Motley Fool Hidden Gems Investing - Salesforce = Less of a Force?

Episode Date: May 30, 2024

Sales slow for the software company. But is that the end of the growth story? At (00:21), Tim Beyers and Mary Long break down Salesforce earnings and ask whether the stock’s slump is warranted. Then, at (17:22), Asit Sharma and Ricky Mulvey take a look at some companies that could ride the tailwinds of Nvidia’s standout earnings. Check out the Range Rover Sport at www.landroverusa.com Companies discussed: CRM, SNOW, AMZN, NVDA, AMD, MSFT, MU Host: Mary Long Guests: Tim Beyers, Asit Sharma, Ricky Mulvey Engineer: Tim Sparks Learn more about your ad choices. Visit megaphone.fm/adchoices

Salesforce slumps, but is the street missing something? You're listening to Motley Fool Money. I'm Mary Long, joined today by Tim Byers. We are both recording virtually, but from sunny colorado tim thanks for being here thanks mary fully caffeinated ready to go love that okay we are having a sunny day here but not such a sunny day for salesforce today we're shining the spotlight on that company because it's down more than 20 this morning that is largely due to the fact that the company just reported its first revenue miss since 2006 that sounds pretty bad, but I will say up top, bottom line results were solid. Gap net income up almost 700%. Free cash flow up 43%. Tim, we've got some good, we've got some bad. The street seems to think there's a

lot of bad. What say you? What stuck out to you as most important in this report? Well, the street doesn't like to be surprised. They were surprised. First quarter revenue coming in at $9.13 billion. That was under the consensus forecast that I got from Capital IQ of $9.147 billion. That's a miss. The street does not like a miss. They like it even less when you follow that up with a Q2 forecast that is way below what the street wanted. In this particular case, the midpoint in the Q2 guidance, Mary, was for $9.225 billion. The street wanted at least $9.34 billion. That is a serious miss. Also, looking ahead to the full year, Salesforce said that their guidance on margins, so this is operating margins. They issued operating

margin guidance in Q4, this is three months ago, and they said, look, we're going to hit our gap. Operating income is going to be a little over 20%. It's going to be 20.4%. Then this quarter, they said, nah, you know what, we have maybe a little extra stock-based compensation. We're going to take that down 50 basis points to 19.9%. Just surprising the street all the way around is really not a great look. Having said that, Mary, what stands out to me is that even when you subtract all of the things, so you subtract everything, you subtract stock-based compensation. You subtract all of the capital expenditures. You subtract the dividends. You subtract a $2.1 billion buyback of shares. And we can talk about whether or not they should be

doing that buyback. But let's just say it's fine. That's a lot of spending. And even with all of that, they still end up with $2.75 billion in free cash flow. This is a very healthy business, but it's not growing as fast as it was. Management's explanation for the slower revenue growth is that enterprise customers are working with tighter budgets. They're reevaluating contracts, taking a closer look at every penny spent. Does that mean that this misses more like a macro story or is it indicative of a longer term problem specifically with Salesforce? Well, this is something. So I'm going to use a dumb term that has been around the tech industry forever. So bear with me here, Mary. Elongated sales cycles. Elongated sales cycles.

This is something that we hear about in enterprise software everywhere. It's been around for decades. It's not new, but it happens to be hitting Salesforce right now. So is macro part of it? Maybe. Also, just in a year of efficiency, are companies going to be a bit more judicious about how they spend their capital, especially when you have investors, board members, customers saying, give me AI. I want AI. Have you spent on AI yet? I'm sorry, I haven't heard you. I thought you said you were going to spend on AI. When that is the narrative that's coming out, Mary, it's a little bit harder to go forth and spend on something that is strategic, like a Salesforce product, but is fundamentally a customer relationship management system. Now,

it's more than that. It does more, it manages what we would call front office operations. Front office, everything facing the customer. Back office, everything that's not facing the customer. We know what these things are. Marketing faces the customer, inventory management doesn't. Back-office, front-office. Salesforce is arguably the most significant and most important provider of that front-office software in the world. But there are other priorities here. You may have AI priorities, AI hardware spending, and so forth, just taking some money out of the budget. Yeah, let's talk about AI, because I felt like listening to this, I counted 57 mentions

of AI or artificial intelligence. That's it? I know, it almost felt like every other word was AI or artificial intelligence. So just based on that alone, it seems like that's quite a priority for Salesforce. That said, looking at this 20% plus slump, I've seen headlines that have attributed the revenue miss to Salesforce falling behind on AI compared to competition. So what's the argument there? Benioff mentioned in the call that other AI players, he called them commodities, says, quote, that's not where the intelligence lies. And his point was that there's an importance to data

here, right? And Salesforce has tons of data. Some of the numbers that they threw out are kind of mind-boggling. They process two quadrillion records of data. Salesforce manages 250 petabytes of data. So they're saying that that's a huge foundation for this AI game that is to come. Do you buy that? Yeah, I do buy that. I mean, I think it's a little bit arm-wavy, but it's also true. So all of those numbers, yeah, that's a lot. There's a lot of data that are in Salesforce systems. But he's making an argument that I think is fundamentally correct, that your AI algorithms, however you build them, whoever you use, whatever kind of hardware you've got is only going to be as good as the data that that hardware and those algorithms point

at. Whatever's pointing at the data can be fast, it can be resilient, it can be really expert. But if the data itself is terrible or not really useful, then what insights are you going to get? And so, what he's saying is that in the fundamental AI revolution where you are trying to automate real good workflow, you're trying to gain insights from data, the things that Salesforce helps you manage is fundamental to that process. So, he's trying to position them in the middle of the AI revolution. And again, I think he's not wrong. Now, having said that, is he being hyperbolic? I mean, newsflash, Mark Benioff is hyperbolic. No! That couldn't be true. Of course, he's being hyperbolic.

But at the same time, he is right that what they're doing here is fundamental. And I'll mention one other thing, and we could keep moving here. They did invent a product internally that is designed to make Salesforce a bit more sticky in this area. In other words, take the data that you already naturally put into Salesforce and then add other contextual data to it very easily, and they call that data cloud. They have a couple of features inside of data cloud that I think are interesting. First is, it's like a set of programming interfaces, APIs, so you're going to bring in data from, say, a Snowflake or bring it in from AWS and just these other sources that add to the richness of those customer records and sales records and marketing

records, all these things that you're keeping. Then there's another feature to it where they have what's called this, I believe they call it no copy or something like that, no copy partnerships. Essentially, what that means is the data comes in natively. You don't have to do any transformation of that data in order to get immediate value from it inside of Salesforce. That strikes me as interesting. It's not clear to me exactly how it works, Mary, but there is something smart about Salesforce saying essentially what Tim White has said, and I think it's one of the smartest things he said, and Tim says a lot of smart things, data is the food of AI. That's what Marc Benioff is arguing. He's making Tim's argument.

Data is the food of AI, and the most important front-facing, customer-facing data lives in Salesforce, so don't count us out. I'm seeing two stories here. On the one hand, we have a perhaps hyperbolic Benioff positioning this company as in the middle of the AI revolution, data cloud being a part of that, etc. Basically, the growth story is not over yet. Then on the other story that I'm seeing is this not insignificant slowing down of revenue growth. And even like, as you pointed out, this is a company that's generating a lot of cash. They're returning a chunk of that cash to shareholders in the form of share purchases, dividends. They're anticipating announcing a quarterly dividend next month. So taking all

this into account, what story do you see as being more true? This next growth story, or is this the beginning of Salesforce becoming a mature enterprise software company? Well, if you listen to the street, they will say, growth story's over. I mean, if we were to visualize it as a meme. You have algorithms as babies throwing cash into a big, giant bonfire. And they're all upset about it. And they're all upset about it. They're all bent out of shape. I think two things can be true here. I think that Salesforce can be maturing, that growth rates can be sliding into the high single digits, low double digits for a long period of time, where there's dividend growth over a long period of time, improving margins

on a company that will just not go away because it's too sticky to remove, Mary, that's not just a good business. That's a great business. That's a great business. Can you name other businesses that fit that same profile? I'll give you one. Microsoft. Microsoft fits that profile. Now, is Salesforce going to be the next Microsoft? No, I'm not going to predict that. But if the features that, if the contours of Salesforce are becoming more like the contours of Microsoft, why would you be not okay with that? That's a great place to be. So yeah, I do think that you should be very skeptical of AI as a catalyst to return Salesforce to 20% plus growth rates. I don't buy that for a second. And I don't think Benioff is arguing that.

What I think he's arguing is we have never been more important. And if he's right about that, and they're sticky, and they can keep disciplined and keep solid margins and grow them just a little bit, just a little bit every year, half a percent, 1%, just a little bit better, a little bit more, they are going to be a cash machine for years. That'll be something interesting. I'm looking forward to seeing how the street re-rates this stock. If they just keep crushing it such that the free cash flow yield just makes it look cheap, then I may be interested in buying some more. I already have a lot of Salesforce stock. Yeah, that works out well for us.

Yeah, I'm okay with that. We were chatting yesterday afternoon about this a little bit. And you said that 20 years ago, if someone had said Salesforce would become a player in enterprise software, most people would have laughed at that comment. But that's what a lot of people did. Yeah, not just would have. They did. They did.

It happened. It did happen. Like spoiler alert, that's what Salesforce has become a player in enterprise software. And that's maybe even an understatement, right? enterprise software player once so difficult to believe because it required a paradigm shift you had to move from this process of buying software and you would buy buy the software itself you would own the code you would own all of the upgrades you would install it yourself you you would maintain it yourself. You had poor souls like my friend Tim going into work on

weekends to do things like major upgrades because the network had to be taken down, and then the new upgrade would be put in place. That was a nightmare. That used to happen for IT people to be like, okay, everybody out by 3 p.m. on whatever day, and then the network's going to be taken down. All these things are going to happen so that you could stand up the new software or the new network upgrade or whatever it is. You had to believe that that paradigm that had existed for such a long period of time would be replaced by, you know what? You don't have to do any of that. You can just rent the software from us. We're going to deliver it over the internet, all the upgrades, they're going to be handled on our end. If you have Salesforce,

you don't even see the upgrade. You might see a new field, but you don't experience the upgrade. You just log into your system here. You had to believe that that was something that people would trust. A lot of people said, that's insane. Maybe there's a couple of small software apps that can be delivered via the cloud. But enterprise software, are you out of your mind? $400 billion plus worth of software sales are going to move to a totally different paradigm? You have lost it, son." That was the argument. My argument at the time was that, well, it doesn't have to be all of it. Even if it's just a small portion of it, this is going to be a much bigger company. It turns out that that actually has become the new paradigm that

virtually all software of any meaning is delivered via the cloud now. But that 20 years ago, Mary, like a lot of people didn't believe that was going to happen. Tim, in preparation for this, I had asked you if you had anything that you really wanted to rant about. And you had said, not really. I don't know that I have anything, but I think we kind of got you on a rant about Salesforce and why it's maybe not as down and out as the street wants us to believe. I mean, algorithms throwing out the stock like babies chucking cash into a bonfire is not the meme I expected, but it feels like it fits. But it's the meme that we got. Thanks so much, Tim.

It's the meme that we got. Thanks so much, Tim, for diving into this with me today. Always appreciate having you on. Thanks, Mary. You just found out that your sales team is at risk of missing quota. Don't panic. Just ask Rippling AI. Since it's built on your real-time people and business data, Rippling AI can pull metrics from Rippling and Salesforce into a meeting-ready dashboard showing quota attainment, headcount

plan, and monthly revenue to quota by region. In seconds, you'll see exactly what's behind your quota risk and fix it before it's missed. Question answered, action taken, crisis averted. When you have critical business questions that need answers, don't just file a ticket and wait weeks for an outdated report. Describe what you need and have Rippling AI build it instantly from your live people and business data. Whether it's a dashboard with detailed charts or automated workflows with the right triggers, conditions, and approvals. Ready to rule your business? Head to rippling.ai slash fool to get the only AI built to give you full visibility and take complex actions across your entire organization.

That's R-I-P-P-L-I-N-G dot A-I slash F-O-O-L. Sign up for exclusive access today, rippling.ai slash fool. Last week, NVIDIA reported some stellar earnings. Again. Up next, Asit Sharma and Ricky Mulvey look at three companies that stand to benefit from the chipmaker's record results. Asit, last week when we were talking about NVIDIA, I had a question about the ripple effects on other companies. And then we just straight up ran out of time, but I still wanted to talk about it with you. There is this boom in data center spending and NVIDIA is not the

only beneficiary. And in fact, there may be some other large and small tech companies that are riding the coattails of this wave a little bit. The one company when we were slacking about this, that you, quote, def want to hit, end quote, is AMD in terms of this data center spending. So how does the boom, how does this data center spending boom affect AMD? Wait a minute, Ricky, before I jump in, I really take issue with you revealing what my normal speech patterns look like to a wide audience. I think it's lovely and personal. I def want to hit AMD.

I mean, this is a company that has a wide sort of play field, playground in the chip industry. It doesn't only make data center GPUs. It makes chips that we use in laptops and computers, field programmable gate arrays, FPGAs, et cetera. So they're really widespread when you think about computing, both in data centers and just in everyday life. But AMD, yeah, is going to have a spillover effect from all of this investment in data centers. It has a competing product, a very advanced GPU, that goes up against NVIDIA's omnipresent, all-powerful, great GPU stacks. And that's starting to get a little bit of traction in the marketplace. They're only doing about $4 billion a year in GPU data center business, AMD.

And by comparison, NVIDIA, as we all know, sells in the tens of billions of GPUs each quarter. So while it is a distant number two, it's starting to see a little bit more of business from just the amount of demand that's out there. So I think what's good for NVIDIA remains good for AMD. This is where AMD still has a strong growth story in spending for data center GPUs, even though it's a smaller part of their business. So let's say I have a large language model, Asit, and I want to run it somewhere. Why might I put that on AMD servers instead of NVIDIA?

Is there a quality difference or is this a pricing game? It's somewhat of a confluence of factors, Ricky, and you hit on one. The AMD processors are a bit cheaper right now than NVIDIA. NVIDIA can price as it wants. Also, they're accessible. if nvidia gpus are in such demand in the cloud provider that that uri is going to use let's say that we're a startup that needs to develop our own llm and train it right now you might have depending on who you're using a bit of a wait to even get to the specific gpu setup that you want

with nvidia so there's that and i would say they're comparable there are some startups now who are trying to test the benchmarks of the NVIDIA GPU stack versus AMD. And really the message is, sure, NVIDIA on most benchmarks is a little bit faster than AMD. And the next generation that you and I talked about last week is going to be even faster. But for a comparable price for something that's available, really you don't have a ton of difference if you're working to train a large language model. And also for the inference part, when people start interacting with that model. So you've got an alternative that people want to use.

You've got one that is being promoted by various cloud providers like Microsoft. So the implication for AMD is that it is going to be able to lean a bit into this demand. And we should see that $4 billion annual run rate with its GPUs start to expand in the next couple of years significantly. So you're telling me that AMD is store brand GPUs. We don't need to go out for fancy GPUs. We have GPUs at home with AMD. We want the expensive GPUs, precious, but we can't reach them just now.

So there is an alternative out there. Let me throw in this quick example, which has seemingly on the surface, Ricky, nothing to do with AMD, but it has everything to do with AMD. We just heard that XAI, so Elon Musk's AI startup using a chat bot called Grok, G-R-O-K, not to be confused with a large language model called GROQ, they're going to purchase, this startup that Elon is backing, is going to purchase up to 100,000 of NVIDIA's H100 GPUs, so billions of dollars of the H100.

Well, that's an older technology now, right? NVIDIA's moved on to the H200. It's got the Blackwell architecture coming out. Why is this startup backed by Elon Musk investing in this older technology? It comes back down to this confluence of factors I was talking about. Price, they're cheaper than the next generation. Availability, they're here now. They're comparable to the future generations of NVIDIA's own product.

So if Elon is starting to think along these lines, you can just picture drilling down to small startups who can take that to the next level of logical reasoning and go ahead with the AMD processing stack. Perhaps the comparison to store brand ice cream is a little mean, especially when the products that AMD are making are good enough for Microsoft, that multi-trillion dollar tech giant. What's going on between AMD and Microsoft, who does have a lot of partnerships with different semiconductor companies? Sure. Well, AMD was actually backed by Microsoft early on in their race to catch up with NVIDIA. This is something that I think many investors get a little confused

by. Why would Microsoft even bother with AMD since they're so deeply intertwined with NVIDIA's platform? The reason is that they want an alternate source of supply, as does Oracle, as does Amazon. All these cloud titans don't want to be dependent on NVIDIA at the end of the day. They like the business. NVIDIA is helping them realize billions in revenue, but they don't want to be beholden to this one provider. And that's also the reason, of course, why they're all developing their own chips and platforms.

So, AMD sort of plays right into that with their MI300X, their GPU accelerator. And Microsoft is sort of looking at the prevailing winds and understanding the same factors are at play. They're starting to reach some capacity limits in what they can offer to customers. So they almost have to move forward with this little relationship they started with AMD for cloud-centered GPUs. And it sort of makes sense when you think about those dynamics of demand and supply. Let's talk about Micron, which is a name that may be a little bit less familiar, but they

do memory and data storage chips, basically. So a lot of spending on data centers. Why is Micron feeling the tailwind of all of this spending? So Micron technology has been around for a while. They're one of the largest memory providers in the world. And Ricky, as you pointed out before the show, they're sort of like a commodity business, not a terribly exciting business, but they offer a very interesting memory component. It's called the HBM3E.

So this is high bandwidth memory. What's so interesting about this component is that it works well with a certain type of chip architecture used by both NVIDIA and AMD. When these companies build their GPUs, they no longer have these single-die components. They're stacking a lot of components. They're stitching together memory, compute, into one really dense package on a slab of silicon. They're building Lego-like structures to make all this function with the densities of these products. You need memory that can interact with the different components on these dies. And Micron has these components that can also be stacked and configure right into what NVIDIA and AMD are building.

So the HBM3e is the latest iteration of this memory module. It's going to be part of NVIDIA's H200 TensorCore GPU. So that second generation of GPUs that we were talking about last week. And I think it's in production now. It's going to be available a little bit later this year. It may already be finding its way into some components. So this is a tailwind for Micron. And I think the demand that we're seeing in data centers shows that it's a new part of

the market the company can play in. The product management has about 30% less power than other offerings, and there really aren't many offerings that compete with it. So, stretching beyond 2024 and 2025, where they are capacity-constrained, I think you'll see here a chance for Micron to have this higher-margin, higher-volume product that makes them a little bit less of a commodity business going forward. As always, people on the program may have interest in the stocks they talk about, and The Motley Fool may have formal recommendations for or against,

so don't buy or sell stocks based solely on what you hear. I'm Mary Long. Thanks for listening. We'll see you tomorrow. Thank you.