Chief Executive Officer & Member of Executive Board, SAP
SAP CEO Christian Klein Talks Company's Q2 Earnings | Bloomberg Talks
SAP's stock climbed 6% in premarket trading in New York after the company reported strong revenue growth in its cloud business. The company's current cloud backlog grew 26% on a constant-currency basis to EUR22.93 billion in the second quarter, above analysts' forecasts of 24%. SAP cut its outlook for the year, citing a EUR100 million impact from the acquisition of data platforms Prior Labs and Dremio, and trimmed its guidance for non-IFRS operating profit. CEO Christian Klein joined The Opening Trade to talk with Bloomberg's Anna Edwards and Guy Johnson.
See omnystudio.com/listener (https://o...
About Christian Klein
Christian Klein, CEO of SAP, stated during a July 24 Bloomberg interview that the company was "very happy" with its second-quarter performance, citing accelerated cloud growth and noting that 90% of cloud deals included AI. He described AI as a "disruption to the software industry" but argued that SAP's value lies in its ability to provide business context and domain logic that large language models lack. Klein emphasized that SAP maintains an open platform with APIs and third-party agent integration, and said the company sees "not one winner" in the AI layer for ERP. Following the departure of chief product officer Muhammed Alam, Klein took over oversight of development, saying that "it's good to be close to development these days" as the company shifts from coding features to coding AI agents. On the July 23 earnings call, Klein said customers are realizing they need to "modernize" ERP landscapes alongside AI adoption, and noted that "a lot of customers" are choosing to do both with SAP. He discussed the company's AI agent hub and its new end-to-end user experience, Jewel Work, which has 4,000 internal users. Regarding pricing, Klein stated that "with AI... we can completely reset the price level" toward outcome-based pricing. He also said SAP had adjusted hiring plans and would "nowhere near" hire the number of people planned at the beginning of the year, citing increased productivity from AI tools.
Transcript (17 segments)
N
Narrator0:02
Bloomberg Audio Studios, podcasts, radio, news.
A
Anna0:07
Let us talk about some of those corporate earnings then. As mentioned, SAP's cloud revenue grew at a faster pace than analysts expected as customers move away from on premise licenses. Europe's largest software company is facing threats to its business from AI, well documented, of course, as it pushes clients to migrate accounts from locally installed programs to new products in the cloud. SAP's ADRs, their American Depository receipts jumped following the results and it looks as if we are expecting to see the stock pop a little higher at the start of trade this morning. Christian Klein joins us, CEO of SAP. Christian, very nice to have you with us. Lots to discuss. So, the good news in your numbers, cloud revenue, better than expected, your customers then migrating to the cloud, that continues to be a trend. What does that tell us about the underlying strengths of the business at this point, Christian?
C
Christian Klein0:59
Yeah, good morning Anna. Yeah, I mean we are very very happy with our quarter. I mean we have even seen an accelerated cloud growth compared to Q1 far ahead of our market and our competitors and even in the cloud 90% of the deals also had AI included. Yeah. So which shows that first the ERP is absolutely needed as this is the brain of every company and then the agents are built on top and that gives us a lot of confidence with such a strong half year one especially in such a volatile market environment.
A
Anna1:34
Okay, so the migration to the cloud continues, Christian, but in amongst all of that, we add on, and you've suggested this includes AI, but we add on the concerns around AI. SAP's share price down by some 38% year to date because of concerns about the so-called SAS apocalypse, what AI will do to companies such as yours. Do you have a much better understanding now of where the vulnerabilities lie for SAP?
C
Christian Klein1:57
Yeah, absolutely, Anna. I mean first of all it's great to see that also our customers are now turning towards SAP as they understand that the LLMs they don't understand business process they don't understand business data and also they don't understand governance. I mean you cannot just have agents doing your financial close not adhering to the governance of a company so that is all sitting in our ERP and with the launch of the autonomous enterprise you know our customers have seen that SAP can deliver accurate and also compliant behind AI and that of course also is reflected now in our Q2 numbers and also gives us a lot of confidence for the second half where the pipeline actually looks pretty strong.
G
Guy2:40
Christian, can I just pick up on that? Good morning. It's Guy. OpenAI is making waves again in this space this week with its new presence model which is causing not your direct competitors but people in the SAS space to see their share prices come under significant pressure. Now, in some ways, you could argue this software doesn't directly affect you, but in some ways you could argue further down the road it might. You've got your dual software. It's the integration kind of layer, the orchestration layer, which is quite a valuable space to occupy. And it's looking to target that kind of space. And I'm just wondering how worried you are to Anna's point about kind of understanding the risks about how big the risk could be of OpenAI coming in and controlling that kind of integration layer, that key interaction layer that we're dealing with here and I'm just wondering whether or not that is a threat that is emerging, a new threat that you see emerging right now.
C
Christian Klein3:38
Yeah, I mean absolutely Guy. I mean AI is a disruption to the software industry without any doubt. But look at SAP. I mean we are running the world's most mission-critical business processes, complex processes. And there is a lot of domain logic sitting in the ERP. And when you now look at our AI platform, you can choose any LLM you want. And by the way, not every frontier model is the best for every agent. You can also use open-source models where you have a much better price to outcome but then what you find on the SAP AI platform on the business AI platform is also the context layer so we are building our own AI foundation where we are building the ontology with our business data with our business processes because again the LLMs don't understand, are not trained on this logic and when the two worlds are coming together this is of course how SAP can really deliver accurate and also very compliant.
G
Guy4:35
But is there going to be a fight over that interaction layer where the two systems integrate? Is there going to be a fight there? Do you need to make sure that you control that point?
C
Christian Klein4:49
I mean we always said our customer's data is the customer's data and we have an open platform. We also include third party agents. There's no problem about that. We believe in the value of our platform because again this business context is only available on the SAP business AI platform and that's why we believe in the value of the platform. We don't want to lock customers into our platform. We want to have an open platform. We have APIs. We have MCP servers. We want to actually run business processes end to end with our autonomous agents. And that's why we believe in the value and there's definitely not a fight about locking customers into our platform.
A
Anna5:29
I wonder Christian, how concerned you are about costs and how much you're having to spend at this point to innovate and to keep up. I started this conversation talking about how revenues came in above expectation. Morgan Stanley suggesting that maybe the R&D costs were higher than they had anticipated. Are you concerned about how much you're having to spend to keep up in this race?
C
Christian Klein5:50
Definitely in the R&D cost you see a higher token consumption as we are using AI to code additional features but especially now additional agents but vice versa you also see a huge productivity increase of 30%. So the last 12 months was really about ramping up our R&D workforce, bringing in some top talents especially on the data scientist side. But from now on it's really about gaining the productivity also by applying AI in our R&D workforce.
A
Anna6:23
Yeah. And Christian on that point then you mentioned there the increased token cost but then the increased productivity. So what is the return on your investments in AI?
C
Christian Klein6:33
The good piece is Anna that we can switch models on our AI platform. Again we are not only using the frontier models we are always using the best model which actually provides the best price to outcome ratio. So that first of all makes sure that we have healthy cost margins going forward both for our external customer business but also applying AI internally at SAP. Second obviously we are also managing tokens as part of our cost budget. Yeah, you have a headcount budget, you have third party budget and you have tokens. So our managers in R&D in the go to market space, they have one budget and they have to manage that also according to the productivity assumptions we reflected in the budget.
G
Guy7:15
Christian, just kind of picking up on all this, do you get the feeling now that we are moving into a different phase? Who's ultimately going to win this battle? Do you think it's going to be Frontier Models? Do you think it's going to be the open-source models? Cost is becoming and you are one of many CEOs that we talked to about this and you're all giving the same answer. We are now managing our token budget. It is a cost we have to manage. Do you think that ultimately points us in the direction of open source models?
C
Christian Klein7:46
Guy, there is not one winner in the tech space. I mean again these LLMs will become more and more commoditized. They are super strong but there are a lot of new open-source models entering the market but now I feel that the value creation moves up from the infrastructure from the chip side into the application into the agentic AI layer and so you need both worlds but I feel now companies like SAP will benefit from the advancements technological on the infrastructure side on the chip side by building now this agentic AI layer for our customers so there will be not one winner, but there is also definitely not an LLM provider taking the whole market, especially not in the ERP market.
A
Anna8:35
Christian, you at the end of June, you announced another top level reorganization at the business, as a result of which you've got a bit closer to managing some of the AI challenges and opportunities the business has. So, your chief product officer, you've got some of his job. Is that sustainable as the CEO and what kind of insight has that given you?
C
Christian Klein8:57
Anna, first of all yes indeed. Muhammed Alam is leaving SAP, he's still around but he's leaving SAP for personal reasons. So I took over his responsibilities, we are in the transitioning period. I guess in this time right now having the oversight about our development organization is actually a good thing because you have to see we are moving from coding features for software into coding agents. So we have to change the operating model, the way how people work. We need to apply AI in all of these jobs. So it's good to be close to development these days. But over the next months and quarters we will definitely also find a successor so that I can solely focus then on my CEO role.