 Welcome back everyone, this is theCUBE's live coverage here in Las Vegas for SAS Innovate 2024. This is the Analyst Roundtable, Analyst Angle. I'm John Furrier, your host of theCUBE with Dave Vellante, Rob Stratche, both with theCUBE Research, and Ray Wong, who's a CUBE alumni and contributor to theCUBE Collective. Also the founder of Constellation Research, legend in the industry. Probably one of the number one analysts in the industry. Ray, we see each other in events all the time. You know, it's season, it's kicking in again. All the events are happening. You have the luxury of having that horizontally scalable observation space of all the different companies, the stories, AI washing, who's got the real deal. You're starting to see a lot of trends. What's your seeing? Obviously the big year, we talked about last time on theCUBE was let's find those workloads in production. That's going to be the test this year. What are you seeing? What's your observations? You know John, it's more than just the workloads in production. You're trying to figure out who actually has the goods and who doesn't have the goods. And you're right, there's a lot of greenwashing going on or AI washing as we call it out here in terms of what people have. I guess my kids would say, look, we could have done this in class with like 10 of our classmates in 30 days for half the stuff that's been announced so far this year. But what we have really is the difference between who can take something from data to decisions. And that's what SAS has here. And what they've been showing is really where models come into play, what you can do with the data. These guys are steeped in this. I'm more talking like decades of experience and what they're trying to do is package it so it's going to be simple for customers to use. And it's been one of the things that have been burning SAS for a long time. They've been seen as the legacy analytics company. They're slow to make a move. But you know what, it's in their benefit right now because what it's showing up as, how do we make things easier with all the knowledge we know to get AI in the hands of people. And if there's anything, that's what they're trying to communicate at this event. And the AI really gives them a real accelerated tailwind to modernize and get their new opportunity to reshape their brand. Because look it, let's face it. You don't want to be grandfather's tool from the 90s and 80s or even 40 years ago. So, but they have all the data. And last time you and I chatted, you know, we were talking about things like procurement, you know, is going to win from this new multimodal AI capabilities. So guys, I want to throw this out on the table and get your all, all your reactions to this. There's this idea that you don't need to raise the house and bulldoze everything. You can retrofit your workload. If you got an end to end workload, like a procurement workload, the last time we were talking or whatever, and you got it, okay, it's a known app. It's a workload. Well, you got to bring AI to it. So you're redesigning or retrofitting versus rebuilding the whole thing. Do you guys see it that way? Or is that just for known workloads? Or are there new net workloads that are being built from scratch? And when do you tear down the house and rebuild? It depends if I may. So for these guys that the challenge that SAS has is they got, and you know this from looking at the data, the spending data, they got a lot of old legacy stuff that is big, it's a big part of their install base and the new stuff's not, it's growing at 30%. We heard that yesterday 30% and several businesses, which is great, it's good growth rate, but it's not big enough to offset the decline. But they are monetizing. The 10% decline, and they are. Okay, so that's cool. But then you take a company like Blue Yonder that you know, and they're kind of rewriting their whole stack. It's like the old Minugistic stack when they're saying, all right, give us some snowflake and some relational AI. We want to reimagine the entire supply chain. So for them, it's appropriate, right? Now they still have to sort of manage the install base, but I think it's a lot easier to do that. They could say all three are in place. It's a lot easier to do that as a private company, I guess is the point. Yeah, but I'm talking about the customer's workloads because there's cost concerns. Okay, I don't know what I'm throwing this against. I'm going to throw workloads at this. Or if I have a known workflow that's scoped and you know what it is, you can guess at least and then figure out the AI so you bring an AI to the table, which is maybe using AI or building AI. So this is kind of like the conversation, Ray. What do you think? Well, I think, again, what they're doing is they're trying to bring in new people. I mean, they spent an awful lot of time talking about how you can manage Kubernetes earlier today in the keynote, which I was like, okay, this is kind of a little bit odd, but they're using it for exactly that, the cost aspect, how do you scale up? How do you scale down? And then they talked about old ways of doing things batch and how you scale out batch. Plus they also then had single store, you know, Raj from single store on stage talking about streaming. So I think, again, they're trying to bridge that gap and do it in a way that says, okay, by the way, we're the guys who like invented doing ML and ML ops. I think that's that. Well, they got a legacy installed base. That's huge. These guys got aspirations. The big thing is they got to grow the ecosystem, right? I mean, if they want to compete with the likes of Databricks, they've got to have, you know, more partners because they can't do it alone. But they got the workflows. They're customers. They have a huge customer base. What's your take on this? You know, I think the important thing is like, they're basically doing three things at the same time. One, trying to modernize everything internally. Two, trying to actually call the AI story. And three, trying to run an IPO. There's replays going on here at the same time. And it's going to be interesting to see how they maneuver that. But if they get the AI story right, the IPO is taken care of. Yeah. What do you think about the IPO readiness? Do you guys think, you thought about that? I guess the point I'm trying to make is that if they can sort of create that balance, that equilibrium between the decline of the old and the rise of the new and there's a crossover, that's a great IPO story. I mean, they have the users. Back to John's point, right? What's going on? Every person in a role related to data, and if you're the expert, you're using SaaS, right? If you're hacking it, you're using something else. And that's really what they're trying to do is take the citizen data scientists, right? To like the experts and bring them all together. And that's why they're coming out across it from all different angles. And they got the workflows and data. That's what my observation is, is that if they could pull that off, and I think the monetization numbers, the 30% and what they talked about, is a good sign. The question that we don't know is what are they losing, right? Because when you go through these transition, we've seen every company that went to subscription or managed service, there's usually a hit to the existing business. What do you guys see? I'm not in the weeds on this. But the beauty of this with these guys, they have been subscription service from day one. The contracts here at SaaS have been like, I'm going to take a 30% uplift from your 1986 services agreement, right? Like every year it just keeps going like that. They were services before SaaS. Right, yeah. And I think that to that cost aspect of it, I think they've really worked with that. I mean, they brought on that this is the first year they've had a distributor and they're really focusing on channel. They're also focusing on people like AWS and the hyperscalers and as their other parts of their channel. And I think, again, to your point, I think they're making that transition and it's not an easy transition to be made, but transparency and their pricing out to the partners, they've had to get that right in the last six months, especially when you bring on a distributor. And they are adding new partners, which is the interesting thing. I mean, for the first time, they've gone outside and they're actually looking at indirect sales as opposed to direct sales, which has been what their natural sales motion. Right, but I guess they need to 10X that in my view and they will. I mean, if they have aspirations of competing, you know, with the big boys, so to speak, they got to do that. Well, this is all the IPO readiness because you brought up the IPO. We haven't talked about that yet on theCUBE. I think they're not denying it, but they're also not admitting it. No, but they're not ready yet because their numbers just aren't there. I mean, they're going to have to have some kind of stronger top-line growth in order to show that. They talked about 30% growth in revenue of Viya-based solutions. They talk about 30% revenue growth on Viya Cloud running on Azure. So that's good. I mean, you can always pick. And now they've got AWS. They're extending the AWS is great, but you always cherry pick 30% growths, right? You got to, at the end of the day, Wall Street's going to look at the top line. They're going to look at the bottom line and say, all right, how's it all shake out? And if they're flat to down, that's not IPO-able, right? They've got to be, I mean, it is IPO-able. It's just not what they want. So they'll, my opinion, they'll settle that. Look at what Michael Dell did as a private company. They'll settle that behind the curtain and then they'll come to IPO. Ray, what's your advice or take on their IPO readiness? What would you advise them? I think they're working on a lot of things, right? They're cleaning up financials. They're getting that up to date. You know, they're actually putting some systems in. The other piece that they've been doing over time is they're changing the way their Salesforce sales comps are actually set up, right? Their product releases are actually much more focused. I mean, they've done a lot of cleanup, but they're also trying to keep their SaaS values, right? This has been a company that's been taking care of their customers, their employees, and every one of their partners for a very long time, and they want to reflect that. This is also a legacy shift, right? This is going from a private company where Dr. Goodnight built, you know, set things up in place, and he's going to want to make sure some of those values carry out forward. So they're putting all that in place, and that's not his easy shift to put, but they've got some of the top people working on this. And the types of consulting, the types of people that they're bringing in tells them they are serious about an IPO. This isn't, you know, this is just an announcement. And they've thought, they've been through this before. They had a window a while ago, and they chose not to take it, but, and it was probably a good choice, you know, given everything that's changed. It should be a monster IPO. I mean, they just, they wanted to stay private. I don't think they ever wanted to go public. They didn't before, yeah, and now they want to go public, and this could be the most, this could be the biggest AI IPO in the market compared to any startup with like crazy valuation. These guys actually have real revenue, right? We're talking like 2 billion, 3 billion plus of real revenue. Right, and real customers. It's not high. It's north of 3 billion, you know, call it 3.2, 3.2, 3.2, 3.2, I mean, I just know. Yeah, right, I'm not under NDA for the revenue. But, you know, but it's again, I'll wink at the numbers. It's flat-ish, you know, right. But, you know, Kara Swisher made this point. She goes, hey, you know, there's a lot of hype and a lot of valuation going around AI. There's not a lot of revenue. These guys actually have real revenues. And I think we're even seeing that with some of the people who are looking at IPOs this year. I mean, outside of Databricks and that one in particular, you start to look out beyond them and you start to go, okay, really at that valuation or they're getting funding thrown at them at like nine, $10 billion valuations and you're starting to go, okay, that's kind of a head scratcher based on what we know about their sales and how lumpy it is or what have you. These guys have had a very good, you know, steady base and really fanatical customers. And Thropix got an evaluation, what, the mid-20 billion, 25 billion? I mean, what's Anthropix revenue? They're going to get smoked, this would be a great idea. All right, guys, I'm the customer. I'm the person I'm writing, big fat checks. I want to go public. I come to you guys, the advisory. What do I need to work on? I got a blank check. I got my team. Where are the gaps? I want to go public. I want to be ready. I want my products to be great. What do I have to do to get from a point A to point B? Ray, we'll start with you. What's the, you know, tell me what I need to work on. They've got to get this AI story down. They've got to show the subscription revenue growth as they're going from on-prem to cloud, which is one of the big challenges, but they've actually been showing growth and the bias you've been talking about, Dave. And I think the third thing really is they have to show that they've got a management team that goes beyond Dr. Goodnight over time and they've been doing that. They've been, they've removed and replaced a lot of people along the way. They've promoted a lot of people within and you're seeing a very, very different team and a very different mentality. We were at their SaaS analyst a couple of months back and I can tell you, it's a very, very different SaaS year than what would be just even three years ago. Yeah, they're on point on our AI stuff. They're on point. And they, from the explore, the last event we had theCUBE to hear, they've already delivered. I mean, they're first to sell models, lightweight models. I mean, that's pretty impressive. And they got the subscription. I mean, so I like what they're doing. They have good product shops. Rob, any thoughts on what you think that you need to work on? I think, you know, building off of what Ray was saying, I think you can look at how their go-to-market is happening. And I think, again, talking about, Dave was talking about partners and ecosystem, they really need to jump in full on with that because I think part of that will help them in that runway to get to repeatability, that ARR and give them that runway that gets them the eight plus quarters of successive growth that gives them the numbers that then they can go out with. Yeah, I agree with that. I totally agree. I think they got to really think about their go-to-market. They got to dramatically expand their alliances and partnerships. I do think, I'm excited about the AWS partnership. I mean, you know, okay, it's nice that they're doing Azure. I think AWS could be even more productive for them. Can you imagine SaaS on the marketplace? Oh my God. Absolutely blow doors. I don't want to see good night at Jassy, like, you know, on stage. I mean, that would be awesome if they know that. Did you see him yesterday at midnight? Hey, he was great. He's snarky as ever. We're not an interpreter. We can actually compile our code. I love that. I'm probably doing the compiling. Yeah. We have one person here that was a programmer in SaaS. Yeah. I did. I started my first job out of college was using SaaS to do data and data analysis. And I think, again, I've been geeking out this entire week here. It's been, you know, you feel the energy of the geekness here. That just is fantastic. Like you said, the semicolon. I mean, I never had a job programming in SaaS, but we learned it in college. You had to learn SaaS. You had to learn SPSS. There was no R at the time. Right. I never used SaaS ever as a product. I was like, honestly. That was before Python and R and all that stuff. Well, I mean, they got work bench. That looks really solid. I mean, the product side looks really good. And they're working hard. And you're right. We've been talking to the team too. We got to know them. Good culture here. This culture is going to have to turn into a growth culture. Ray, this is not the country club, right? So this is going to have to move from kind of that HP, the old school IBM with the big campus, and great people. Pull it off. HP did it, obviously. You know, HP is there. So all good stuff. The thing that I've been thinking about in this industry, Ralph, this is that is AI, I'm not saying a redo. Is AI an opportunity for the game to change the pecking order of the leaders? Because if you look at SaaS, and what I've been bullish on SaaS is that they got the customers that have the data. They got the workflows. And AI is absolutely a lever for them to catapult them to another level. And it's gettable. Ray, guys, what companies are out there that could really move the needle? Legacy companies or existing companies that could use AI to springboard and propel them to the front of the pack versus the hype side, which is like the anthropics of the world which have massive valuations. Who's right for AI to go from challenger or pretender or, you know, forgotten, steady as she goes? Strangely enough, Oracle, right? Oracle's sitting on the same kind of assets that SaaS has across the board, whereas Oracle Cloud is actually operating at one seventh or one tenth the cost of everybody else. So they've got all the pieces in place to actually do that. And they've been running ML for quite some time. And they've also, they were smart enough to buy a lot of NVIDIA GPUs early at a really lower price than everyone else. So much like, I mean, we saw Bing searches now run on Oracle Cloud. I mean, that's a laugh every time I hear that. I think it's a great point. I mean, to me, it's not strange. I think Oracle, Larry's always invested in R&D. So you got to give him props for that. It's geek to geek, yeah. I think, yeah, totally. I think I do think IBM, I've said, I haven't been excited about IBM in 10 years, but I think actually the mistakes they made with Watson 1.0, trying to put it in places where it didn't belong, I think they learned a lot and they've actually got some pretty good products and they could have PLG. When was the last time IBM had product-led growth? It's been a long, long time. I think beyond the picks and shovels, you're starting to see Adobe. I think Adobe has done a pretty good job with it. It really comes down to who has assets that are in line with AI. That seems to be the trend. Yeah, I agree. I think IBM's in a great shape to do that. They've rationalized what they're actually doing. They know what they're not going to do. I think that's a very important point. You guys have talked to Roger Premo. I mean, these guys are focused. You see what's happening on Rob Thomas' side and you also know, like, I mean, they're investing heavily. Arvin's investing heavily in assets that actually make things easier for customers, not harder. And I think when you think about the stuff that's going on in the research labs, Dario Gill and Rob Thomas in the go-to-market, they're very much aligned and that's been misaligned for years. They say, hey, we're going to do all these science experiments and they're never going to get to market. And that's always been a huge problem. I think Arvin's fixed that. Rob, what's your take? Yeah, I think, again, when you get below that level, I think there's people like Teradata, Informatica, obviously SAS and a number of others that AI is going to help. These are people who have the pedigrees. Also Oracle on the Exadata side, they're sitting on so much data and they have customers that use that because the stuff works. So I think there's a lot of really good pieces out there that could be enabled by AI. What's your take on Salesforce and AI? Salesforce has the pieces in there, but they've got to clean up their shop internally. So the data cloud stuff is really smart. It's the first place that can finally do all that. And it's doing well in the market. And that's actually doing very well. We'll see if they do this Informatica deal. That's going to be interesting to watch. I mean, we're kind of like, are you buying growth or is there really strategic asset here and we'll have to find out where that actually comes down to? Could it be both in your opinion? It could be both. They could be trying to do that. And then back to you on companies, like Bumi, Bumi's another one. Anyone's in this integration space that has to handle data movement is going to benefit from AI and in the process space as well. Companies like a Salonist doing on the process mining side and of course, UiPath trying to get that automation in place. They all have a big role in this AI. Okay, so what industries kind of fall by the wayside or better yet, which Magic Quadrants don't become relevant? Is ML Ops dead? Is it change over? Is AI Ops change over? I think traditional RPA is a good target. I think UiPath's done a good job of moving beyond trying to get to enterprise wide, but there's a chunk of their business in the old legacy RPA that Gen AI can do a lot of that. I'm waiting for an agent's tragic Quadrant in squares of the speaker. I'm not sure if there is a whole point out. Yeah, I think ServiceNow is another one that's leaning in where they were big into the RPA, but now they're leaning into the AI and trying to reinvent themselves. ServiceNow is interesting. It's that layer that abstracts all the crap from all the old transactional systems. You have to do that first. And so I think that definitely has a role. Guys, great analysis. I mean, action packed. I think we just basically solve all the industry problems and the final question as we wrap up, where are we on the industry change if you had to scope it or even kind of describe our historical life in tech? I mean, for me, I feel like this is a really weird time in a good way that it's changing. What's your thoughts, guys, on what is actually happening in the industry? I'm talking about all actors, all participants. What is your take, Ray? Because, I mean, the script has been flipped. Things are going to get really weird and hairy very quickly. I don't know about you guys, but I'm getting six to eight calls a day about people losing their jobs in the valley. And it is just happening. We're seeing hundreds of people being laid off every day. I don't know how they're being replaced, but there are three factors here. We're in the year of exponential efficiency. The consulting firms like BCG, Bain, and McKinsey all went and said, if you don't cut your cost, your stock price is going to tank. Everybody ran that playbook. We're now also in the middle of AI arbitrage, right? Where we've had this pyramid now being carved out into a diamond because we're actually using AI to do a lot of that work. And then the third piece is a margin compression we have never seen before. And I'll give you a simple example. Like, you take UPI in India, that's their payments gateway. It's operating at pennies per transaction. Our payment gateways are running at $1 per transaction or more, and that's actually creating compression. You're also seeing it, for example, you could buy Zoho at $100 per user versus buying all the other players that you need for $1,000 per user. Margin compression is here, and it's going to be interesting. And that's what's driving everything in the industry. Where tech used to be the disruptor, it's no longer the disruptor. It's the barrier. It's holding things into status quo. And so something's about to come and knock that over. And we're not sure what. And Google's the one company that hasn't taken a ton of cost out, but still could potentially. I think when you see these waves, it's like Kara was saying, the revenue's not there. And so the new stuff's not big enough to offset the decline in the old stuff. And so as a result, they're going after labor and to drop into the bottom line. You bring up Google, which is the one that was all over the news this week. About the fact that they're replacing people in finance with AI. And I find it really interesting that they're all, again, looking at the cost and taking that all out. It's not just them. I was talking to a head of Rebops at a company I won't name. They had it 10% automated. They expect to be 50% automated by the end of the year. They're going to have one half the finance people they need. And if you're not in sales leadership or in development, you're out of the valley. You need to be somewhere else like Austin, North Carolina, or New York. Ray, I mean, I think you bring up a good point. From my observation, I think we're in a tech where the tech is the glue, holding things together to your point. But also it's in everything now. So the growth has already happened. Now it's ubiquitous. So this incremental value, it's going to be workflows and data is the new IP. And I think that's a business leadership challenge. And everything else will be a subsystem of some hybrid automation. One more thing. Every customer is pissed at their vendor. They've got a gun to their head. No one's negotiating. People are trying to drive down costs and the vendors are trying to jack up costs. It's a tenuous relationship going on now. It's going to be a revolution and the customers are in charge. Especially in sales. Okay, so it's a revolution and the customers are in charge. I think customers feel like they're not because there's only two vendors to choose from in every category. Oh boy, it's going to be interesting. When that stuff gets set up. And only one in GPUs. Only one in GPUs. Okay, great. Everyone is down in the math there. It's going to be a cold little war going on. Guys, great analyst angle. That was super exciting. I'm John Furrier, Dave Vellante and the CUBE team. We'll be right back with more after this short break.