How SAP's CEO Is Remaking the European Tech Giant For The Age Of AI

0:00Every day, the people of Meta work to protect over 3 billion users around the world. We are responsible for so many products that so many people rely on. When you have that level of responsibility, you take every single decision very, very seriously. I'm Abhishek, and I work on risk and compliance at Meta. We are definitely ahead of the pack here. We invest a lot more. We really sweat over every single detail. Get the facts at meta.me. You're dogfooding, I imagine. Oh, yes, yes, yes. That favorite Silicon Valley phrase, test on yourself. We call it drink your own champagne, yeah, because it hopefully feels like champagne.

0:46This week on Bold Names, I'm talking to Christian Klein. He's the head of one of Europe's biggest tech companies, SAP. Christian rewrote the company's business playbook when he took over almost six years ago. He's going to talk to us about why it paid off and how he's working to future-proof the company now. From the Wall Street Journal, I'm Tim Higgins. This is Bold Names, where you'll hear from the leaders of the bold name companies featured in the Wall Street Journal. This week, we ask, how is one of Europe's biggest tech companies navigating the arrival of AI?

1:26Thank you for joining us here on Bold Names today. You took over SAP in 2020 as the solo CEO. One of the first things you did was pretty dramatic. You announced a complete change of the business model, and your share price dropped something like 20 % in that day. What was the plan? Yeah. I mean, I started my career here, Tim, 27 years ago. And some people are saying I'm more a child of SAP. I started here with actually even as for 15 years as an intern. And when I became the CEO of SAP, it was pretty clear to me that actually the ERP, which made SAP so successful for 50 years, it's not going to last in the future.

2:10Let me stop you there. ERP, that sounds like a very technical term for the layman out there. It is actually enterprise resource planning. So when you want to do finance, supply chain, payroll, HR, everything integrated, then you need to buy an ERP. But that ERP was not really helping to grow our business in the future. And that's why I knew that I have to make a radical change. The cloud is the future. That was really evident. But then you need to disrupt yourself. You need to disrupt the way how you develop a product. You need to disrupt the way how you service your customers, deliver the product, and even all the internal functions.

2:49It's a massive transformation. And I was convinced that it's the only way to make SAP mid - and long-term successful. And then you have to take the risk and you have to make sometimes in life these bold decisions. I mean, really, the key was just for the layman out there. You were selling software and clients were using that on their own hardware. and you were moving to a business model where the software would live in the cloud, kind of very much like what Microsoft did for its enterprise business, right? I mean, lots of companies were moving to the cloud. Today, it made sense. But back then, perhaps, there was some nervousness.

3:26This was in the heart of COVID era. Shares fell, as I said, 20 % that day. What was it like to go home that evening and see how investors were reacting? Yeah, I mean, I probably didn't twink a glass of water at home. I probably needed two or three glasses of wine. But then obviously also one day later, also all the managers, the leadership team of SAP wrote me a nice, I would say even love letter and say, Christian, it's the right thing to do. We believe in the strategy and we're gonna follow you. You know, when of course the share price was down, when we did this radical change, I mean, people, there were of course concerns, questions.

4:04Hey, what about the future of this company? Are we are going to make it? And in these times, you need to actually give people confidence. It doesn't help now to put even more pressure into the system. You need to be positive. You need to show a clear plan. You need to over-communicate. You know, literally, I was always there every month. We reported progress and see here, positive step, positive step. But what then also finding the right balance in different times than to push the bar to also show the people, hey, this is not an industry where we can just stay in our comfort zone. And then finally, what I say was very important is also the mindset of the people.

4:47When you are running a cloud business, it was always about customer satisfaction. Don't get me wrong. I mean, otherwise this company would not have been successful. But now it's really about adoption. It's not about selling a piece of software and the customer is implementing it. Now we are on the side of our customers with every step they go. And that was actually also a big shift in the way how we work with our customers. And that also, of course, required a lot of change management to really get this into the DNA of this company. You know, I don't know if, you know, looking at this at first blush, that you would necessarily be seen as a change agent.

5:24I think a common business school perspective is you need to bring an outsider in to shake up culture. But I know that you've talked about this idea that actually being an insider helped you. and I wonder why. I mean, when you look back into this transformation first, it's always helpful to have a strong network because in order to convince people, you need to understand how do the people work, how do the different functions of SAP work. Here in Germany, we have things called the Workers' Council. So you can do a lot of things, but you cannot do everything against the Workers' Council. Let me slow you down there for a second.

6:02Workers' council, for those not deeply into European labor practices, a workers' council is essentially union in the U.S., right? Exactly, exactly. And then you have the ecosystem with the millions of partners, and they also saw a big change of their business module. And you have to understand how does this all play together in order to make this transformation work. If you lose one of those stakeholders, be the employees firsthand, but then obviously the customers, the partners, the investors, and other parties like the unions, then you have a problem. And that's why I feel it was actually a big plus that I knew the company inside out.

6:42Am I saying that you always need to hire a CEO from inside the company? No. But at a special time like that, I would say it's rather beneficial.

6:56Speaking of culture, Christian is pivoting SAP yet again. to AI services. How is he guiding the company through this moment? If we are now not delivering a strong, agenda AI layer within our software who can run those business processes, then we have a problem. But we have this really, this big advantage of having the data and the business process and a good understanding of how industries run that. That's next.

7:30Harvard Business School Executive Education creates powerful connections for leaders from around the world. Their programs strengthen organizations and individuals by deepening relationships and fostering new ones. Participants leave with lifelong friends, new potential business partners, and a powerful, globe-spanning network of fellow changemakers. Learn more at hbs.me slash breakthrough. That's hbs.me slash breakthrough.

8:08Okay, so we've talked a lot about culture change here. And I'm interested in this because I think a lot of companies are facing this really big change ahead with the rise of AI. Some have to look internally to make that change. Some are looking externally. Ultimately, this probably means culture change for a lot of companies, right? What are you hearing from customers on what they see as the biggest opportunities for AI in 2026? In the B2B world, I mean, obviously what moves the needle is supply chain. I mean, you know, there when you look into manufacturing, inventory, logistics, I mean, companies spend a lot of money for production, for the inventory.

8:50and everything what you can really change from a business perspective with AI is actually resulting in huge, huge benefits in a lot of value. The second piece obviously is when you look at the top line, now with AI and with the business AI and with the LLMs, you can detect consumer trends even better. You can actually tailor your offerings even more personalized to the consumer's needs. And then these agents, when they talk to each other, they can actually deliver you the best consumer experience because they know exactly what is on the inventory, what can be shipped by when. And that is, of course, when you think through how SAP did run such processes for over 50 years.

9:32I mean, this will fundamentally change because with these agents, you can run that more tailored, more personalized, more intelligent. Yeah, when you say the word agents, you mean AI agents, This idea of essentially AI programs designed to do specific tasks for the consumer or the client or whoever, right? Yes, exactly. Or the employee. As you know, I'm based in San Francisco, kind of the heart of a lot of conversations around AI, startup AIs in particular. And I feel like I've had a lot of meetings in the last few months with AI startups who think they have the next big idea to revolutionize those boring functions within business, right?

10:15Really the things that your company is known for doing. And I presume that you think you're probably better positioned to implement than they are. Why is that the case? because history would suggest that green shoot companies can make bigger changes, can move faster, that sort of thing, and established companies struggle. But I think you probably have an argument for why you're better positioned. Yeah. I mean, Tim, when you look into what SAP does, I mean, we are running the world's most complex business processes. And to understand the business process, you need the logic. You need the data module.

10:52You need to also understand who is allowed to do what. I mean, not everyone in the company is allowed to run every transaction or to see every number. So that business logic sits within SAP. And I absolutely believe that all of the companies, the startups you are talking to, can build an agent on top of such a system. But this agent is really lacking the context. It's lacking the semantical data context. So all the things what sits in the software, in the business process need to be understood. And there's only one tech company who wants this mission critical processes end to end, and that's SAP now.

11:30But also, to be honest, Tim, if we are now not delivering a strong, authentic AI layer within our software who can run those business processes, then we have a problem. But we have this really this big advantage of having the data and the business process and a good understanding of how industries run that so that our agents, when they really have to work together seamlessly across those business processes, we definitely believe they know the most, they are the smartest, and they can actually really interact really well across actually the whole value chain of a company. What are some of the tangible things that you're seeing right now that your clients, your customers are able to do with AI and your system?

12:13You're sitting on a trove of huge business data, right? You have these companies' data in your systems. Now that it's on the cloud, in theory, the AI agents can better access that data. Where are we at? Let me share with you, Tim, just two concrete proposals, two use cases, which we just showed two of our customers last week. First one, a retailer. You can shop commerce online or you can go into a store. In the future, when you go into, be it a store or the online shop, actually you're gonna get a very tailored offering based on okay what did you buy in the past is it raining today is the sun shining is really making you know probably a difference on what you want to buy today for your outfit the third part is then obviously okay once you actually then have the perfect offering for your consumer what is actually now you're running out of stock in the store okay which warehouse is now closest how can i make it work that when you arrive at home Ideally, we can already deliver that product, that stuff.

13:17And so that is something, you know, where we can now say, hey, there are smart agents who can actually tailor this for you for the consumer when they're entering a store, plus, you know, really arrange the whole logistics, the whole inventory and run that. And then another one is cash flow. Why are certain consumers not paying? I mean, you know, there are a lot of people doing cash collection in many companies. And cash flow performance matters for your investors, for the financial market. And now we can actually have very smart agents who really can track root causes. Why is this really an unpaid invoice?

13:55Is there an issue with the product? Is there an issue with the commercials? Did we not ship the product in time? And then really figure out, okay, let's really go and collaborate with the supplier, collaborate with the service in order to get this outstanding payment really in. much faster than in the past. And these are, of course, super attractive use cases for a CFO, for a treasurer. And we're actually delivering those agents now in every function. And I have to say, we are really amazed about the progress the technology is making, but obviously then how we can plug it into the businesses of our customers.

14:31You know, one of the things I think is interesting is that you're not out there trying to create your own model, right? Like what OpenAI is trying to do or Google is doing with Gemini or you're partnering. You've partnered with Anthropic on Cloud, but what's to say that these models just can't replace your business going forward? I mean, Tim, actually a very good question. And here, I mean, we are partner indeed. We are not building our own large language model, but what we are building is our AI foundation for business data. And because, I mean, the LLMs are super smart and understand unstructured content and really actually also, of course, helps a lot with human process language.

15:12But at the end of the day, they also, you need to enable them to correlate it with business data. Because the good piece is that these LLMs have no access to a financial data of a company or HR data or payroll data. And that is what an SAP brings to it. So that we can contextualize. It's a wall, if you will. It protects the sanctity of that data. Exactly, exactly. And there is billions of data points inside a company and how does the large language model when you actually ask tool this is our digital assistant similar to co-pilot when you ask a question hey and for this piece here what i need to procure i mean this product give me the the five suppliers who can deliver it in with of course in time at the lowest cost with you know which we also knew They have great quality and maybe even the most sustainable way.

16:08And in the past, actually, there was, of course, procurement searching through thousands of suppliers. Now there is this digital chatbot called Joule. Joule is now enabled to understand your question. But in order to understand your question, you need to be able to correlate the chat GPT with the business data, with your supplier data, with your financial data. And this is where SAP, where our AI foundation kicks in. And that is exactly why we have such a strong belief that SAP will be a leader in business AI. Coming up, Europe has lagged behind the US and China in the AI race. But Christian's skeptical that playing catch up is the right game for European tech companies.

16:52We are not having, you know, an open AI in the country. We are having not an entropic. So training LLMs is not necessarily something what we need in this country. How does he think Europe can compete? That's next.

17:15I'm curious, as you talk to your customers about AI, what are you doing internally? How are you using AI to change the way you are operating as a company? You're dogfooding, I imagine. Oh, yes, yes, yes. That favorite Silicon Valley phrase to test on yourself. Yeah, we call it shrink your own champagne, because it hopefully feels like champagne. No, look, again, in order to be credible to our customers, I mean, you don't get any software deal anymore through the door if you cannot really have a convincing, authentic AI story and how it helps their business. So we need to be a role model and apply AI by ourselves.

18:00And we actually going, it goes across the company. Take, I mean, software development is probably the easiest one. Code generation, automated testing, self-configuration. I mean, a lot can be really automated already today. So we are using our old code generation tools, Joule for developer, we have GitHub and other tools. Second piece obviously is when you're in sales. I mean, every deal we are doing right now, there is an AI agent helping to figure out the right price, the right package. oh we sold this deal a million times in this industry let me help you I would actually configure that with these and these services because it makes your offer even more attractive so that is what is happening in sales and then downstream there is a genetic ai helping us said okay here you have this position based on the skills based on the experience based on some other factors what you are looking for here are the three top talents in this company which we actually believe could be, you know, a candidate for that role in the future.

19:00And or if you do recruiting, we have a lot of jobs which we now need because of AI. And this is where we actually have a recruiting agent helping, you know, to screen and making sure we are finding the best talents in the market. And now I could go on and on and on, Tim. But it's very important one thing, and I guess this is something what everyone needs to understand. This is not actually implementing a piece of technology and magically it works. I mean, there's a lot of change management. You need to train the people how to use those agents. What is changing then for myself, for my daily life?

19:32So there is not only that you can just say, oh, I'm implementing today this AI use case. No, there is a huge change management effort. What you have to do in order to make sure that the end user is also adopting the AI. Let's zoom out a little bit. We'll zoom out a lot. SAP, global company, obviously. Founded in Europe, you're talking to me in Germany, Europe is a part of the world that's not necessarily seen as the big tech hub. It may be lagging in the tech space, if you will. Why is that? What needs to change? Yeah, I mean, Europe is definitely a place in the world where I feel oftentimes we think about the risks first and then regulate before we even start innovating.

20:17And I see, you know, in the United States, but as well as in Asia and China, I actually see that it's going the other way. Let's first innovate and then we can still, you know, do the regulation if we see, you know, there is a bad outcome for societies, etc. But this is, of course, not helping, especially not young startups where you have to move with agility, where you have to move with speed. And then finally, also the European Union is not really a union, also not for tech companies. I mean, every member state has its own regulation. And then Europe, Brussels come with one on top. So that is not what you need in order to scale your business fast.

20:58And these are some of the reasons why I would say where Europe definitely has to change in order to hopefully at some point also have another SAP here. Brussels, where the European Union is based, the European Commission operates out. The European Union has been aggressive in some tech regulating in recent years, including passing legislation about AI, AI rules. Recently, the European Commission has talked about pausing some of those rules, had concerns about competition. Do you think the EU is moving in the right direction now? Yeah, I mean, I definitely feel they're eager to listen. And, but, you know, listening is a first good step, but then the second step is now taking action.

21:42And that is now the proof is still, is still out and the jury is still out. So Tim, I would, I would love to say to you, it's moving in the right direction, but I would say when we talk again in three months from now, probably I can, I can really say, is it really moving in the right direction? Well, maybe we should talk bigger picture than just about the tech, the European tech industry in general. What does the European tech industry need to do to not just catch up with the U.S. and Chinese competition, but to beat them? Yeah, it's a good question. There's a heated discussion even in Europe about that because, you know, the European Union decided to fund certain gigafactories to actually also deploy AI across Europe.

22:25but I'm actually I was a bit skeptical I'm still skeptical because look at the energy costs in Germany they're super high and we are not having you know an open AI in the country we are having not an entropic so training LLMs is not necessarily something what we need in this country because we're not having these companies now what Europe should do is not just you know taking you know the same route as the US does or China does what our strengths are. I mean, we are good in manufacturing. We are still good in automotive, despite any huge transformation as well. We are good in utilities and chemical.

23:04And we have a lot of data there. So we have expertise and data. So we should be the world's leader in applying AI in order to be the best in running these utilities in the future, in order to produce the best EV cars in the future at lower cost than today. And this is something what we should focus on. So let's build really these modules by industry, applying them in our business context, because this is where Europe can still win. And then, of course, when infrastructure, when gigafactories are needed, okay, build them, but we should rather do it the other way around and start with applying AI first.

23:43Christian, thank you so much. This is a great conversation. I look forward to talking to you again in the future. Yeah, thanks a lot, Tim. one quick note earlier in the conversation we referred to the works council as the workers council we reached out to the european commission and a spokesman said that building large-scale ai infrastructure in europe is quote central to support research startups and industry while maintaining strategic autonomy. End quote.

24:19And that's bold names for this week. Our producers are Danny Lewis and Alexis Green. Our video producer is Kasia Brusolian. And our fact checker is Aparna Nathan. Jessica Fenton is our technical manager. Jessica and Michael LaValle are our sound designers. Jessica also wrote our theme music. Our supervising producer is Katie Ferguson. Our development producer is Aisha Al-Muslim Chris Zinsley is the deputy editor And Falana Patterson is the Wall Street Journal's head of news audio For even more, check out our columns on WSJ.com We've linked them in the show notes I'm Tim Higgins, thanks for listening

25:05at janice henderson we strive to deliver better outcomes through insight-led investing innovation for clients and a partnership approach that's core to our identity i'm jessica lianzini head of fixed income client portfolio management i'm passionate about my role in investments it allows me to partner with our clients simplify fixed income and support them in achieving better outcomes for their clients. Janice Henderson, investing in a brighter future together.