Morning, everybody. Welcome to day one of Citi's tech conference. My name's Tyler Radke. I co-head the U.S. software sector here. We're happy to have SAP's CFO, Dominik Asam, here, and you might be wondering, what is the U.S. software analyst doing up here hosting SAP, but my colleague Andrew, unfortunately, could not make it over from Europe, so I gladly stepped in his place. We're very happy to have SAP. I think it's been about six or seven years since I last helped cover SAP, so hopefully know enough to be dangerous, but really looking forward to this conversation, so Dominik, thanks so much for joining us.
Thanks for having me.
Before we get started, I'm just going to read the Safe Harbor statement. Please note that except for certain information, matters discussed during today's presentation may contain forward-looking statements, which are subject to various risks and uncertainties that could cause actual results to differ materially from expectations. The factors that could affect SAP's future financial results are discussed more fully in SAP's most recent filings with the SEC. Okay, I'm glad I got that out of the way. So Dominik, let's start off. You joined SAP around six months ago. You've already gone through the process of updating some of the midterm financial targets. Could you just talk about your experience? You're the CFO of a software company. After leading Airbus and Infineon, which aspects of SAP's business impressed you the most? And where are you going to be more focused going forward?
Yeah, I mean, the fact that I actually joined SAP was due to the fact that I had a deep conviction for quite some time that the SAP offering can actually deploy a lot of power and bring a lot of value to the end customer. And I think it's really fascinating to see how that equation is actually moving to become even more favorable. Why? Because on the one hand, there is the exponential power of that digital technology becoming visible to the corporates, some of them being more conservative and cynical about things you can do there. But with really the very dramatic use cases that have been demonstrated with things like large language models, everyone is waking up to the reality that there might be massive productivity gains available.
So the openness of this corporate customer base to really discuss measures to leverage that technology is increasing strongly. And then, of course, there is the question about what is the data that is most relevant in these companies. And I firmly believe that the type of data which is hosted, so to speak, in the SAP systems is extremely valuable in that context. And there's also a stronger focus on questions like smart data management and integration, and integration not only within the kind of SAP solution suite, but also how can these SAP solutions be integrated with other sources so then you can run use cases on top of all of that. So the velocity is huge.
I think the conviction that this has to be done in the cloud, because it's impossible to deal with that velocity of innovation in an on-prem world where upgrades take half a year or longer, is very obvious, so I think since I joined six months ago, starting from that, things have actually very much moved in the favor of the company, I would say.
Yeah, yeah. Awesome. Well, so maybe give us your view on what you're seeing from customers. Obviously, it's a challenging macro environment out there. IT spending is constrained. But just what are you seeing in terms of demand signals over the last two to three months where sales cycles are still a bit longer than what we've typically seen?
Yeah, I think there's two types of business we have, of course. It's the subscription-based business, more recurring in nature. And then there's the transactional business, which is more dependent on macro, I'd say. You know, we have applications like Fieldglass, which are correlated with contingent workforce numbers. And not surprisingly, that kind of transactional business has been affected somewhat by the macro headwinds. On the other hand, the fundamental move to the cloud and the interest of the customers to really future-proof their businesses is unabated. And we see that also in the numbers. I mean, our cloud revenues have been growing by 22% in Q2, despite that transactional sluggishness, because while it's not a subscription business, it's still a kind of cloud business. And it's included in cloud revenue. So it's diluting the cloud revenues and still 22%.
Even more importantly, the forward-looking indicator, the current cloud backlog, has actually remained in kind of mid-20s, 25%. We also see a decent outlook for the second half of the year. This is why we're not too concerned about it. We also see that the products that are, from our perspective, the best indicator of future growth are growing best. I mean, if you look at the growth rates within the different subsegments of the cloud, i.e., SaaS, PaaS, Infrastructure as a Service, if you back out the Infrastructure as a Service business, which by intention we are shrinking, we are seeing a 26% growth of the SaaS and the PaaS business. Very encouragingly, the PaaS business is growing much faster. This is, from our perspective, a leading indicator, because people, once they have installed the platform, they tend to really leverage it.
And we can use it for cross-sell, upsell, for making the sale of partner applications on that platform easier for customers. So it's really going everything in the right direction. So the most important target for me personally is really to make sure that we deliver to that 25 outlook. And I think with these numbers, we have a good trajectory on that. Also, if you really want to be a little bit granular, the first half of the year, in terms of cloud revenue growth, was a little bit negatively affected by some basis effects, because last year we still had Russia and Ukraine in there, which went out. And we divested a business called Litmos. So that's probably 1.5 percentage points. So actually, that 22% cloud revenue growth is somewhat understating the real underlying cloud revenue growth.
So we feel pretty good about the real clean underlying cloud revenue growth with the PaaS really pushing hard here.
Yeah. So it sounds like a lot of stability. You're still chugging along towards the financial targets. I guess when you talk to customers who, I imagine, other vendors aren't exactly experiencing the same, right? What is it that's giving you that resiliency and that stability? Obviously, the ERP business is incredibly sticky. But why is SAP such a high priority, even on your new businesses, which are growing quickly?
I think it's really what I alluded to very shortly before, which is that all these management teams and their board of directors are looking at a very public discussion now about what's possible with digitalization, and they think about how can we future-proof our business and what are the key ingredients for that. And productivity becomes more important than ever, because it's not an easy environment. We've seen a lot of inflation, labor shortages to some degree, so it's all about productivity. We are in a brutal B2B business. Our customers don't care about blah, blah. They care about real value for them, and I think a lot of stuff has also been demystified, and it's really waking up to that reality. The data which we have in an S/4 ERP system is extremely valuable to exploit when we leverage these new digital opportunities. We need to drive productivity.
We don't want to hire people anymore at these high costs, which are actually accelerating. I mean, inflation is accelerating on labor cost. Hopefully, it's now slowing down. But in Europe, still, it's very high. So I think there's a lot of ingredients we come together. So the key task we have on the go-to-market side right now is to convince the customer base that the kind of SAP journey is the core pillar, or at least one of the core pillars in the artificial intelligence strategy. And we are quite successful at bringing that point across by virtue of the technologies we have built by now.
Yeah. So talking about margins for a moment, as we think about all the puts and takes, right? You had high inflation. You had some softer transactional revenue. How are you thinking about margins heading into the second half? As you think about product mix, headcount changes, just walk us through the range of outcomes.
Yes. I mean, we had an extremely strong year-on-year growth in non-IFRS operating profit, especially in Q2. Already in Q1, we exceeded that famous 10% threshold. You know, we had a little bit of a valley of tears, because since October 2020, the company invested heavily to really provide that platform, that credible platform that customers can build their roadmaps around. And now we start, we said in 2023, we want to see the non-IFRS operating profit go back to a double-digit growth, which is 10%. And we have exceeded that in Q1. I said we had constant currencies, 28% in Q2. But I have to be honest, there was a little bit of a kind of easy comparable effect. In the first half, for instance, we had quite some charges because of Russia-Ukraine in the first half of last year.
And the second half, actually, some of these were reverted. So the comps become much more difficult. But still, with the update on the guidance we've given on the earnings where we slightly, very slightly added, we are now staying on that kind of 10% growth. You mentioned also the consolidation of our cloud infrastructure. That program is largely completed. So we are now through that journey. So don't expect wild improvements there. The lion's share of the improvements has been already materialized in Q2. On the restructuring side, we made good progress. I have to say, though, that we had a kind of high-quality problem, which is that our attrition was extremely low. I mean, normally, we see some attrition, which we plan for in our yearly planning. Now, with all the competitors also becoming quite strict on hiring, we actually didn't see the attrition we've been seeing.
So this is where there was a very slight roll on of headcount. And in the second half of the year, you might see some investments in core capabilities like artificial intelligence, but we will be extremely restrictive on others and try to contain that cost growth. But to cut a long story short, the guidance is robust, we think, for 2023 on non-IFRS operating profit. And I think the midpoint of the guidance is staying on the 10%. So I feel quite good about that today.
Yeah. And if you could talk a little bit about pricing, just how much that's had an impact this year versus prior years? And then how are you thinking about the ability for pricing to contribute to SAP's growth this year and beyond?
I mean, in prior years, so before the whole inflation story was really bubbling up, SAP has been quite accommodating in terms of not charging inflation, despite some contractual rights to charge inflation, so it was almost like a gentlemen's agreement, and then we had to, of course, change that, given that we really came to high inflation levels, we've done that for the first time, 1st of January 2023, where all support was put on 3.3% inflation, and now we have just announced another round of hikes on on-prem support to the tune of 5% max, so that's the cap, so we charge basically CPI increase, but cap it at 5%, and depending on the country, depending on the respective inflation level, we can get there, so that is, of course, giving us some top-line stability, but let's also face it, we have also significant cost increase on inflation.
If I just look at Germany, I think the compound inflation index, CPI, since early 2000 was maybe 16%. If you compound 3.3 plus 5, it's way less than that. But we also get some productivity. So I'd say it has been adding more to revenues than to margin, because some of that kind of revenue increase has been absorbed in higher cost. But we will work very hard on the cost side going forward. So I really spend a lot of my time today to what I call de-risking the 2025 non-IFRS operating guidance. And what it really means is trying to go beyond in terms of profitability to make sure that if any other topics happen, that we are really secured with that. That's really the aspiration of the entire organization. And I think we have quite some leverage.
And also being slightly more rigorous, but still in a partnership level, but more rigorous on the inflation side. And also, of course, being also reasonable in terms of competitively. We also have to look at what type of CPI increases our competitors are charging. That will help us to some degree.
I see. Okay. One of the other big growth drivers that stood out is the Business Technology Platform. Could you just talk about how you're thinking about the attach rates and how customers are actually consuming?
Yeah. I mean, when we talk about the more complicated customers and larger enterprises, they go on what we call the RISE journey, which is a private edition of our cloud offering, and the attach rate there is extremely high, so it's 80% or so, and these customers, once they have the so-called BTP Business Technology Platform, they tend to also be easier to cross-sell, and it's much easier to cross-sell on that platform in the cloud than to move an application of SAP the customer is currently not using on-prem. It's just a much more intuitive, low-cost incremental move with a lot of benefits for the customer. It's an easier sale, so that has been working very, very effectively.
I can only encourage you to look back at the slides we've shown at Sapphire, where we showed different vintages of the kind of wins on the cloud and how we have been able to kind of upsell on that customer base quite successfully. So that is an extremely powerful tool. And maybe for those of you who are not so familiar with what it is, the logic is the following. I mean, on the one hand, there is an optimum process, hopefully the SAP process, because we spend day and night to create processes that are really and cast them into IT that are really efficient, make sense. And then there is an infrastructure footprint of the customer. And they don't always look the same. So you can do two things. Either the customer is customizing his software to fit his footprint.
This is what had happened a lot on the on-prem side. And of course, all the system integrators were very happy because they could bill endless hours to the customers. So on these transitions, basically 20% was maybe spent on the software and 80% on system integration. And now we ask the customer and say, look, customer, maybe that's not efficient anymore, because if you want to be at a high velocity of innovation and want to use all these AI use cases, we have to do it in the cloud. There's no choice. So you have to move that way. And the BTP tries to do the following. It tries to say, we are creating what we call a clean core.
So a minimum set of standards in the system, which is really depolluted of your customizations over time while you are still in the private cloud, because the ultimate goal for you is also to go to the public cloud. When they don't go to the public cloud, they don't do it because they don't want to. They don't do it because it's just too much invest now. But they understand sooner or later they will have to take that move. And we have a lot of customers really already kind of doing zebra strategies where they pick certain instances and say, we bring that to public cloud to really start experimenting with this, even in large industrial companies. On large professional services firms like a PwC, they go straight to public cloud because it's an easier business.
So BTP is basically the platform to partition the clean core versus extensions. And then these extensions can be either customized stuff for the customer, or it can be add-ons programmed by system integrators. And then the system integrator would be able to sell that not only to that customer, but kind of really do a business with us and say, this is a module I can sell to 10, 20 customers. And so we create an ecosystem that is composed of a clean core, which we upgrade rapidly, where you have a high speed of innovation, and then upward compatible extensions, basically. So this is why it's so crucial. And this is why I'm also so fascinated about the company, because back in 2020, October 2020, when management said we need to change strategy, that was simply a dream and not something that you could do.
Now we have it. We have invested heavily in these really basic things, and now we're starting to see the benefit of that.
I gotcha. Yeah. And I think at Sapphire, you shared some math about the mix of PaaS, mix of PaaS within SaaS growing from 20 to 50. How are you thinking about the contribution from also new customers in that segment? Because obviously, it's huge migration opportunity and conversion opportunities you talked about. But how do new customers play into that?
I mean, new customers, where we have been historically not very strong, was in smaller customers and simpler businesses. We loved complexity because of the ability of the scalability of the product and the ecosystem that was very much geared towards customization for every customer need, no matter how complex. And now we also offer these public cloud solutions. And we've created an initiative, a motion called GROW, where we basically target smaller customers with simpler business models. And that is bearing nice fruit. So we see a high % of net new customers also in the mix. These are smaller ones. So while it might be 50% in the kind of number of customers we add, it's not as high yet on the revenue side. But it's, of course, another increment. So I think about how do we solidify our cloud revenue growth trajectory.
It's that kind of increment that we are tapping into in a more effective way. Plus, of course, converting that still huge legacy customer base on-prem into the cloud. There's still a lot of people that need to do that. Remember, there is the 2027 cliff on our support, and they need to find a solution by then, and the only solution we can offer and support is the solution to go on a RISE journey, so we think we have double support there, kind of converting the existing customer base plus increments. It is quite difficult to eradicate an existing ERP system anywhere, so there's often a debate about Oracle versus SAP and who is stealing what customers. I would still, of course, claim that we steal more from them than the other way around.
But I have to admit, it's difficult because there's a lot of cost associated with the customer to change that. Where you have that sometimes is when companies merge. And then there's a lot of benefit in the customer consolidating from kind of a Zebra strategy into something unique, unified. And I think we can say that we have the upper hand on that. So I think this is what gives us the confidence when we say, well, 25 is an interesting data point to look at, but the real fun starts later. We think we can sustain the cloud revenue growth. And as the software part is shrinking more and more, the headwind from the shrinking software part becomes smaller and smaller. That means mathematically, we should see an acceleration there.
Yeah. You mentioned Oracle, and as somebody that covers Oracle, would love to just get your perspective more broadly on the competitive market within ERP. Obviously, they like to mention you on their earnings calls more than you like to mention them, but even beyond Oracle, there's a lot of private ERP vendors like Infor. I guess, are you seeing any meaningful share shifts now that you've really kind of reinvigorated this migration to the cloud?
I dare say that we are gaining share because simply of the math, if I look at the kind of pace and SaaS offering of ours and the growth rate we achieve compared to competition, we do quite well. Yes, on-prem, we have a certain decline, but it's actually slower than what we thought, so I think, yes, just back of the envelope, it's clear that we are doing quite well on that front, but of course, the competitive pressure is huge, and I think disruptive technologies like AI are a fabulous opportunity for competitors. If they're leveraging that smarter than we do, they can kind of cause market share losses.
And that's exactly what we need to focus on now to really be humble, fight for every customer, and make sure that we are really, really paranoid about the technology, invest sufficiently in research and development to make sure that we keep our technology edge. And we have not always been the fastest, but I think once we woke up, as you've also seen with the cloud transition, we've done it in a very kind of Germanic structured and effective way. And so from that perspective, while I take our competition extremely seriously, and honestly, we have made mistakes in the past where we've left a feast on the table for competitors where actually, if we had acted more smartly and faster, it should not have happened.
So we've learned the lesson that this is a kind of a shark pool, and you have to be super fast and bite faster than the others. But I think the data is showing that we are on a very good trajectory right now.
Yeah. And as you think about the broader portfolio, certainly there's also opportunity in the HR side, SuccessFactors, and the CRM side, kind of more front office. Certainly, there's market leaders in those categories in terms of SaaS market share. How do you feel about SAP's competitive positioning relative to, say, a Workday or Salesforce?
It's a very interesting discussion. I think every vendor first would like to do everything himself or herself and say, look, I want to be a captive platform and integrate everything. But honestly, you have to realize this is not possible anymore. So you have to be very granular and say, what are the things where I have to federate with other people? So take our Datasphere platform. We basically federate our ERP data with other vendors, analytics vendors, and so forth. And we have to make sure that the contracts we use in these circumstances are structured in a way that we basically get our fair pound of flesh and the monetization on what we can bring on the table. And what we see in these types of discussions is that we're an extremely attractive partner, also for people who do have a different footprint.
They want our data. They want to see and talk to SAP and say, how can I federate my data with yours so that the customer has the least effort in data management possible to leverage the data? So I think we are really sitting on a treasure of data. And we can talk to all these vendors. And we have to decide, where do I keep the cards close to my chest? Because I think, actually, I want to do that business myself. And I think we are close enough to the best-in-breed offer to really insist on that being SAP. Where do I kind of find a relationship, a good partner where we're complementary, where we can offer a solution to the customer to federate? And this is then decided case by case.
But coming back to your initial question, when I look at the type of dialogues we can have with some really leading innovators in the space, I'm quite impressed about how much people appreciate that data that we have in our system is really key to so many use cases. The breadth is important. I think with AI, extracting value from what I call connecting the dots in different fields becomes even more tangible and more interesting. Think about customer-centric data versus supply chain data. I mean, we've all seen in the kind of crisis of the supply chain that good management from kind of projecting customer needs versus ordering long lead time items five layers down the supply chain, that's a pretty complicated job where you can leverage AI. I always bring a very simple other example, HR versus finance.
I'm always frustrated about budgeting, forecasting discussions where the guys who've under-spent in the first half tell me second half they spend like hell. And I know that they have not even the kind of the job requisitions out to even hire these people. And I could, of course, with predictive capabilities, easily demonstrate to my people, look, between HR and finance, the machine forecast is simply better. In every forecast, there's a big negotiation. How much time do I dilute or lose in finance running these negotiations? So then there's the forensic stuff. I mean, assurance becomes more and more important. We have so much regulatory stuff flying in our face where we have to deliver to the assurance expectations of our stakeholders, shareholders in particular. And in AI, you have super use cases across different verticals where you bring data together and make fraud visible.
All these use cases show the value of integration of different fields.
Yeah. So you talked a little bit about it on the regulatory side, but sustainability has been a huge topic, especially among your customer base, which has a lot of physical goods and waste and emissions. How is SAP monetizing or approaching that with conversations? And then how should investors track kind of that progress or contribution?
I'm personally deeply convinced it's a tremendous opportunity for SAP for several reasons. First, you have to ask yourself, how important is that market, and right now, I know it's a little bit out of vogue in the United States, I mean, but I'm deeply convinced that the price for every ton of CO2 will go up to the degree that it will actually pass the price of the fuel itself, and if you make that assumption, so you just think, whenever I burn a ton of kerosene or fuel or whatever, so and so many tons of CO2 are emitted, and then you have to ask yourself about what's the price for that, and you can calculate that the technologies you currently deploy to abate CO2 are already more expensive than the fuel itself, and that means that market will be bigger than the oil market.
And if you add all the market caps of the oil majors together, it's a pretty significant number. Now, will stakeholders tolerate that you're estimating what you have there? I mean, with oil, it's not a problem. There's a market price. So you don't need to account for that because it's in your P&L, basically. But the problem is there's no homogeneous market price for CO2. It's kind of depending on regulation. So the workaround is that you count the tons of CO2 to make it comparable on a worldwide basis. And we are deeply convinced that when you can account for VAT and euros and dollars, and you have a lot of money at risk, you should also account for CO2. And I think there's a very special animal within the ESG family because it's so big in terms of value. And the risk of fraud is so huge.
The risk for assurance is so high. We have a deep supply network from Ariba where we can really offer the transfer of data between the different parts of the food chain. The double, triple, quadruple counting in this world, I mean, I'm basically, when I delivered aircraft and Airbuses, I was accounting for all the emissions of the aircraft for the next 20 years or plus. The oil majors have accounted for the kerosene. The engine manufacturers account for it again. You could create really a system where you eliminate these effects and come to a really clean system where you know exactly what ton of CO2 is coming where and what's the price of that and so forth. I'm deeply convinced that this will happen.
It might take a little bit longer, and regulatory things will play a role, and that SAP is super well positioned. It's early days. Right now, it's more on an estimate basis that people can still survive. I think the E.U. regulation is 2028. It should be reasonable assurance, which means a kind of audit process like for financial data. And we are currently heavily investing, but you will not see that move the needle a lot this year or next year because it's really an emerging business. But I'm super excited about it because it's a combination of, A, a very visible market that I cannot imagine will not be created, and, B, our set of capability, which is simply we need to integrate a new field in our data model and account for the CO2 through the supply chain the same way we account for euros.
This is why I'm excited about it.
And I guess, would that be a separate product, or would that just kind of be built into?
We actually integrate in our cloud offering. By the way, we say we're not going to do it on-prem. We're not wasting a single dime because it's a future business, and the future is cloud. We're not investing anything in making that accessible on-prem. So we are really focusing that. And it will be in a bundle because it doesn't make sense to do it as a pure kind of, I mean, there are intrinsic solutions which can be sold separately. But the real mid to long-term strategy is clearly an integrated ERP, full integration ERP.
Sticking on the cloud theme, I guess, what would be the factors that could cause cloud growth to exceed the 25% that I think you talked about? What should investors keep in mind for that upside case?
I mean, what we guided for 2025 is actually, if you take the guidance for this year and you guide it for 2025, you see it's like 8% growth, something like 24% remaining to do. What could exceed is an acceleration on a kind of move of our installed on-prem customer base to the cloud when they kind of all then need to come 2027. It is really, I think, AI itself. I mean, the AI, we think that AI will give us an opportunity to add significantly to the enterprise application market. So the enterprise application market growth should accelerate. Why? Because enterprise applications are a B2B business where it's all about value creation, and if you can push productivity like hell, if you are the enabler of that productivity, you should get a fair share of that.
And so we have said that the market enterprise application market could double come 2027. And that is basically, I'd say, already now without that kind of booster, you would see a 60%-70% growth over the time frame, but that's the header. Now, we've also talked about in our premium bundles where we are going to embark some AI. We think that these premium bundles ultimately will carry maybe a net of about 30%. I have to say, though, that this kind of header is kind of a compound of two things. You will need to do a broader product sale to the customer because if they want to leverage that AI tool, they need several applications. So it's also a little bit of a cross-sale mix in there. But if you, yeah.
I'd say for the guidance for 2025, to secure that cloud revenue and maybe create upside, it's all about how successful we as SAP are that the kind of pre-invest every corporate will need to do to future-proof for AI. They will say, I need SAP as the first thing to do that. There's many other things I can invest in AI. But if I don't have the data management in the right way, if I don't have the contextual awareness, all the technology, I cannot buy the technology. I mean, technology itself is not worth much to the customer. And their ability today to consume that raw technology is quite limited. So even we were kind of toying around right now with seven different large language models to figure out which one are we using for what application. So we need the contextual awareness, the use case.
We need the technology, compare the cheapest available technology for that problem. Cloud compliance is a huge problem. So in every country, we have different rules. Infrastructure needs to be cheap. So it's extremely difficult for our customers to consume the raw technology, to just buy a foundational model and say, we are going to train that ourselves or so. So they have to take these off-the-shelf solutions. And then you look at solutions like Copilot, Microsoft. You see that's why that's interesting because I have the same discussion now internally. Are we going to invest in that? And I'm actually a big fan of that because I say, look, we want to make our own staff. Our own staff needs to be kind of AI native.
They have to demonstrate that they themselves can leverage AI to become more productive, to be a role model for our customers. And see, this is what you can do. And of course, with our applications, you want to achieve the same. So I'm really deeply convinced that there is an acceleration because of much more value being created with these new tools. And it's complicated for every single customer. Maybe the very biggest ones can do it. But hiring the right talent, ensuring the compliance, data privacy, all this stuff is a pretty complicated game. And also just understanding what foundation models are out there and what are they exactly doing. Good luck if you are a kind of, I don't know, even if you're a big company and you have hundreds of millions IT budget, you will actually need billions of IT budget to do that.
So I guess just to put a finer point on the generative AI opportunity, do you envision releasing new product SKUs that you'll charge for that will be similar to maybe a Copilot? Or how do you kind of envision the packaging and monetization?
It will be more a package thing. And that package will include a certain consumption of AI. And then you will probably be able to add on by selling extra units if you use it more extensively. And in terms of the kind of three fundamental thoughts where the value lies not going into specific use cases, but the more generic things, it's, of course, still even smarter automation opportunities to drive productivity. So really, when you talk about foundation models, I always say the guy who has the best data should train. So if it's all about its process optimization, ERP, we should train probably our own foundation model. So really showing ways to the customer, how can he create a self-learning process optimization journey for himself or herself? That's a key focus point. Then there is the interaction with the system.
Frankly, SAP has not been always known to be the most user-friendly interface, and there's a whole ecosystem having exploited that to kind of sandwich itself between the user and SAP to say, well, this is clumsy, and we need something that's nicer, and of course, with all these large language models, I mean, I always jokingly say, I want to try to make my own job redundant, and that basically Christian Klein can talk to the machine and say, can call it still Dominik if he wants to, but give me that and that analysis, and then it's really popping out as opposed to having these guardian of the temple ERP experts who know where to pull and where to get the data from, it's quite complicated, so we can democratize the data, and the last thing is, of course, coding can become much more efficient.
That can be a productivity boost for us, and I'm currently in the operating plan, discussing how much productivity requirement I hang around the neck of our developers on that stuff, but also for our tools. I mean, I talked about all the nice benefits of BTP, but BTP is still a kind of bespoke art of very highly trained people, how to use it, how to create that clean core, and if you can create also more intuitive ways for our customers to improve and leverage our platforms, so it's more a kind of low-code, no-code approach. And they can intuitively describe what they're expecting from the system, and then they don't need hordes of super well-trained IT experts, but they need somewhat less, and simpler stuff can be dealt with automatically. These are kind of three avenues which I see as very promising.
We have also, of course, the assurance topic. I think forensics is a wonderful field for these tools.
Yeah. So it almost could help accelerate the migrations.
But let's look. I mean, we have committed to what we said for 2025. And my job is to kind of create as much upside with my colleagues on the top line, bring the cost as low as possible, cash conversion as high as possible to protect that kind of number as much as we can and deliver it with the highest degree of confidence. And let's see what we can do beyond. But I think the current trajectory is kind of fully confirming what we had in mind when we communicated with SAP.
Okay. Just the last topic I wanted to hit on is capital allocation. So I know you recently did announce a re-up of the buyback, I think, EUR 5 billion, but you still have a considerable amount of cash on the balance sheet. How are you thinking about the balance of returning back to the shareholders versus acquisitions? Obviously, there's a lot of interesting but expensive stuff in the generative AI front, so.
I mean, acquisitions. I mean, first and foremost, my experience is the most value-creating stuff is organic growth because you keep all the upside for yourself. If you buy a company, the selling shareholder tends to want to have a pound of flesh in terms of participating in your synergies and so forth. So I like organic growth, but there are limits. On the other hand, we do see a lot of opportunities where we can invest at returns higher than cost of capital. Now, M&A is a little bit hard to predict. I mean, as you say, or let me start differently on M&A. We are currently in a situation where we don't need to be desperate in the sense that we absolutely have to buy something at a crazy price because we have blatant portfolio gaps and we need to plug them. We are growing very fast.
We have a consistent, very logical portfolio, so we can certainly add with interesting acquisitions, but still the business case has to fly because if the business case doesn't fly, it's not creating value, but we are certainly exploring, so you will see us exploring opportunities again and again and then from time to time also doing deals. Now, I still hope that this leaves money, then we have, of course, the standard recurring dividend policy of returning 40% of the pre-tax profit back to shareholders. We also have some capital optimization in terms of we still have, to my taste, a little bit too large a bond portfolio, which creates negative carry between the cash we carry and the liabilities we have, and yeah, on shareholder returns with the EUR 5 billion, we have decided, you see, that share repurchase is not a taboo.
And to the contrary, we are doing it. But let's not kind of now jump the gun and say we're going to do another one because we have not even executed that. It's actually it has been started and it's starting at a good pace. But it's clear if we're not going to hoard cash. In that context, I want to mention that we got an upgrade recently from S&P, which was good. I mean, we don't really need a higher rating. But what I found positive about it, that they basically said you can have a higher rating or you can have you basically have a better business model today. So they said we can give you better ratios because we have a higher faith in the kind of robustness and resilience of your business model.
So it also, but on the other hand, also means I might not need to hoard so much cash anymore because the cash generation of the company itself is strong enough, even if we have to do an acquisition, that we can easily pay it down. And we have quite some headroom. So it's a high-quality position to be in. But we fully understand that just putting money in a bank account for nothing without any strategic thoughts behind forever is not very shareholder-friendly and that we should then return the money as we have decided recently on the €5 billion.
Yeah. I think you timed that perfectly. It hit zero rate as you finished that last thought. So thank you so much for coming.
Thanks for coming.
Thanks, everyone, for joining the keynote today.
Have a good one. Bye-bye.