap or are completely unrelated” to its uses for Aladdin, that manager can benefit because they “can connect the data behind Aladdin to its own private environment” (assuming contractually that’s allowed). Though it doesn’t sound to me like this is the norm.

Instead, the key thing for Snowflake at this time is its partnerships with important data and tech firms that allow it to position itself as a “global data network”.

“Aladdin, State Street, FactSet, S&P—all of our publicly announced and private clients—all connecting to a node in the data network that we call a data cloud,” Glickman says. “Some are running on nodes in AWS, some in Azure, some GCP, and in regions around the world, but it’s literally one big global system that we’ve connected.”

He contends that the company has built the infrastructure, works across the Big 3 cloud providers, and can help an individual organization pull its various data siloes together. The next phase then is to create seamless interoperability between, say, BlackRock’s Aladdin Data Cloud, FactSet’s Concordance and various other data and analytics offerings.

If Snowflake can get these platforms it powers to become interoperable, and create cohesion between important industry platforms, that will be the ultimate differentiator for the company.

“At the end of the day, we have to find a common way for a client to interoperate with FactSet Concordance and a common way for a client to interoperate with Aladdin in the sense that you can basically collaborate and share data and functions on data between them in a common, standard way so that you don’t have to reinvent the wheel,” Glickman says. “Part of the reason why we’re doing these industry cloud launches is to frame [this idea of] what are the pieces that can interoperate together, and how do you do that? … That’s our big push for this year.”

He continues: “We realize we need to meet our customers closer to what they’re solving for, instead of just providing these better pieces that we’ve been doing all along. We’ve assembled the pieces. I think we’ve gotten to a place where we really can do that, and meet them where they are, and be a partner in helping them solve reg reporting, or ESG, or quant trading analysis, or Customer 360 for personalized banking.”

If you read this column enough, you know that I love to bang on about interoperability. More specifically, I often mention three companies in particular: OpenFin, Cosaic and Glue42. With what Glickman was saying, it sounded like there was some overlap with what those vendors are trying to accomplish, even if they’re going about it via the browser, containers, and the open-source FDC3 standard.

Glickman says those vendors are more about interop at the application level, while Snowflake is more about interop at the data services level. When I pushed back that these vendors are working with many of the data services platforms and end-user companies with whom Snowflake has partnered, he said that that is a fair argument, but he believes that Snowflake is solving for interop at a higher, enterprise level (due to the sheer amount of data that it works with), rather than the user application/workflow level.

“I’ve got to basically get my reference data in. Get my fund custodian to send me my latest net asset value and positions, every day. I have to build a pipeline, run risk on that, enrich that with my latest signals, and run my rebalance,” Glickman says. “I need the scale to do that in the data pipeline space, and now I want to train a model that’s going to make predictions on what stocks I should be picking for my clients based on historical patterns. It’s moving that integration from the application user space, to the enterprise data management space. We don’t see them as a competing force; I think it’s just a mind shift—interoperate sooner and lower.”

As it would seem to me, Snowflake believes that it can create a network of data and service providers using its own cloud network that also manages to run across and between the other major cloud providers. The thinking behind that—high-level interoperability so that firms can use real-time data to run analytics, construct portfolios, and measure risk, and then get that internal and external information in one spot—makes complete sense.

But it also sounds a bit audacious to me. In the back of my head (and this might not be reasonable) I can’t help but wonder what happens down the line as Amazon, Google, Microsoft and IBM take greater notice of their opportunity to be disruptive in the capital markets. Cloud is the key to change in our world. Cloud disruption—along with the additions of wider API usage for data delivery, open-source tools that have democratized things like machine learning and natural language processing, and even the burgeoning low-code/no-code movement—is leading to a larger interoperability push throughout the industry.

As we’ve written quite often, the days of monolithic, closed-off systems are dying (sans Bloomberg—and to be fair, even BBG is opening up more and more). Tomorrow it will be all about cloud, SaaS and managed services. It’s all about interoperability.

From end-users to vendors, everyone in the capital markets is realizing. Does Snowflake—a company that has other industry verticals outside of the capital markets—have the staying power to combat major cloud providers if they decide to throw their weight around? Or if the major platform providers like Bloomberg, SS&C, Ion and the exchanges that have been buying up data providers decide to not through their weight behind Snowflake’s vision of a global data network? Or if those pesky regulators start to worry about “systematically important” tech companies?

Or…maybe I’m just being cynical.

Glickman, who worked at Goldman Sachs for more than two decades, is right on several fronts: buy-side firms want easier access to real-time, live data; they want that information to be available in one place where everyone throughout the organization (who has been permissioned) has access to that data; and they want a single source of truth, for data siloes to be broken down, and they want to more easily connect with third-party data analytics tools. Why can’t Snowflake be the one to unify the masses in the name of enterprise-level interoperability?

Have thoughts? Let me know: anthony.malakian@infopro-digital.com.

The image accompanying this column is “The Woman with the Spider Web between Bare Trees” by Caspar David Friedrich, courtesy of the Cleveland Museum of Art’s open-access program.