On this episode of After Earnings, Snowflake, the cloud-based data platform that enables organizations to store, process, and analyze massive amounts of data. I'm joined by Sridhar Ramaswamy, Snowflake's CEO, to discuss Project Snow Work, the company's newest AI platform that brings its products directly into the hands of users for the first time. We break down the threat AI may pose to all software companies and his belief that the speed of innovation is a company's best protection. And we dive deep into Snowflake's focus on GAAP profitability as well as its approach to M&A and the competitive acquisition and compensation of top talent. Let's get into it.
So, I'm delighted to welcome Sridhar Ramaswamy, CEO of Snowflake, joining us from Menlo Park. Thank you, Sridhar, for taking the time to join us today.
Excited for this conversation and thank you for having me.
So, if you don't mind, help us understand your products with some depth, Sridhar. And the reason I ask this is we're in a moment in time where the markets are saturated with news about software and technology. Our listeners are saturated. And what we're really trying to do is sort of cut through the noise and understand with some specificity the products of those who welcome to the show and how they're actually defensible and sustainable over the long term. So, in your own words, tell us Snowflake's core product line and why it is standing to win.
Absolutely. Snowflake is an enterprise data platform. We, in fact, call ourselves the AI data cloud. And our mission, actually, is to help enterprises realize their full value from data and AI. In practice, this means that enterprises use us for everything from storing transactional data to storing analytic data. Historical data that helps people understand how performance happen. And often people will do things like build prediction models, build machine learning models on top of this data. And increasingly, we see AI as an enabler for faster access to this data directly by business users as opposed to going through analysts or going through BI dashboards and such. But we are the end-to-end data platform for enterprises. And over 13,000 enterprises from around the world trust us with their most valuable data.
And so, talk to us about who's adopting Snowflake's offering, Sridhar. And let me give you the context of the question. I work with a lot of companies, manufacturing businesses, industrial companies, that have a lot of data, but it's not necessarily, at least right now, in the most usable format. There's a lot of cleanup that's needed. How much is Snowflake supporting enterprises in the cleanup and making useful part of the data agglomeration part of it.
That's right. That is, in fact, our specialty. And we work with companies in many different verticals. A lot of financial institutions for whom data is incredibly valuable, but also manufacturing companies that uses for everything from supply chain analysis or price optimization or manufacturing optimization, to folks like Disney. It's a theme park and they use this for visitor experience optimization. And so, our users span the gamut and it is all of the things that you said, which is how do you bring data in, how do you clean it up, how do you deduplicate it, remove spam, and organize it in such a way that analytics becomes easy, prediction becomes easy. And that's been our specialty, which is the most valuable data belonging to all of these companies is in Snowflake. And we've also helped people collaborate. In fact, our motto is easy, connected, trusted. And the connected part means that it's very easy for one department in a company to share data with another department, but in a governed way, so that only the right people have access to the data. Plus, a ton of our customers get data from their partners, get data from data providers. We provide much of the data sharing that happens in the financial sector in the US, for example. And so, it's really our mission is centered around clean data, great data that you can run analysis on, that you can share responsibly without making copies, so that you have the provenance, you have the right data for the right person at the right time.
And so, when I think about your head count, walk us through what sort of proportion are engineers who are actually the designing the software. What proportion are sales talent going out and selling a pretty complicated story? I mean, it's easy to intuit, but in terms of the execution, I'm sure it's complex. And then, what proportion of your head count is sort of account management in a very hands-on way that's working with enterprises that are perhaps not as sophisticated who need a lot of help with their data cleanup.
This is a great question. Roughly the split, we have about 9,000 people and roughly our split is 2,500 folks in product and engineering. This is the engineers, this is the product managers that help us design our product. Little less than 5,000 people in our broader go-to-market function. We also have a small, but very powerful industry team. We have a services team that helps our most important customers with their implementations. So, we are a full-service firm, even though, obviously, we work with a lot of system integrators, the GSIs, others, to help implement Snowflake.
So, there's a new and exciting development at Snowflake, Sridhar, and that is Project Snow Work. Talk to us about why this is an exciting new chapter.
As I was saying earlier, companies store their most valuable data in Snowflake. Typically, they've been accessed through relatively inflexible means like dashboards. But if you want to ask for something new or you want to take action on it, you'd have to do it differently from the dashboard itself. We rolled out Snowflake Intelligence a few months ago and it was our agentic platform for accessing data. It means, among other things, that all of our sales data, for example, is right there on my phone. I can ask about any customer. Snow Work represents the next big step in that and it's born of our own experiences using AI. Our thesis is that AI is going to be incredibly powerful if you take governed data that's in a platform like Snowflake, combine it with a premium model, and attach other SaaS products as tools into the same platform. This means, for example, that how many launches we have coming up next week or what launches we have next coming up next week, that's not a dashboard that I go to. That is a simple question that I ask. And if my CFO wants to know how our travel and entertainment expenses trending, that's again a simple question to ask off of the Snow Work platform. Not only does he get the first answer, but he can also do things like dig into it. Are there anomalies on or is there an excess of spending in one group versus the other? And he can also take actions on it, share the report with other folks, send email to individuals or to teams. Basically, work becomes a set of capabilities that are provided by this platform. And we're uniquely positioned to do it because we've been this enterprise source of truth and all of the data that's in Snowflake is secure, it's auditable, and it's also been governed since the beginning of time when we created Snowflake. All of this makes this a great product offering for enterprises that want to transform how they operate. Many teams, like my services team, has switched over to using these tools. My support team has already switched over to this mode of operation of always being inside this AI agent with access to the right data and tools. It's the future of how work is going to get done.
So, talk to us a bit about the partnerships that have helped to get you here. You have partnerships with Anthropic, with Google Cloud, with OpenAI. Frankly, the kinds of players that the market is nervous will look to supplant its software colleagues, as it were. So, we've got partnerships today, the market's worried about displacement in the future. Where are you in the partnership point in that cycle?
So, we have excellent partnerships with all the model makers. I think, having the best models power your AI is not an option. They make a material difference to how our products like Snow Work or Snowflake Intelligence operate. And we see ourselves as this amazing platform in which AI is going to be more effective. That's the reason why we launched Snow Work because we want our customers to take advantage of the data and these models. But multiple people are going to have similar ideas. There are products from several of these companies and the hyperscalers, in fact, that all attack the same problem because there is a dawning realization that it is the combination of great data, great models, and tools that are at the heart of how enterprises operate.
And now, a word from KeyBank. Successful businesses don't spin their wheels, they evolve. And in order to keep up, you need a bank relationship that evolves with you. Every business is inherently different and that's why with KeyBank, you'll never receive one-size-fits-all solutions. Their expansive team of experts across industry verticals develop bespoke solutions to your ever-changing challenges and opportunities. Reimagine business as usual, KeyBank opens doors. Learn more at key.com/b2b. That's key.com/b2b.
Why don't you address directly, Sridhar, the concern that investors has? We've seen the software sell-off, and again, it's this concern that partnerships will be supplanted by replacement. What exactly is it that protects you from your partners, A, expropriating the data that's been shared with them over time, and B, developing the specific capabilities that you have, and therefore, replacing Snowflake?
Well, first of all, I think software is evolving at such a rapid rate, and the cost of creating software is going down so quickly that no one should feel safe. That actually includes the model companies. They have to worry that there is a new model, open source perhaps, that is just as good as the models that they're making. I think this is a time of incredible disruption. So, the first thing that I'll say is I'm always paranoid for our success, and paranoid for our future. And while we partner with these companies, we don't give them the data that is in Snowflake. That data belongs to our customers. We're stewards of that data, and because we work closely with our customers, because we have additional enterprise context that a model sitting outside doesn't necessarily have, we have this ability to create great products on top of this data. We're the ones with the knowledge about governance. We're the ones with the knowledge about access control, and what constitutes trust in these environments. And as I was saying earlier, we also have relationships with the 13,000 customers. It is the combination of all of these things, plus the ability to create cutting-edge products that deliver value, that is the protector of success. Used to be software was hard to write, and previously, companies could just say, 'No one can write software that is as good as something.' I think that moat is eroding. It's the relationships, it's the insights. Part of what we also do is make our AI products better as a result of our customers use them. And only we have access to things like those improvements that are going to make Snowflake better on customer data. I would in fact go one step further and say that no software company is safe just by the virtue of them being able to write software. It's by the virtue of the value that they create. The unique secret sauce. For example, if you create a marketplace, a new LLM is just not going to recreate that marketplace because there's a lot of humans coming together, or you need some kind of feedback loop like the one that Google Search had where people using search produced information that made search better. That's the mentality that we take. It's people using AI and data on Snowflake that makes AI and Snowflake the platform better for the future, and that's the only protection that there is today.
And just to push on this point again, Sridhar, which is you're very clear, and I think it's understood that Snowflake does not share the data that it is the custodian of with the LLMs. But there is a fear that those LLMs will get sufficient exposure to the data as it's being used that through inference or other means, it's able to basically replicate certainly some portion of the data that Snowflake does have custody of. So, is that just a completely false preconception? Is there absolutely no way that these LLMs could have backward engineer the data that's fueling it in these partnerships?
I don't think they can reverse engineer the data, but on the other hand, it's a free world. They can go seek relationships with our customers. We cannot wall off data that is in Snowflake. Customers can choose to take the data out if they have better products, if they deliver better value for our customers, and that's why innovation, whether it is Snowflake Intelligence, which is our agentic data platform, or Snow Work, which is our conceptualization of how work itself is going to get done, this innovation matters more than anything else because all of these model companies and the cloud service providers can always directly go to the customers and buy their business perfectly. I mean, you know, and get their business, and that's why the innovation and speed of innovation really matters.
And you've been unafraid, Sridhar, to turn to M&A to support the speed with which you seek out innovation. Tell us about that. February was a very busy month. You acquired TensorStacks, you acquired Observe. Talk about these acquisitions and where else you're hunting for targets.
This is a time that is ripe with opportunity. TensorStacks eventually became Cortex Squared, which is our agentic coding platform. That's the basis of a lot of the work that we're doing in AI. It was a small talent acquisition. Observe is a bigger acquisition, and it was born out of the belief that something like observability was going to be a key component of a data platform. It is an incredible product, and at higher data volumes, it can be an order of magnitude cheaper while being just as effective, and it's built right on Snowflake, so the integration work to integrate Observe and Snowflake was not a whole lot. This is the reason why we acquired Observe. We will continue to be pretty active about M&A because there's a ton of opportunity, and it's honestly tough for small companies to make a big mark right now.
