Snowflake Intelligence is an opinionated agentic platform focused on creating value from structured and unstructured data, freeing users from the inflexibility of 2D dashboards.

(00:00:06) Hi, listeners. (00:00:07) Welcome back to No Priors. (00:00:08) Today, I'm here with Sridhar Ramaswamy, the CEO of Snowflake, the former founder of Niva and the SVP of Google Ads. (00:00:16) We will talk about his first 18 months of being CEO, the incredible execution over that time in shifting a company at scale to being AI first, where the enterprise ROI is, (00:00:29) and what happens to the cloud service providers and the ads model in the age of AI. (00:00:34) Welcome, Sridhar. (00:00:35) Sarah, really excited to be back. (00:00:37) Well, it's a pleasure to talk to you as an old friend and colleague. (00:00:41) The last time we spoke, you were on the entrepreneurial journey doing search still. (00:00:46) You're now 18 months into being CEO of Snowflake. (00:00:50) It has been a very eventful 18 months. (00:00:54) Tell us a little bit just about the journey from taking the mantle from Frank to the first few months to where you guys are today. (00:01:02) I think the market's reacted in many ways, most recently incredibly well to the execution, but it's been a journey. (00:01:10) That's right. (00:01:11) That's right. (00:01:11) Snowflake has always been an amazing product company. (00:01:16) The original product that Benoit and Theory conceived of 10 plus years ago, (00:01:22) was many years ahead of its time. (00:01:24) It took the world by storm. (00:01:26) And obviously they had the storied IPO, the biggest software IPO at that time. (00:01:33) I think what happened was the company was a little slow to reacting to changes from things like machine learning and AI. (00:01:42) And that was a little bit of, honestly, the reason why Frank voluntarily pushed for the change, because he felt presciently that we were headed into a time that was just a lot more tumultuous from a product perspective, and he wanted someone that was product first to be in charge of the company. (00:02:00) And the last 18 months have really been about embracing that wave of change. (00:02:05) And if you look back to what's happened in the last two years, it's crazy how much change has happened with respect to AI, how it's become commonplace every day in all of our lives, and then the speed at which things are still getting driven through. (00:02:20) I think the really amazing thing about Snowflake is the company embraced this change, transformed itself, (00:02:29) And then showed that not only can we do it from a product perspective, which one could have expected, but we've also done significant things to retool our marketing, our go-to-market overall. (00:02:40) I think that transformation has been pretty amazing to watch. (00:02:44) But times can be difficult. (00:02:47) Last year, there were a lot of doubters. (00:02:49) But (00:02:50) There were a lot of us who believed both in the value that Snowflake was already creating. (00:02:55) And the reason I took this job was because I talked to a whole lot of customers before I became CEO. (00:02:59) They all loved Snowflake. (00:03:01) And that was a big motivation for me to take this job. (00:03:04) So I think we have sort of successfully ridden through that and are now at the cutting edge of data and AI for enterprises. (00:03:12) It's been an amazing journey to have gone through. (00:03:15) Walk me through just some of the orientation, prioritization you did in the first six months, and then a little bit more about the long-term vision here. (00:03:25) Yeah, the first six months were a lot of tactical changes, which primarily around accountability. (00:03:32) Like every company that goes through (00:03:35) essentially a rocket ship phase of growth, growing at 100 plus percent year on year. (00:03:41) Snowflake had basically specialized at every layer possible. (00:03:46) And there was a very long distance between the engineer that did a feature and the customer that made use of the feature. (00:03:53) And there were like 7 to 10 layers of teams that were involved. (00:03:56) That works fine and you have perfect product market fit and you're trying to optimize for every function. (00:04:01) You're just the winning cloud data warehouse. (00:04:02) Yeah, drive a truck through that. (00:04:04) But on the other hand, if you're working in the world of AI where we can barely tell what's going to come out next month, forget next year, this is the wrong structure to have. (00:04:14) So we did a lot of organizing by different areas, making sure that there were accountable people (00:04:20) For example, in product engineering, that was among the first changes, organized into different product areas like AI or the core warehousing analytics product. (00:04:31) But then we also wanted a straight line over to our go-to-market team. (00:04:34) So we created these specialized teams that work closely with product and engineering and marketing to take these new products to market. (00:04:42) And that was a lot of the early phase of Snowflake with an emphasis towards faster iteration. (00:04:49) This is something that I've believed in all my life, which is speed wins. (00:04:53) Ability to iterate always Trump's carefully laid out strategies. (00:04:58) Yes, you shouldn't do dumb things, but on the other hand, realizing any kind of gain requires a lot of iteration. (00:05:06) So we made a number of changes on that side, both with respect to how quickly we created products, but also how quickly we iterated with customers. (00:05:14) And I would also say we took a little bit of time to find sort of our sweet spot in this AI space. (00:05:20) As you know, that itself has evolved A lot. (00:05:22) We are not a CSP. (00:05:23) We are not a foundation lab. (00:05:25) So what are we? (00:05:26) And there was that discovery of ourselves as the AI data cloud as opposed to the data cloud. (00:05:32) And I think it's that kind of (00:05:36) clear product insight into what value we add that is setting up the stage for the earlier paths of the year or even what we are about to talk about today. (00:05:45) You know, now you're announcing Snowflake Intelligence. (00:05:48) Tell us about that and how it fits into the broader vision. (00:05:50) I mean, first of all, when it came to AI, as I said, we had to look hard at ourselves. (00:05:54) Early last year, we actually went down the path of creating foundation models. (00:05:59) We created a credible MOE model. (00:06:03) This was early last year. (00:06:04) But we also quickly realized that our ability to compete with the likes of OpenAI or Anthropic was going to be really hard. (00:06:13) We simply did not have the capital to be able to invest meaningfully in things like that. (00:06:20) So we pivoted away from that into much more of a, how does AI massively accelerate what can be done with data that is in Snowflake? (00:06:33) And over time, that can become a reason to bring more data into Snowflake, which is the phase that we are in now. (00:06:39) But a lot of our AI product strategy was actually quite humble. (00:06:44) It didn't say we were going to rethink everything. (00:06:47) It said enormous number of customers. (00:06:50) Something like half of the qualifying Fortune 2000 companies on the planet are Snowflake customers. (00:06:56) They have their most valuable data on Snowflake. (00:06:59) What does AI mean for that? (00:07:00) And so we systematically invested in the components, whether it was search or whether it was text to SQL, in ways that added on value to the things that people were already doing with Snowflake. (00:07:14) And SI is an agentic platform, but it's actually an opinionated agentic platform. (00:07:20) A lot of agentic platforms, for example, from the CSPs, will basically say, oh, you can bring in data from anywhere. (00:07:26) You can imagine any kind of workflow that you want to do, and the one agent will rule them all, which is nice in theory, but in practice, when you have an infinity of things that you can do, it's also hard to figure out what you should actually do. (00:07:38) Snowflake intelligence is very focused on how do you create value from data, whether it's structured or unstructured, a whole lot faster. (00:07:46) And so the kind of use cases that got us really excited, honestly, internal ones were things like, if we were to take all of the different dashboards that we used in sales and put it into one single interface, what could that be? (00:07:59) We had done two, three versions of this, but eventually that culminated in this internal product. (00:08:04) We call it Raven, but it's basically the sales data assistant. (00:08:08) Then we started working with early customers, whether it is a Cisco or a Fanatics or the USA Bobsled team, (00:08:15) to figure out what does this all mean for them. (00:08:17) And the theme again is get away from the inflexibility of things like dashboards. (00:08:23) A dashboard is a 2D view of a complex surface. (00:08:25) It just has no easy answers to the many questions that any reasonable person, you or I, is going to have off of that. (00:08:32) So we wanted to create something that freed people of the 2D style of thinking, was much more flexible in what it gave people access to, but also knew its place. (00:08:41) This is not a general purpose agent tech platform to do (00:08:46) This is an agentic platform that lets people realize value from data faster and is a great foundation for people to get value from data like really quick in a meaningful way. (00:08:56) I think having this sort of an opinionated framework for AI has been super helpful for us. (00:09:01) How does a user consume Snowflake Intelligence? (00:09:04) Is it like, I ask a question, I get push and answer, it builds dashboards for me on, like, how should I imagine that experience? (00:09:11) Yeah, so we should show a demo Snowflake Intelligence to you. (00:09:15) But yes, it's an interactive interface. (00:09:18) You can ask questions. (00:09:19) There are a set of canned questions to make sure that you don't have block when it comes to being able to ask questions. (00:09:26) You can ask it to say, hey, what data sets do you have access to? (00:09:29) What kind of questions can you answer? (00:09:31) It'll do a perfectly reasonable job of that. (00:09:34) And our aspiration was for this product to be used by every single employee in the company. (00:09:41) So it's not for people who can write SQL. (00:09:43) It's not for people that can write SQL. (00:09:45) We wanted it to be enough of a daily use product for every single person. (00:09:50) There is not a single customer meeting that I'm going to have without quickly checking up on what's the latest with this customer. (00:09:57) And so Raven that I talked about, our sales data assistant, absolutely has things like, what's our relationship with the customer? (00:10:04) What kind of contract have they signed? (00:10:06) What is their consumption looking like? (00:10:08) But also things like, what are the most recent conversations that we have had with them? (00:10:11) What came out of these? (00:10:13) Are there any outstanding tickling issues? (00:10:16) And so it is a lot of that. (00:10:18) And like many good products, there's breadth, there's value driven to many, many people within a company. (00:10:27) On the other hand, we don't pretend it's a BI dashboard. (00:10:29) There are more things that you can do with a Tableau or a Sigma than you can do with SI, but that's not the goal. (00:10:34) Because this product also lets you do a bunch of things that you could not easily do in a dashboard, and is really meant for any business user. (00:10:42) And we place a lot of trust, and we place a lot of emphasis in all our AI products on trust. (00:10:48) Meaning, I tell people, we need to think of AI the same way we think about software engineering, which is there's a right and there's a wrong. (00:10:54) It cannot be this mode of like YOLO AI, you can get some good answers, some terrible answers, it's your problem. (00:11:01) And so we very much emphasize you need an eval for every single new thing that you're going to launch. (00:11:06) If you want to change the underlying model, you need to be able to quickly verify that you didn't blow up on the things that you are already doing. (00:11:12) And so we want it to be the trustworthy product for every employee, which is actually a new thing for us, by the way, because Snowflake, for pretty much all of its history, has always been used by the data team to slap a dashboard on top, which then gets exposed to end users (00:11:28) This is a very different motion. (00:11:29) This is why we are working on things like identity provider integration, so that you don't have to set up Snowflake accounts for each of the many users you're going to have in your company. (00:11:38) We are also by the fact that there is subscription fatigue. (00:11:41) And so Snowflake Intelligence is very much a consumption product. (00:11:45) People pay for what they consume. (00:11:46) And we are experimenting with a bunch of things there in terms of how do we drive broad and deep adoption without having people worry about runaway costs and things like that. (00:11:55) The way you describe Raven or the sales assistant agent use case, it sounds (00:12:00) like an application or a lot of applications that I get pitched, like how do you draw the line between like data and agent system and app today? (00:12:09) Back to my point about execution, I tend to be like completely emotionless about where the strongest current is. (00:12:19) On the other hand, it's absurd if we think we are SAP or Salesforce. (00:12:23) We are not. (00:12:24) There are somebody managing $100 billion supply chain ecosystem (00:12:30) with a complicated software provider isn't saying, hey, I'm going to use SI and I don't need that. (00:12:35) That's really not the goal. (00:12:37) But on the other hand, I think the line between what an agentic system like this is going to be and what like pure software is going to be will be, absolutely is going to be bloody. (00:12:49) And I can imagine a lot of easy use cases, like my sales team has to go update Salesforce quite often, you know, because we force them to update whenever there's use case transition and stuff like that. (00:13:00) Can that be done with APIs? (00:13:01) Absolutely. (00:13:02) Should you be able to file a vacation on top of Workday using our HR agent? (00:13:07) I would say that's sort of a reasonable thing. (00:13:10) And so we very much take this approach of be opportunistic, but again, operate from a position of value and strength and not just on naked ambition, because I think that doesn't work out. (00:13:21) But if on the other hand, you focus on how do you, like what does value creation mean, what do these users really want, I think that tends to be much more durable. (00:13:30) So you're describing a bunch of changes for the organization that you executed on very rapidly, right? (00:13:36) Both in... (00:13:38) Always feels entirely too slow. (00:13:40) Yes, I have always experienced you to be quite impatient, but you know, for scale seems pretty fast. (00:13:45) How do you, what is 1 tactical thing you were doing, are doing from a leadership perspective in terms of like move faster or communicate new direction internally and get people on board? (00:13:57) Because you, know, this is a (00:13:59) This is a broader and different vision for Snowflake than before. (00:14:03) Change is hard. (00:14:04) You have to acknowledge that. (00:14:05) And driving behavioral changes from lots of people is incredibly difficult. (00:14:12) We were measured about how we rolled out changes. (00:14:16) For example, (00:14:17) Among the first changes were leadership and alignment changes and clearer accountability. (00:14:23) That happened within a few quarters because you're not dealing with as many people. (00:14:27) Yes, you organize the teams under them. (00:14:29) But I would say that change also what we then called the water room or the pod model of product and engineering and our go-to-market functions will all work together. (00:14:42) That was, again, an early change. (00:14:44) And it was done with small groups of people without necessarily (00:14:48) literally disrupting lots of people. (00:14:50) I would say other things, for example, rolling out coding agents to our engineers, that was a project. (00:14:58) Not everybody wants to do it. (00:14:59) Some people are skeptical, some people are not. (00:15:02) I'm a big fan of combining bottoms-up with tops-down approaches. (00:15:08) The example with coding agents is that Benoit, our wonderful founder who fell in love with coding agents, (00:15:14) I did more to drive coding agent adoption with engineers and any number of words from me. (00:15:20) You sort of have to find the right people, find the champions. (00:15:24) My take is that every large organization has these forward-thinking, curious, I'm going to work over the weekends to figure out how to do something kind of people. (00:15:33) You need to find them, you need to encourage them, you need to elevate them and use that to drive change. (00:15:38) Top-down change can be helpful, but it really needs to come from a bottom-up perspective. (00:15:44) out coding agents to all of our solution engineers. (00:15:49) And they are excited because that just dramatically lowered the amount of time it takes to create a demo. (00:15:55) Usually our demos used to be canned and they were not always customizable to a particular customer. (00:16:02) But we can now be like, okay, we know the kind of data we think Elad and Sarah are going to have as part of their podcast. (00:16:10) Let's create a demo with, you know, synthetic data sets just for that. (00:16:14) I think that's the kind of ability that we have gotten. (00:16:17) Change is hard. (00:16:19) When you and I first met and got to work together, you were an investor, then you were an entrepreneur. (00:16:24) Did either one of those roles change the way you are a leader at scale or a CEO? (00:16:29) I think these things are accretive. (00:16:31) They add on ways. (00:16:34) tied on to things in ways that you don't always appreciate or like then or ever. (00:16:40) It is what it is. (00:16:40) I always complain to my family about the 10 years that I spent doing research and getting a PhD. (00:16:48) I was like, that was a waste of time, but not really. (00:16:52) For example, doing a PhD teaches you to focus on ideas, teaches you to focus on how do you convey them crisply. (00:16:58) You often spend enormous amounts of time writing four-line abstracts. (00:17:03) But it actually turns out that's incredibly powerful to be able to convey ideas in an easy way. (00:17:09) New was among the hardest and most heartbreaking experiences of my life. (00:17:14) It is what it is. (00:17:15) Sometimes you are too early. (00:17:17) But on the other hand, I probably learned more about hustling, took success far less for granted, learned more about social or marketing or any of these other things. (00:17:30) But you kind of take for granted if you're at Google. (00:17:32) At Google, whatever you did, my first launch at Google, which was entirely my work for three months, one person, was covered by the New York Times. (00:17:41) Okay, yeah. (00:17:42) So you just have immediate scale with anything you do, your distribution. (00:17:46) Yeah, distribution. (00:17:47) Your ideas, your products. (00:17:50) Doing a startup makes you realize that that's actually special. (00:17:54) And so I think I bring quite a lot of that when it comes to what does it take to hustle? (00:17:59) What does it take to win? (00:18:00) Honestly, I think both the Google and the Niva experiences make me somebody that's just a lot more grateful for my job. (00:18:08) We talked earlier about how you have to deal with a bunch of stuff that you don't really want to deal with when it comes to doing something big that you like. (00:18:16) I'm a lot more gracious about that. (00:18:19) because it is just such a privilege to be at a place like Snowflake, to be having the kind of impact that we have. (00:18:27) I remember what you told me after Summit, you said something along the lines of, thank you for inviting me to your rock concert. (00:18:36) There was a line that was more than two blocks long of people waiting to get into Javits Center in New York for this little conference that we were doing. (00:18:45) I'm a lot more grateful for things like that. (00:18:49) There's nothing ardent about it. (00:18:50) You, I think, are perhaps the world's expert on game theory and strategy with the tech elephants, because you have led the elephant, fought the elephant, and now are built on top of the elephant, right? (00:19:04) And so I think this is the analogy broke down at some point. (00:19:08) But in terms of (00:19:10) The experience of building Snowflake on the cloud service providers and both for you today and then the analogy for anybody building on foundation models as you are as well, how do you think about that? (00:19:25) What's a framework for creating durable value there? (00:19:28) I think product market fit continues to be magical. (00:19:33) It's the reason that Snowflake exists. (00:19:35) Think about it. (00:19:36) The 3 hyperscalers would love to just own the data space like they own any other space. (00:19:43) But yet, there's no flake. (00:19:46) There's Databricks. (00:19:47) And so that sort of redeeming value is quite unique. (00:19:53) And we should all have a lot of humility about what it takes to create that lightning in a bottle. (00:20:01) Having said that, (00:20:03) I think the model companies, especially OpenAI, is super interesting because they are in that phase of their growth where they literally, like, they don't think they can not do anything. (00:20:18) Yes, yeah. (00:20:19) I joke to people that these are like empires that have not met their oceans just yet. (00:20:24) And so I think you do have to pay attention to what is likely to be in their immediate path. (00:20:31) So for example, I think (00:20:33) coding agents are particularly interesting from this perspective, because it is very clear that both Anthropic and OpenAI are going to be laser set on having the best one that there is. (00:20:45) So I think thinking about what is the likely trajectory of these companies, and do they really have a right to win? (00:20:54) Or is it something that is different enough (00:20:57) that you don't really have to worry about. (00:20:59) Google, for example, stopped at information. (00:21:01) God knows I spent enough time trying to get into physical things like shopping or airline purchases or hotels. (00:21:07) We didn't really succeed because we didn't have core competence really in some fundamental way beyond the world of information. (00:21:16) I think it's going to be fascinating to discover what that kind of a boundary is for an OpenAI or an Anthropic. (00:21:22) But I think there are lots of areas that can be reasonably guessed at with respect to where they're going to go. (00:21:29) I think that's the one that's tough. (00:21:31) And early patterns of if you are, for example, a set of prompts on top of one of these models, that's a problematic space to be in. (00:21:40) You kind of need to add value. (00:21:42) I also think a lot about what differentiates (00:21:45) us from these model companies? (00:21:46) Is this an area that they're likely to be interested in? (00:21:49) How do we make sure that we have distance with respect to what we add? (00:21:53) And a lot of the urgency and change in the products we create in collaboration with these folks, it all take us towards the data platform as a durable category. (00:22:03) But this is also a time where literally no software company can feel secure about their position in the sun. (00:22:10) And I actually think that (00:22:12) And that perhaps is just as important as anything else that I just said. (00:22:16) Do you have that orientation of like, we need to continue to earn it? (00:22:21) We need to continue to earn it. (00:22:22) And if there is anything that all of us have learned, say from the CSPs, it is that they have infinite budgets, they have infinite patience. (00:22:32) And unless you innovate and stay ahead, not just be ahead, but stay ahead, you will be intel. (00:22:40) I think that's another really useful lesson to remember as a company like Snowflake navigates the current realm. (00:22:46) And this resonates hugely with me, both on the dimension of like, I'm thinking about one of the founder CEOs of one of my favorite companies that's now a public company that I thought was unassailable. (00:22:56) And AI, I hate to be the person who'd be like, well, this changes everything, but they feel threatened for the first time in many years. (00:23:03) I think that's a pretty common experience right now as a software CEO. (00:23:06) I think especially when the technical environment (00:23:10) is so fluid, defensibility is built, not strategized. (00:23:14) That's correct. (00:23:15) It's built every single day. (00:23:16) You have to keep moving. (00:23:18) One of the... (00:23:19) questions I would have on the data cloud is like you, even if you don't have the CSP's budget, you do have the ability to make multi-year plans. (00:23:30) And you joined Snowflake because you saw like a vision for it to be much more than the data cloud. (00:23:35) As A technologist, like when you look out three to five years, like how do you expect people to think of Snowflake both, you know, in the ecosystem and then customers to use it most? (00:23:46) Our core strength comes in that data platform layer. (00:23:50) I sometimes internally talk about being there for our customers from inception to insight. (00:23:56) From when data is first consumed.