Latest / Fit Happens: Why CEOs and Leaders Succeed or Fail / Why 85 Percent of New CEOs Are First Timers, with Anand Eswaran

Transcript

Why 85 Percent of New CEOs Are First Timers, with Anand Eswaran · Fit Happens: Why CEOs and Leaders Succeed or Fail

• Jason Baumgarten 00:00 So there's a kid in Mumbai in the 80s taking the family radio apart, then the fans. And soon nothing in the apartment is safe. At 15, he gets his first computer, 40 megabytes of hard drive, which he still chuckles about. And then taking things apart becomes a bit of an obsession. Forty years later, that same kid runs a 15 billion dollar enterprise whose entire product is the exact opposite of taking things apart. Jason Baumgarten 00:26 Veeam doesn't take things apart. It puts them back together after ransomware attacks, outages, some of the worst days a company has ever seen, I guess. Today is Anand Eswaran, the CEO of Veeam. And here's what interests me so much about Anand. He has said out loud that his whole career set him up for this job. Jason Baumgarten 00:47 And that's a big claim. Most of us can only draw that line backwards at the end with the benefit of hindsight. And today I want to talk about that. Anand, welcome to Fit Happens. Anand Eswaran 01:01 Jason, pleasure is mine. Jason Baumgarten 01:01 And boy, have you done your research. Well, I will say I started reading about you taking things apart and we had a chain in the US in the 80s, Radio Shack. And my local Radio Shack had what they lovingly referred to as a junk table. And they would put things that didn't work anymore on the junk table and you could buy them for a dollar, 50 cents. It was very cheap. Jason Baumgarten 01:26 So I would come home every single week with a huge bundle of junk to take apart, to try and fix. And my friends at the time would laugh that every piece of electronics I ever had was something I figured out how to fix. So that resonated with me. Anand Eswaran 01:48 That's cool. What I would say is we talk about curiosity, we talk about all of these things right now. I think that's the biggest, probably the biggest gift my parents gave me, which is the permission to be curious without any outcomes. And it wasn't easy because we used to live in a multi storied apartment building. And I have often felt curious as to why things fall the way it did. And I have been known to throw a few things at the same time just to figure out what falls down first. So they didn't have it easy, but they gave me the space to actually do what I needed to do with boundaries. Jason Baumgarten 02:16 When did that instinct of taking things apart shift to the instinct of putting things back together? Anand Eswaran 02:29 I think it comes back together. It's always something which happens at the same time. The first is you're trying to figure out, oh my God, why does this work? And then once you take it apart, then it becomes a challenge. How do you bring it back together? Can I make it work again? So I don't think it's one or the other. It always was. How do you figure out why is something really working the way it is? And how do you bring it back to what it did before I took it apart? Anand Eswaran 02:56 So it was all these things which actually happened together. Jason Baumgarten 02:56 I certainly remember my first computer. So I'd be curious if you think back to that era, of that first machine, which I'm sure was a wonderful color of beige, like most of them were at the time. By the way, I'm very impressed, you had 40 megabytes of hard drive, I think I had five. You were well ahead of the curve. What did that let you do at that time in your life that nothing else around you could let you do? Anand Eswaran 03:14 I remember with crystal clarity when I got my first PC. Things were pretty expensive in India at that time, so getting a PC wasn't a straight line. And so 40 MB, yeah, this was 91 probably, I think. And you know what was fascinating was, for me, the entire construct of the PC was around the games you could play there. That's what sort of made it fun. But pretty quickly, it transported you into a different world. The game I remember playing the most was the Prince of Persia. Anand Eswaran 03:56 I don't know if you remember that, Jason. Jason Baumgarten 03:56 I was going to ask you. King's Quest or Prince of Persia. Anand Eswaran 04:16 It was sort of Prince of Persia back in the 90s, playing on MS-DOS, completely pixelated, before computer graphics was a thing. And the manner in which you could construct entire worlds around you for nothing. That was probably the beginning of my obsession with computing. How you could actually manufacture an entire world which was capped by just your imagination and what it let you build and what it let you explore was actually fascinating. And then, as you see in LinkedIn these days, every little thing people write about, they talk about the five things I learned, none of those. But simple things like the game of Gorilla, which was just a simple game in which two gorillas were throwing a banana at each other. Anand Eswaran 04:48 But you had to put in the angle and the trajectory and the distance and all of that. You don't realize some of those concepts actually hold you good for a very long time to come. So I think my world revolved around this little pixelated PC. But it led you to imaginations far beyond what you realized and how your brain was being wired by it. Jason Baumgarten 05:18 Well, one, I will now know never to get in a food fight with you because you will know the angle to throw the banana. But it is incredible how many of us learned in that environment. And then you think about, and we'll talk about AI later, but the worlds we can now put ourselves into, and the capability is so amazing. You then went from Mumbai to Missouri, which is not a path everybody goes on. And that's not a small move. Jason Baumgarten 05:37 What did you have to unlearn in that first year that had nothing to do with the academics you were coming to study? Anand Eswaran 05:37 Yeah, I mean, it was different in so many ways. I just had met this professor in the University of Missouri in Columbia who was working on robotics, who was working on data, and those conversations just led me there. So it wasn't like I had this process I was in trying to get to the US, it just happened before I realized it. And so I hadn't even had the time to process what it means, you know, moving to a different country. Anand Eswaran 06:16 Culturally, very different. Very different for food. I was a vegetarian all my life, and I still remember coming in, walking into a McDonald's in Missouri and asking for a veggie burger. And I still remember that deer in the headlights look, you know, of that individual, like, what the hell is a veggie burger? So everything was different. Anand Eswaran 06:36 You know, I had to think differently about how I would work with people. The IQ of the person in grad school was very different. So here you are, used to a different environment, and you're suddenly coming in. People learn differently. The classes were different. You know, the professors were not telling you this is what you do. And there was not a fixed syllabus. They would pick up a concept and talk about the concept, and you were sort of on your own, trying to figure out how they create a roadmap for you. But your journey of learning is your journey of learning. In India, that's not the case. Anand Eswaran 07:14 You're told exactly what you need to learn, what books you need to learn from, what are the specific questions you need to learn about. So the whole construct of learning, I think I owe a lot of my opening of the minds to the university system. The academic system here in the US is absolutely brilliant because it just opens you up to a new way of learning, which I think is critical to the world we live in. So very, very different. Anand Eswaran 07:40 And I'm assuming it was different for you, Jason, because you grew up here, but when you transition from high school to university, did that change as much as it did for us? Jason Baumgarten 07:40 You know, it's interesting. I think that for me, a lot of it was around being able to explore the things that I was really passionate about. I think, you know, most high schools around the world are a little bit more, you will take this because we've dictated. This is the curriculum that you're supposed to know by the time you're in 12th grade. And then you get to college and you have choice and you have the ability and the freedoms to explore. You know, in my case, I did tons of cognitive science and psychology. I did tons of economics. And it was so much more fun because I was learning things that I really wanted to learn about. Jason Baumgarten 08:24 Similar to your experience, a lot less specific in terms of like, thou shall learn about this and more, we're going to point you in a very broad direction. I mean, I loved college, I loved grad school after that. I think there's still a real shift in environment. And it's interesting, when I went from the east coast to the west coast, some of what you're describing of like, landing, you know, the stranger in a strange land moment of like, I'm asking for the veggie burger. There's certain things you don't realize even within a country are so not normal in another region. Jason Baumgarten 08:50 Right now, here on the west coast, you would never think about getting a styrofoam container. When's the last time you went to a restaurant and got a styrofoam container? You go to the east coast, there's a lot of styrofoam containers, little things like that that you don't even think about become very normalized. And I think that as you move to different regions for school or through work, I know you travel constantly like I do, you start to pick up on all those nuances. Jason Baumgarten 09:23 And I think we'll talk about you as a human leader, because I think it's one of your distinctive qualities is how much you care about the people and how you lead with humanity. But I do think that picking up what's different and acknowledging it can be very powerful for people. Anand Eswaran 09:23 Absolutely, absolutely. Jason Baumgarten 09:23 You've said at times that your whole career has kind of set you up well for Veeam, and I laughed in the intro that we all think careers are very straight lined after the fact, but at the beginning or in the middle, sometimes it's just not quite so clear where it's headed. Was there a moment along that line where it did not feel like a line and it felt like, oh, my God, I've got a wrong turn, however it resolved? Anand Eswaran 10:04 I'll be honest, Jason, I don't think I went along my career trajectory with a plan. I really didn't. I wasn't like, I've got to get to this place, and hence these are the next six moves which will actually make sense. My career was doing things I enjoyed doing. Anand Eswaran 10:24 So if I look back and look forward, this is probably one thing I can say here. When I talk to interns and fresh out of school graduates, there's one thing I tell everyone, but that wasn't so obvious for me then, it's obvious for me now, which is I tell them, treat your career like a series of nonlinear steps. Because that nonlinear step is what gives you range. If you just go, you know, what is my next step and how does that get me a promotion, and I'm just going in a very linear fashion, it gives you altitude, but it doesn't give you range. Anand Eswaran 10:56 And so when I look back at what I did, it was just that it was a bunch of nonlinear steps which wasn't honestly planned, but it was just things I like. I did roles from when I came out of school, I was actually doing research, building a data warehousing practice in my first job after my master's with what was then a company called Braun Consulting. And then I moved from there to consulting to sales, went back to services. When I came to Microsoft, I came in through leading the entire services business. Anand Eswaran 11:33 But what I probably remember most fondly was setting up Microsoft Digital, which is moving Microsoft from not just selling tech, which is what they are so good at, but being a partner with every company they worked with on their digital transformation journeys. So that was very different. And then moved on to run the enterprise for Microsoft, went on to run RingCentral as a chief operating officer, which Jason, you were the liaison which brought me to RingCentral. But that was very different. A company focused predominantly on the mid market in an area which I had no idea of, unified communications. Anand Eswaran 12:12 So a variety of different roles, not planful, a variety of different customer segments. I think I've worked in every segment from mid market to SMB to enterprise and a variety of different products. I started my research, even my master's was in very large databases, VLDBs at that time, went from there to setting up data warehousing research practices, to applications. A large part of my career was enterprise apps to cloud digital transformation, comms, and now what feels like entire circle back to data. Anand Eswaran 12:46 And that's why I said it feels like my entire career was with this, which is bringing me back to data. You can talk about everything from frontier AI models to compute to OpenAI, Anthropic to Nvidia, all of that. But what fundamentally doesn't change is the value of, they all create for you is based on data. What was structured, what now is unstructured. Anand Eswaran 13:13 And so it feels my whole everything I've done is set up for this moment to be with Veeam and to make sure that we actually do what we do well, which is make sure that data for every company is the most valuable asset. And you position it to create the maximum value out of all new technologies coming forth, whether it's AI, machine learning, or whatever comes next. Anand Eswaran 13:52 So this wasn't a fit to a spec. Honestly, you talk about fit so much, because the spec for Veeam CEO, actually, I read it, it said a proven CEO, a product engineering pedigree as the first lens, and a backup industry veteran. I was none of the three. And so for me, I felt that the conversations led me here. Fit wasn't a match to the spec. In some ways, the fit was a match to where the company needed to go or wanted to go, which is literally what came out as part of the conversation. Jason Baumgarten 14:24 Well, and I think the other thing that is very clear to me, looking outside into and knowing you and the company, is you are hired also for who you are. And I think sometimes, and I talk about fit a lot, and people always assume, I mean, you know, experience fit, and actually, experience fit is often not the right thing to look at. And I often use the example of Lou Gerstner. When we were privileged enough to recruit him to IBM many years ago, people were sort of surprised. They said, well, he's not a tech guy. The thing that Lou Gerstner was amazing at was seeing the pattern in a business and taking a very complicated situation and zeroing down on the thing that needed to happen. What was the change that needed to happen? What was the nucleus of value creation? Jason Baumgarten 14:47 And he was a great simplifier as a leader, which is super important. I mean, I think a lot of leaders make things more complicated, then realize nobody's following them because they've made it too complicated. And so when you look at all the things Lou Gerstner did in his career, they fit that spec. But if you had made the spec about IT services or hardware or whatever else at the time, you wouldn't have stumbled on him. Jason Baumgarten 15:14 And I think that looking not only at the service level characteristics, but the underlying ones are very important. And I think for you, some of how you lead and who you are clearly ended up being even more important than did you have a data recovery background, or were you a proven public CEO, or some of the other things that were likely in the spec, particularly going from founder led to founder inspired. You've got to be a very good human leader. Anand Eswaran 15:56 Well said, well said. But one of the things you did, some research you published in 2025, Jason, where you said 84% of the newly appointed S&P 1500 CEOs are first timers. Does it still continue that trajectory? Because I'll tell you, while I loved the conversations, and at the end of six months I was ready to come in and be a part of Veeam, I wasn't sure they would think the same way. The board, Insight Partners, Mike Triplett, who is the principal for Veeam, would think about it the same way, because I was a first time CEO, and it takes courage to bet on first time CEOs. How do you see that trend happening right now, given that you probably have the ringside view for most of the big CEO positions coming in? Jason Baumgarten 16:30 I'll back up a little bit. So the first thing I always ask people is what percentage of firm performance can you draw back to a CEO? It's a very important question because people assume, well, I'm hiring the CEO, it's 100% of the firm's performance. It turns out, and there's been literally hundreds of studies on this, the CEO accounts for 15 to 30% of firm performance. Now that's a big difference. Jason Baumgarten 16:51 If you're an investor, getting 15 to 30% lift from one company to another can be the difference between a top tier investor and a bottom tier investor, but it is not 80 or 100%. So the first thing is to acknowledge that the CEO only has a portion of the lift. They don't have the whole thing. The second thing is to recognize that for larger companies, that decreases. It's on the lower end of that range. Jason Baumgarten 17:17 For very, very small companies, it increases. Right. So if you're running a 50 million dollar company as CEO, you actually have a bigger role or bigger impact because you're standing in for more of the executive team. When you're at a multibillion dollar company, you've got a very seasoned team, they know what they're doing. I don't want to say they need you less, but they need you a little bit less than a 50 million dollar company. Jason Baumgarten 17:39 So that's very important. The other thing that's important is the longer and more successful a CEO is, the less they tend to matter. Which is ironic because most boards think, well, we couldn't possibly lose Joey, Joey's created all the value. Well, it turns out that's exactly when you can afford to transition away from Joey. Jason Baumgarten 17:57 So that's just important context when we think about the impact of a CEO. And by the way, still huge impact, but it's important to acknowledge not 100%. So then you look at the data in public companies, 85 roughly percent, so very similar, are still first timers. And there's two reasons for that. If the company is successful, boards are much more likely to promote the insider. And the insider almost by definition has never been a CEO. Jason Baumgarten 18:25 Right. Whether it was Satya at Microsoft, never been a CEO. I would argue, you know, almost no founders have ever been CEOs before. So they barely had jobs before. So we have this situation where most people are in their first time job. And then if the company isn't performing well is when they want the proven CEO the most, but where they have the least leverage to get it. So it's also often the case that they get a first time CEO. And what's important is there's very little statistical difference in the performance of a first time versus repeat CEO, in large part when you control for the ingoing momentum. Jason Baumgarten 19:02 Because again, if you're a first timer, you probably got something that is going well. And if you're a repeat, you probably got something a little more beaten up. So the reality is that there is no real statistical difference in performance. Yet I think emotionally, particularly for investors, it is much more comforting to say, I've got somebody who's already done it, and largely because their ownership periods are often very short, and so they don't want to go through the period of a new person. But again, you look at tech, I mean, you work in the tech industry, if you made a list of the 10 most valuable companies that have contributed the most to market cap of last decade, they're almost all run by first time CEOs, most of whom weren't qualified for any CEO job, let alone the one they're in, because they were, you know, students before that, or dropouts, or never run a P&L before. Jason Baumgarten 19:54 So again, it's why I always say you got to look a little below the traditional spec and be very thoughtful about what is the quality you need. You might need curiosity and problem solving, not experience dealing with the board of directors. Anand Eswaran 20:26 Totally, totally. But Jason, there's one other thing you said I completely agree with. Intuitively, I didn't know the stat that the impact of a CEO is 15 to 30%, because I tell you that as I look at Veeam, the success we've had is driven by the quality of the talent we have across the company. Anand Eswaran 20:27 And especially as I look at the leadership team, and when I say the leadership team, it's not just people who directly report to me, it's people in critical roles, the impact they have. The key test I have is, does my head of technical support know more than all of us on technical support? If we have to educate him or her on what that is, we have failed. So I think part of it is just that, which is, do you have a team which comes together in a good way to move the company forward with the right friction, constructive friction between us? That's why I think it's the system. Anand Eswaran 21:03 One of the things which I believe Microsoft's success is, is that one of the things they have going for them is that the system is larger than any one individual. The whole system, the way it's constructed, I think that's the key. If we can leave the business better than when we came in, then we've done our jobs well. It's not just what happens when we were here. And building that system is actually the most important thing which I can think of, when I think of how I leave this business eventually at the right time. Anand Eswaran 21:35 It's about creating the system and creating the processes and creating the workflows. Jason Baumgarten 22:01 No, it's very true. And I think obviously it's in the limelight right now because everyone's talking about process flows and workflows and automation and AI. But how you get work done, I mean, if you go back to Coase's theory of a firm, it was the firm should exist to reduce the friction of things that would be too expensive or too impossible without a firm. Jason Baumgarten 22:02 Well, what are you doing to make that better? And, you know, often it's like, well, we're just doing it because we did it yesterday. It's like, okay, what are you doing to innovate on that? Going to your point on systems, though, you know, at Microsoft, you had tens of thousands of people. Jason Baumgarten 22:16 And then you get to Veeam, which is a fraction of that. It's an amazing company, and you've killed it. But when you got there, a small fraction, what did you look around and realize that that muscle of the scale wasn't right, or wasn't fit for purpose in an organization that was smaller? Anand Eswaran 22:41 The first thing is, I think there's a scrappiness in a startup, and I wouldn't even call Veeam a startup, and I came here, you know, we were a billion in revenues, you know, we are going to be two and a half billion at the end of this year. But there's a certain scrappiness which is pervasive in a founder led culture, and I did not want to lose that, which is different in a company like Microsoft. The second is, I had to sort of fire the operator inside me. Anand Eswaran 23:04 That's how I woke up every day, whether it was Microsoft, whether it was SAP, whether it was RingCentral, I was the person who executed a plan which the CEO came up with. The CEO muscle is doing, the CEO muscle is deciding whether something is worth doing and repeating it until people are sick of hearing it. And so nobody actually tells you that my job is actually the chief repeating officer. Anand Eswaran 23:33 So I had to find a way to fire the operator within myself and sort of pivot to a different muscle, which was hard initially for me. And then I had to sort of unlearn certainty. Everything I did in the past was about confidence in what I was trying to do. And the reinvention was saying, I don't know, but let's find out. Let's find out how we are going to go do this together. Anand Eswaran 23:59 I don't know if this is the right answer, but we're going to sort of pivot to it and figure out if we can together find out what's the right answer. So you're going from the biggest chair in the room and discovering when you do that, you actually buy more trust than any specific precise answer which I could give my team ever did. And by the way, AI makes the bluff worthless anyway. The smartest thing in the room is not a person anymore. Those were two big things. Anand Eswaran 24:27 And outside of that, I had to reinvent my relationship with speed, because in Microsoft, my job was to make a big machine move. At Veeam, my bias for action is kept tipping into impatience and too much detail. And that reinvention was critical. Delegate what is important, but don't abdicate it. And there's a very fine balance between delegating and abdicating. Anand Eswaran 24:53 And don't abdicate it, and obsess over the why and the who. And so I am a lot calmer about being wrong right now, and a lot fiercer about being slow. So I had to completely rewire my relationship with speed, with which how we operate here at Veeam, versus what I could have afforded to operate at Microsoft and SAP or an HP. Jason Baumgarten 25:29 To pull on another thread, I mean, you spent an enormous portion of your career in services, and then you shifted to product, and obviously data as a component of all of those. But I think that for many years in software, services and product were always a little bit more at odds. I think at the moment we're in this unique time when there's more coming together. But where has that services brain been a huge advantage as you've tackled this current role as CEO, and where do you have to fight it a little bit? Anand Eswaran 25:57 Yeah, I mean, you're right. I mean, services was sort of, I used to call it a necessary evil within the concept of software companies. Anand Eswaran 25:59 You're at a point which is very interesting. The software companies need to move a little bit towards the services element, because in the construct of the businesses we are in right now, it's not getting a new customer which matters, it actually is expanding and creating continuous value for those new customers which matter. So the biggest benefit of a services brain is a new. I don't know if you remember you had all these sales kickoffs, and the CRO would go stand up and say who is in this and who's in customer support and who's in something else and who's in sales? Anand Eswaran 26:35 And the expectation was everybody would stand up when he asked the question, who's in sales? The question is different today. The question for which everybody needs to stand up is, who is in customer success? Because landing a customer is the smallest component of what you do in the construct of a SaaS business, a software business. Today, expanding the value you create for your customers is life itself. Anand Eswaran 27:02 Retention and expansion is the new growth, and the foundation of that is how you operated when you were a services business. And so it's fascinating how I think again we are coming full circle on retention and expansion being the most important thing. And the second thing, which always was more critical for services and now you hear it everywhere, which is culture. Culture is a cornerstone of services, because services is a people business. Very different when you were running sales in a software business. Anand Eswaran 27:34 But culture is probably the most important and sustainable thing for a company's success at this point in time. And when I say culture, culture is not just the softness. It's not just am I nice to you and do you like me or any of that, which is a misnomer. Culture has specific elements to it. When I think about culture, I think about, do we have a culture of zooming in and zooming out. Anand Eswaran 27:57 You don't have the luxury of either being the strategy person floating at a 30,000 foot level or only being in the details. How do you zoom in and zoom out. How do we create clarity for every single person on the planet? How do we make sure that the system is larger, as I said earlier, larger than the individual? How do we make sure that our culture is very grounded in data being the cornerstone of every decision we make? Anand Eswaran 28:21 So culture was critical to make sure that a 25,000 person services business was moving and growing in the right direction. And that is huge. And I think that's been a cornerstone of my leadership or how I think about leading. And that has paid rich dividends in building that cultural thread in every company I go to. And that is a large part of what we do here at Veeam as well. Jason Baumgarten 28:44 This idea of things like culture being way more important when you're trying to land that value. And it used to be, if you sold it in, it was very hard to pull it out. SaaS churn was never, at least conceptually, as easy as it is now. I think it might have been a little overblown, but certainly conceptually never as easy. Jason Baumgarten 29:05 We talked a little bit about AI. I know you're using it a ton personally. How do you think about it in terms of how it's changing what you do as a human? And then as you think about your leadership job, how is it starting to shift there? I know some things are quicker than others, and then maybe where do you see it going? Anand Eswaran 29:24 Yeah, it's an interesting thing, and I think we're all early on. I'm sure you have very interesting observations given the range of companies you work with. The way I think about it is, one, AI makes a material difference. It raises the floor of incompetence at its most fundamental level. Anand Eswaran 29:44 Everyone is able to do things slightly better, slightly quicker, slightly more productively, and there's huge value in that. So everyone should be on the journey of learning with AI, on the journey of how do you apply AI to your job in a better way? Because the reality is AI is not going to replace you. A person leveraging AI better is going to replace you. Anand Eswaran 30:06 That's the first layer of how I think about AI. The second layer of how I think about AI is the impact it has on companies. And when I say that, what I mean is just using AI to do things quicker is a very small percent of the impact AI can have. The bigger impact AI can have is we should be rethinking and reimagining every process and every workflow we have in a company. How should that look in an AI first era? Anand Eswaran 30:36 And that reimagination of process is the biggest impact AI can have on how a company operates. I'll give you a simple example. We have an AI bot we have paired to every seller. And what it does is the seller has this conversation with the AI bot before every meeting, before a big high stakes sales opportunity. And the bot helps the seller get more precise in the conversation they're going to have and get more productive in the outcome we expect them to have. Anand Eswaran 31:08 And so that just changes the dynamic and the workflow of how a seller engages in the sales opportunity. Massive things. As a company, in the last two years, we've gone from what historically was a CAC payback of 30 months, to we're going to finish this year at roughly a CAC payback of 12 months. And that impact has been driven by reimagining the workflow of how we train our sellers, how we pair our sellers with interesting tech, and how we basically look after the fact to make sure that what are we learning from these conversations to change the workflow of the content we create in the first place. So that's the second thing, which is super critical. Anand Eswaran 31:50 And then the third thing is how do you as a company use AI effectively in the products you build for your customers? And that's where we are fortunate to be in the sweet spot. I mean, we live in a world of data. And what we now see is, you know, I think, you know, Sam Altman said this, which is the eventual bottleneck is not going to be models and compute and any of that. The eventual bottleneck is, you know, our opportunity is going to be, can you trust the data the AI is actually acting on? Anand Eswaran 32:20 Can you actually trust it? And that's what we are doing as a company. We are basically working to create that trust across the entire lifecycle of data, from the time it's created to the time it actually engages with anyone, whether it's a human or a non human identity, to the time it's protected and it's in cold storage. And so basically making sure that trust exists along the entire lifecycle of data is the key to unlock AI, not just for your company, but to unlock that entire data access to all the partners in the ecosystem around you, which is your force multiplier. And that's what we stand for as Veeam, which is accelerating safe AI for every company at scale. Anand Eswaran 33:02 So we feel privileged that we are able to actually work on all stacks of this AI layer, which is personal productivity, reimagining the company's workflow, and then creating products which allow every company to take data as a trusted element and unleash it without concern on the AI they use. Jason Baumgarten 33:50 This is such an interesting area. And one of the things that struck me is last week there was an article about one of the large language models buying the data of a defunct company. And I thought it was such an interesting article because on one dimension, I'm sure there's great rich data that is worth buying. On the other hand, I thought to myself, boy, we're going to teach a model based on a company that actually was unsuccessful. Jason Baumgarten 33:51 And as you think about, because you've worn the hat of advising companies on their digital journeys, and you've now worn the hat of empowering those journeys, and we've all lived through the era of like, everyone has big data and big data is great, I think we're all sort of realizing some data is great, some data is not so great. How do you think we're going to have to rearchitect the learning loop or the metadata around all this data that people are assembling and trying to learn from, so that we're really tuning models between the bad example and the good example? Because certainly if you have a thousand sales examples without the extra, did they succeed, you could very easily train on the wrong stuff. And that's something I don't think we've fully addressed in our environments, in our corporate environments. How do you think we're going to have to change that? Anand Eswaran 34:49 It's a great question, Jason. It's a great question, and I don't think there's a clear answer to it yet. Anand Eswaran 35:46 The way I think about it for Veeam and my company is that we are creating a version of the frontier AI models for ourselves, because as you said, rightly, the large frontier models are trained on all available data, which doesn't discern the good from the bad. And not only doesn't discern the good from the bad, but humans are a very biased and flawed species, as we know. And so these training data sets also bring in all of those biases along with that. And so it will eventually, if you're not careful, lead to all of the same questions on fairness and equity and all of those questions which we grapple with as a human species. We are actually training the models on those very problems which exist. Anand Eswaran 35:47 And so as you think about then, how do you take it and how do you personalize it within the construct of a company to make sure that doesn't happen? Number one, what you said is spot on, which is how do we take the right examples and make sure that those right examples are seeded with a better weightage on these training models. And that's the first thing. And that's one of the key things we're doing at scale. And we are on that journey right now. Every company will have to have an SLM, which is derived from some of the LLMs, but the SLM, which is specific to your company, elements which, you know, you can put guardrails around like bias, and then making sure that you actually train it on the right outcomes, which you have had as a company. Anand Eswaran 36:23 I think that is super critical. The second thing is, still life itself, which is you automate the reversible, you've got to own the irreversible. So there are elements like, you know, I would use AI in an interview process to understand research and maybe connect dots, which I can't do in an easy manner. But the final judgment of who you hire and why you hire that person has to be human. Anand Eswaran 36:52 And it's not just a human, the process is important, as important as a destination, which is you bring people together who have had this conversation. We have a live conversation and we make a live human judgment. So you have to also figure out what other processes and the decisions which are absolutely essential for the final judgment to be human, and what are the processes which are irreversible. Anand Eswaran 37:18 And it is absolutely critical for the final judgment to be human. So I think we are on this journey, we are very early in its innings. We are on the first inning of a very, very long game. And I think we're all going to learn as we go on this one. Anand Eswaran 37:31 You have this ringside view of every big company and small company out there based on the work you do. Jason, what do you see which is an interesting lesson or a good few nuggets you picked up on this? Jason Baumgarten 37:41 I think what we're seeing in general is a lot of companies at scale really struggle with adaptation in a large way. So small companies or AI first companies are designing their whole system around the AI engine. Jason Baumgarten 38:01 Large companies are really saying, boy, we'll implement it in our contracting process. But they're not realizing that the contracting process might start much earlier in the sales process or the marketing process. That happens even before that. They're sort of addressing a piece of it without addressing the whole. Jason Baumgarten 38:20 And unfortunately, it's very similar to the individual becoming more productive, but the organization not. When you do that in pieces of a functional workflow, I think you see pieces getting more efficient. But you and I both did lots of consulting work, if you can't actually take humans out, or you can't take capital investments out, you don't get savings, you just get dead air. And so I think that more people are going to have to take bolder approaches to the whole way in which we work. So that's one. Jason Baumgarten 38:53 And then the second is, I don't see as many people scoring the outcomes. I see a lot of people recording the workflows, and lots and lots of work on getting the workflows more automated, but without as much thinking about the downstream effects. And the reason is those are harder, we often don't have them for a while, or they're disconnected with the work being done. So even in areas like sales, where you have a number at the end of the day.