STRATECHERY.COM
An Interview with SAP CEO Christian Klein About Enterprise AI
An Interview with SAP CEO Christian Klein About Enterprise AI
Friday, May 9, 2025
Listen to Podcast
Listen to this post:
Good morning,
Today’s Stratechery Interview is with SAP CEO Christian Klein. Klein started working for SAP as a 15 year-old student, and became CEO in 2019 at only 39 years of age. That means that Klein is actually eight years younger than the company he leads, a rarity in tech; SAP was founded in 1972 and has been the leader in business process software, particularly ERP systems, ever since.
In this conversation, conducted 10 days before SAP’s annual Sapphire Conference, we get into Klein’s path to the top, which is intertwined with SAP’s path to AI. Klein’s day-one focus as SAP CEO was accelerating the on-premise software company’s move to the cloud and unifying its disparate offerings — including a host of SaaS companies acquired by his predecessor — onto a common platform. That approach could not be better timed given the arrival of AI.
To that end Klein paints a compelling vision of why SAP is well-placed for the AI era: the company holds its customers’ most important data, and as companies realize the importance of a common data layer, the natural choice for all of their back-end processes will be SAP. And, when it comes to unstructured data, SAP earlier this year announced a fascinating partnership with Databricks for a product they are calling the SAP Business Data Cloud. We get into all of these topics and more, including SAP’s social network for supply chain, why Klein wants to co-develop with customers, and SAP’s competitive positioning.
As a reminder, all Stratechery content, including interviews, is available as a podcast; click the link at the top of this email to add Stratechery to your podcast player.
On to the Interview:
An Interview with SAP CEO Christian Klein About Enterprise AI
This interview is lightly edited for clarity.
Background
Christian Klein, welcome to Stratechery.
Christian Klein: Yeah, Ben, thanks for having me.
I usually like to start these interviews by asking about your background before your current company, but in your case that background is very short. So I guess what did you do in the very brief number of years before you started at SAP?
CK: (laughing) I indeed started my career as an intern here at SAP. I grew up in the region here, still my parents, my family and my friends are living here from school days. And then, yes, I started my career as an intern, and then I have seen so pretty much every function.
You were 15 years old, right?
CK: I was actually 15 years old. Actually, I started in the cellar in the IT department carrying big displays around and when everyone needed a new monitor, this was Christian getting called.
And these are big CRT displays, these are not flat screens.
CK: Yeah, they were not like the flat screens back in the days, they were pretty heavy, I’m not sure if I could do the job today.
(laughing) Well, I don’t know. I have to say, before the call I had the CEO of SAP helping me figure out Microsoft Teams. So, I think your youthful acumen is probably very extensive.
CK: Thanks for the compliment.
So you mentioned it was an internship. How does a 15-year-old get hired at SAP? Is this a German thing? I’m very curious, that’s such an early age. At 15, I had a job, that was our generation, but I was making pizzas.
CK: You can also do pizzas in Germany as an intern, for sure, but here and during school time, actually, and when you are actually in school, oftentimes you have a week, two, three weeks where you’re getting asked to find a job and the company’s here in this region, then they offer internships and it’s actually a pretty standard thing. I benefited from that because, believe it or not, even with 15 years, you can network, you can build your first connections. I got the flavor of what SAP does and it helped me to shape a little bit my thinking about what do I want to do in the future.
So did you start there and you liked it and you go to college and you’re all along, “I want to work for SAP,” or how did that path go?
CK: Actually, I was 15, 16 years old, I still had the dream to become a professional football or tennis player, but I realized very fast that my talent was not good enough. So when I then started my time at the university, I really figured out what my passion is, and my passion is to talk about businesses, business processes, business modules and bring them together with great technology, and that was actually fascinating.
Wait just a second. That sounds great for a reflection back as CEO of SAP. Was your passion as a teenager actually business processes?
CK: Honestly, yes. I had also my legal courses at the university, I learned many other things, but my passion was always business administration, and then really focusing on manufacturing, retail. I was always fascinated about, “How do these companies run?”, “How do they transform?”, and then of course with SAP it’s a good fit because then you can also fit the technology to it, and because the technology plays a key role on how these companies should transform.
Well, I do buy it because I think it was about that age that I was very much getting into technology, and it seemed very odd at the time, but I was very interested in the business side of things, even from then. So whereas there were other people that were into technology, but they were nerds, they wanted to actually program the computers. I’m like, “How do these companies make money?” — Microsoft dominance, how chips work to an extent, but Intel versus AMD, those sort of fights. So I can buy it that you started out early.
CK: Back then my next internship coming from it came when I was a vocational student at SAP. For example, I was working in the financial accounting department and I was fascinating by doing my general ledger recordings, my bookings in the system, and I was fascinated on how does this all fit together so I could apply my business knowhow from the university to the real world, and then how is this getting transacted, how does this finance data then get over to HR for payroll data, etc. So that is actually, I learned from the get-go on how companies won, and that fascinated me.
Well, very cool. You are, as of this week, 45 years old, happy birthday! Which is to say that May 1980 was a good month for birthdays as that was my birthday month as well. However, as we just discussed, I’m a blogger sitting in my home office, you are CEO of one of the oldest and largest enterprise company in the world. We started at 15, but it’s only 30 years and now you’re sitting on top. How does that happen?
CK: It was definitely not my plan when I then started my job at the age of 21 to one day become the CEO. But at the time when you are going into SAP, and I guess this is also what makes SAP special, I started my work at Active Global Support, so supporting our customers if they had issues with our systems. Then I moved over to development, to finance. I was moving to the US wanting their success factors, back in the days we acquired this company and I got really good insight on how to run an end-to-end SaaS business. So I had different insights into different functions, I had different mentors, and no one asked me about my age. No one asked me at the end, “What did you study?”, they all looked at, “What can you contribute, how can you help us to steer the business? How can you help us to code better software?”, and that was always fascinating to me.
At that point then I would say from a personality perspective, I’m pretty competitive. Not competitive in a way that I needed to always shine within the team, but competitive in a way to do my job better every day. And at a certain time, finally when I became the Chief Operating Officer of SAP, I said, “Hey, this would be great to run this company because I have certain ideas, I have a certain vision where I believe we can, SAP move to the next level with all the new technology which is on the horizon”. So I moved my career ahead step-by-step, and then finally I thought, “Hey, yeah, it’s fun to run this company also as a CEO”.
You mentioned this, that it is a compliment to SAP, and that’s certainly the way it comes across to me, that someone — to be clear, you’re eight years younger than SAP, which I guess for traditional industries, that’s not odd that the CEO is younger than the company, for tech, it’s certainly very odd — but was there some sort of program that early on sets you on this track, or is it just a matter of things have changed so much particularly in the last, I would say, probably 15 years with the cloud and whatnot, which we’ll get to all these bits and pieces, that it was just a particularly great time for someone young to seize that opportunity.
CK: The vocational student program I actually attended at the very beginning is something which we brought now also to the United States, to India where we are cooperating with the best universities, because I believe in this concept, the concept is about your working actually for four or five months in a department of SAP. If you study computer science, of course it’s more engineering-focused, if you study business, then it’s more like you are working in corporate functions, in sales and marketing and consultant, and then you actually go back to university.
While I finished then my study, I had a) a great network, I had b) a great understanding of the culture on how SAP works. And again, the start into the real life into my real job was much easier. SAP knew me, I knew the company, and the likelihood that it fits, it’s of course much better when you’re just starting without any knowledge about a company just fresh from the university, and so we brought this concept now also to other countries in the world, and a lot of great talents from all over the world are working now in top positions of SAP. We have a pretty young leadership team and that is, I guess, also a benefit of the vocational programs we are wanting.
Becoming CEO
Do you think there was an aspect where after a decade of a brash American outsider, I talked to Bill McDermott on Stratechery a couple months ago, was there a real hunger for someone homegrown, and that was a real allure to people in the company?
CK: I would say it was not actually about needing someone homegrown. Bill was a great CEO for SAP, he started the cloud transformation, he did many acquisitions, which we are still benefiting from. But Bill and I are also different. Bill, of course, can get very excited about the customer and the business itself, and he was very much focused on driving the top line growth. I’m probably a CEO who is connecting the product with the customer. So I always ask myself, “How can we use our technology best to solve the biggest needs of our customers?”, and I guess that was probably the right time to have someone after all the acquisitions to see, “Hey, how can we get all of these mission-critical businesses on one platform? What can we do with our data? How can we bring all of these great assets together and actually really start with our cloud transformation so that all of SAP is SaaS, and not only the acquired parts of it”.
Stepping back, and you were a COO for a portion of this time, you’re buying all these cloud SaaS applications, what was the thinking there? Because today, we’re going to get to your vision of integrating everything more tightly together, I think it’s really compelling. But I just want to understand how we got to this position. You have this on-premise software, is it just that you have a channel, you have distribution so you can more easily sell these cloud applications? Or was there a vision of this integration and it just maybe took longer than you thought?
CK: Look Ben, being a child of SAP, starting my career so young and knowing how the company works, I knew pretty well what customer enjoy about SAP and didn’t enjoy so much.
Tell me what those are.
CK: When I became Chief Operating Officer of SAP, I was responsible for our IT, and obviously we are running SAP solutions, our ERP, our CIM, everything. And I felt, “Oh, I have to transform this company”, and the ERP plays an instrumental part, because in the cloud you sell differently, code differently, you service your customers differently. I had a very homegrown, customized ERP, a very complex system, extremely hard to upgrade. It served our needs very well, but it was not on the latest and I had all of this great new technology, we will definitely talk about AI in a second, but I was eight years behind the latest release and I thought to myself, “Hey, this business model, highly profitable, highly successful, will not lead SAP into a bright future, because there are all these best of breed competitors coming in”. All of these lessons which I learned, I brought into my role as CEO and said, “Hey, look, wait a second, we are running here a good business, but we have to disrupt ourselves because customers need way more agile systems. They need the latest innovations. They don’t need to spend hundreds of millions of upgrading a system, they need to be on the latest”.
What is also different about SAP, is SAP wants business models, business processes so we cannot only sell a piece of technology. We have 400,000 customers, we know how Nvidia wants their supply chain, we know how Exxon wants their kind of business for also now renewables, we also have to exchange and be much better on business knowledge, exchanging best practices and helping customers to drive their business transformation, while moving them to the cloud to give them also the agility they need because definitely with the on-prem software, this time is over.
The benefits of getting people onto the cloud today with AI is super clear. But you take over in late 2019, was the AI vision already in place, or was it just a sense that, like you said, you were CEO, you’re running on an eight-year-old version of SAP, what was the push? Were you worried about losing customers to competitors? SAP is so sticky, is that really even a concern? What was the motivating factor at that time?
CK: Yeah, SAP is sticky because it’s mission-critical software, which is great, but you cannot count on, “Oh, we are sticky, customers will trust us forever”. So the move to the cloud is existential, and we are on a very good track. Most of our customers are on the way, and as we are running their mission-critical businesses, of course it takes time.
But the good pieces, we are not only selling now a piece of cloud technology, we are also helping them to transform a retailer around omnichannel, resiliency of supply chains, the oil and gas about renewables, the utilities about having much more flexibility in the way how they attract and do the meter to pay for new energy sources, so that’s SAP.
Now, AI, Ben, to your question, of course, in 2015, 2016, we already had our first machine learning modules and so on, but what I didn’t like is we played the “me too”. We played the “me too” of other tech companies’ offering on our platform machine learning services. You could build your own modules, but when you are SAP, think about that — you’re running the most mission-critical business processes of the world, you have access to so much business data. So you have a suite, you have mission-critical data, we have to embed AI right into the business processes of our customers so when you do financial planning, you don’t have to code another AI module for that. We are actually infusing the intelligence into your financial planning so that you can simulate right away the impact of tariffs on your financial guidance, or when you then run a supply chain and you demand in supply chain planning, don’t code with Gen AI or traditional AI, another AI app embedded right away into our solution so that end users can use it out of the box. Obviously they can do some fine-tuning, but this is what SAP can do and as we have the business data other than the consumer-driven LLMs, that is what differentiates us. So we are now playing Business AI on our terms, on our strengths, and that is really infusing AI into the businesses of our customers.
You’ve talked about having a suite-first approach, and this applies to the AI bit, getting a common data layer between, again, you had on-premise SAP installations, you had all these cloud SaaS applications or companies that you bought and then you also needed to move SAP itself to the cloud. Was that one process together or were those two work streams that had to happen in parallel?
CK: Very good question, Ben. It’s exactly the work what we have done starting in 2019, we started exactly what you just described. We brought all the data together, and not only bringing data together, technically, when you have HR data and financial data, it has to match, you have to run payroll, it has to match to your P&L. If you’re running CIM [Central Invoice Management], sales order data, your consumer data has to match from ERP [Enterprise Resource Planning] to CIM. When you deliver it, manufacture it in the supply chain, your consumer and your supplier data have to match. So we brought this together semantically, that was always the strength in on-prem, we had it in the cloud.
Now in the next step with AI and our new offering, Business Data Cloud (BDC), we are saying suite-first because we have the very mission-critical SAP data. Then Ali Ghodsi, the CEO of Databricks, once reached out to me and said, “Hey Christian, I have so many SAP customers and your data is the most mission-critical in their company, but I have a lot of non-SAP data and I feel we are creating all of this data in big data lakes and then very expensive data scientists are coming to make somehow meaning to this data, but we can do this way smarter”. I said, “Absolutely, Ali, we can do this way smarter, why we are not building this one data layer where we can semantically match SAP structured with non-SAP also unstructured data?”.
That is now the evolution in the data game that we are really building this one data platform and we started this with Databricks and it’s very exciting, because then when you think about it, about AI and agentic AI, we are of course building our modules with SAP data, so far but suddenly an AI agent can talk about cashflow collection also based on non-SAP data, because maybe some of your sales data, your commercial data is sitting somewhere else and that is of course also super, super important to make our AI even more powerful. When you want to do data engineering on the Databricks side, same there, you can have access to these semantical data products and you can really create AI based on this semantical data layer with BDC.
Is there a bit where you’ve gone full circle, where initially the thought that, “We need to integrate with other data places is important so let’s buy some of these companies in this area?”, and actually it turned out you had your bread and butter, the mission critical part of the company that mattered, and the key thing was maybe partnerships was a better approach all along, and you’re getting that almost through underneath with this Databricks partnership?
CK: I would say Ben, yeah, this is right, I always believed in the benefit of integration. When you also look by the way into the success of Apple and Microsoft and you see how these assets are integrated, integration is key also for customer experience. But in the on-premise world, it was integrated in a monolithic landscape, and this monolithic architecture.
With very hard walls around it.
CK: With very hard, not open, hard walls, then you put ten times more custom code around it, and at the end, no one knew how to ever upgrade this ERP again. So now in the cloud, we bought some cloud solutions, but we renovated them, we put them on one platform, but not monolithic, you have it modular. So with that, we can always make sure when the customers are moving now to the cloud, we can keep them on the latest, they see a new SAP, agile with speed, latest innovations, fantastic.
Then you have this common data layer underneath, and this common data layer was all about SAP. But the world does not consist only of SAP data, so we are opening ourselves up and say, “We are doing this very impactful partnerships to give you even more value”, and especially data as we all know is super important. In the B2B world, it’s really important also about high quality data, that is actually very much different to BDC because when you deal with financial AI, HR AI, supply chain AI, you need very accurate results at the end of the day so the data accuracy is super, super important.
Business Network
Can you tell me about the Business Network? This was something I actually was not super familiar with until prepping for this interview and talking to some of your colleagues, but I thought it was very compelling. Explain it to me.
CK: The Business Network actually, we brought the Business Network in with an acquisition we did with Ariba. The Business Network was there, you are a buyer and you have your suppliers, and you can digitize the transaction, you can digitize the documents. Great, you have much less paperwork, you are more productive, you can procure faster, good.
But then COVID hit, and in times of COVID, a lot of our customers came to us, take for example the vaccine, a lot of pharma bioscience companies came to us and said, “I have no clue how to get all of the ingredients to my factory in a certain location because I have no visibility in my supply chain”, and they said, “But Christian, wait a second. You are running us, but also probably all of our suppliers procure and buy with SAP software”, I said, “Yeah, that’s right”. I have millions of suppliers in my network, so what we started to do for pharma, but also for automotive, we connected the suppliers. So suddenly, for example, take a vaccine, you can see all your suppliers lined up in your supply chain down to the raw materials and really down to the very single ingredients to really track and trace, “Is every supplier able to deliver at what time?”, and, “Can we also make the logistics work”, which was super important in COVID, but still today. Think about tariffs, you also want to understand where to procure from to avoid getting too hard hit about tariffs.
So the Business Network is an extremely powerful network to not only sell supply chain solutions or procurement solution, it’s also about the transparency and the resiliency part. Then last piece, in the automotive sector for example, we brought together all the large OEMs, we brought together the brake manufacturers from Italy down to the raw material providers for lithium in Brazil and now they’re not only having the visibility, but they said, “Hey, wait a second, now when I produce my car, I can actually not only see what my own carbon emissions are, I can actually also track and trace what are the carbon emissions from my suppliers down to the raw material and with that, I have for the first time a clue what the emissions actually are for my electric car”. That is of course, still very valid to not only having the transparency, the resiliency, but also getting more insights around the sustainability of a product.
It’s amazing, it’s like you have a social network for the supply chain, it’s not what you’d expect.
CK: Yeah, you could actually start in our marketing, our branding — exactly, that’s what I always called my market. We should call it the social network for B2B! I feel it’s still a very hidden crown jewel but more and more customers, like take for example a Microsoft and Nvidia, they use that as well, so we are bringing up this industry supply chain.
They’d be your sellers, because they’re going down into their supply chain and say, “Look, you have to integrate SAP because we want visibility”, and they become some of your biggest champions.
CK: And Ben, if you think that’s true, if you are in there as a supplier, and we are now developing very smart search engine so you can find more buyers. Buyers can go in and say, “Hey, I have maybe a shortage in my inventory here and my suppliers can’t deliver, maybe there’s in the SAP network more suppliers and vice versa”. The suppliers can also find more buyers, so it can really become a very, very powerful B2B platform also for all buyers and suppliers in the network.
What I think is pretty interesting is there’s two layers of integration we just touched on. We have this suite-first integration where of course you’re an enterprise seller, so you will let anyone bring to the table what they want, but it’s going to be a much better experience if you use all of SAP’s product, particularly as you get the common data layer and then partnering with Databricks to get the same data everywhere. You also have this, again, social network for business angle, which I think is super interesting. Like I said, I didn’t really know about this until about a month ago. These are very high levels of integration.
On the flip side, you decided to not build your own cloud, you are partnering very strongly with AWS, with Azure, and you can run SAP on Azure for example. So it’s like you decided, “Okay, horizontally we’re very tight together, but vertically it just makes more sense to partner”, walk me through your thought process around that.
CK: I would say in the software world, in the tech world, no one can code or deliver everything on their own. In every piece of the stack, there are third party components in, and I think where it matters to us most for our customers, where they see SAP, they see actually SAP starting on the platform layer and in the cloud, it’s not monolithic anymore, they say, “Give me the best integration layer — by the way, give me the best integration also to non-SAP, give me the data layer with BDC now, give me the extensibility so that I can code with AI coding tools, with tool for developer, industry verticals”. That is super important.
Then the apps, of course, they need to be world-class. Give me embedded AI, give me this supply chain resiliency, give me financial and integrated planning cross-company, give me omnichannel, and that is of course where customers see and need SAP.
On the infrastructure, on the commodity level, we decided indeed to rely on the hyperscalers, because they have a very powerful offering, they are multinational. We still have our own data centers, by the way, we have 50 data centers around the world, but we give customers choice. Why? Because we believe it’s not so strategic for SAP. We can partner and partner and our customers love to having the choice. Maybe you have already Microsoft 365 and you want to combine your workloads on Azure, we give it to them. Maybe you already want some of your data in BigQuery so you can combine everything on GCP, and the same with AWS. I feel this is a winning formula that we are saying, “Look, we are open for these partnerships”, but what we also of course do, Ben, and especially in this geopolitical times we are having right now, which are by far not easy. Of course we have various sovereign cloud offerings in the EU, in the United States, in China, in Asia, so we can give our customers all what they need and focus our R&D on the stuff where customers really see us on the platform and then on the application layer.
Best-of-Suite vs. Best-of-Breed
You’ve brought up AI a couple of times, obviously that’s a big deal right now. You have this AI-first, suite-first strategy, you’ve talked about building this common data layer. Where does AI sit relative to that data? On the consumer side people go to AI for answers and which can be mostly helpful and sometimes a little risky, I think there’s other angles where AI is just a user interface element to get access to what’s there. How do you think of it in the context of your enterprise offerings?
CK: Where we are going with SAP and what our strategy around AI is, I definitely believe the best-of-suite is now even more important, and that the end of the best-of-breed is pretty close. Because why? You can have of course as an end user have access to thousands of copilots and digital assistants, etc., but at the end, the way how business runs, business runs across end-to-end business processes.
Think about a process like when you source-to-pay. You source, you need to find the right supplier, you put up an order, you actually pay the supplier, you want to have it connected to your supply chain, and this is SAP. So the best-of-suite matters and because you need the data, you need the business context, that’s the suite.
Then when you put a Joule, a digital assistant on top and say, “Hey, you know what, Joule? I want to source a supplier for this material, give me a choice of supplier based on cost, based on availability to deliver quality, etc”, we have all of this data. We are not only having procurement data, we have the suite data and with BDC, we have sometimes then also external data. Then Joule, our digital assistant, can help to source. But Joule then and our agents then in the supply chain can also take this new material and say, “Okay, supplier will deliver by date X, let’s make sure that it goes to our manufacturing team so that they are able to understand how do they do the factory planning, the shop floor planning”.
So you see how important the suite is so that a digital assistant can really work across. When you think now about agentic AI, I mean all of these agents, they are great, but the agents need to be able to talk to each other, to run across end-to-end processes. At a certain date, I’m pretty sure we will have see all agents wanting businesses, processes end-to-end, and for that you need the data, you need to understand the business context and there is no need absolutely to put another layer on top of more agents, we can embed that right away in the businesses of our customers.
It sounds like you could go back to the on-premise SAP era, you talked about it being monolithic, and which was the case. We go through the cloud SaaS era, everything breaks apart and there’s best-of-breed for all these different areas and it feels like SAP is a dinosaur. It’s like, “Oh yeah, there’s the ERP system that we’re stuck with it because it’s sticky, but ideally for these new functions, we can do X, Y, Z”. The story you’re telling here about this common data layer about how agents can work better if it’s all unified, the same semantic index, are we back to monolithic actually being the answer for most businesses?
CK: I would say it’s not monolithic, it’s a modular ERP, but it’s sitting on one platform. So you have modules, you have the agility, you can decide to consume HR, then you go to Finance, then you go to Supply Chain, you go to CIM, Sales, Commerce, you do this step-by-step. It’s not monolithic, but there’s still a platform underneath that gives you the out-of-the-box integration, the integration of your workflows, the integration of business processes, the integration of data. On top, we are building these agents who then can use this high quality data and then Joule who actually becomes our UX because in the future, there is no need anymore to type data out or get data out of an SAP system, you are actually running SAP.
SAP and UI, very famous.
CK: Exactly! So Joule will become our new UI, and it’s very much you use SAP via your human language, but there is not anymore this need to type data in, get data out, screen documents. That is what Joule can do for you in the future.
Do you see an opportunity for SAP to be winning core business functions where people had you for ERP, but now it’s like, “Actually we should do CRM?”, or, “We should do all the different various things that you have so that we can reap this advantage”?
CK: I’ll give you a real life example. Last week, I was in Switzerland, talked to a company. They are producing elevators and they said, “Christian, even the elevator business, it’s changing a lot, we want to package services to that, we want to infuse our AI for predictive maintenance. But guess what? Our frontline, our sellers, they configure something and our supply chain, our people in the supply chain functions have no clue about what has been configured at the frontline. Help me, I want to get rid of some of the best-of-breed, I want to go to a best-of-suite where my sales, where my services engine work seamlessly together with my manufacturing, with my logistics engine”.
So we are bringing this together in the cloud, modular, but then on top we say, “Hey, let us infuse AI. When you configure a new elevator, our configurator should tell you what is the best service package? What is the best price? Are you able to deliver at that time? Let the AI agent for sales do this job for you”. Then we transfer the data to the fulfillment functions of this company, and then it says, “Hey, there’s new demand coming and by the way, do we have everything on stock? How does this configuration need to be translated into the factory line?”, then we ship it and then we already plan the shipping. All of that will be done on one data module, consistent workflows and AI embedded to automate and to drive this way more intelligent than in the past, and this customer told me, “Christian, you know what? I get rid of four or five best-of-breed software prodcuts because I would definitely believe in the best-of-suite, because I know I grow better and I actually be more productive if I want this end-to-end in the core with SAP, obviously there will be also non-SAP, but in the core, this is where I need the suite”.
Change Management
I mean that’s the whole thing, is we mentioned the “sticky” angle before, you own the most critical functions. Literally no one in the world wants to tear out SAP and put something else in, no matter how frustrated they might get but if this is the case, if it’s not just you’re talking your book, but thematically that to take advantage of AI, it needs to be this common data layer and integrated. What are your obstacles? What are your challenges? It sounds like you could sweep the space.
CK: Ben, look, it sounds so simple. It’s oftentimes not that simple also, to be very frank here.
(laughing)Yes.
CK: Why? When I’m running in such IT projects and I’m only seeing IT people around the table, they are needed, but in order to do what I just described, and in order to transform a business model, you need a business at the table. You need to think through in the digital world, how will this company sell? How we will service the customer? How do you want a price? How do you want to fulfill? How do you want to deliver faster? And for that, it’s not only that we are coming with our best-of-suite and saying, “Here we are, let’s just implement it”.
You need the business at the table, you need to have a change culture which comes from the top and says, “Similar to what we did, SAP or what Microsoft did or what others are doing, we are disrupting our own business module and the customer needs to be up for that”. So don’t see SAP as another IT project, we are more than infrastructure, we are more than a move to a new hyperscale environment, we are running your businesses, and you need to see this projects as a travel project and I need change management. I need the people who lead that together with IT and we come in with our tools to tell you, “You know what? We have 20,000 other customers in this industry, in your segment. Here are the best practices. Let’s benchmark your business processes against the best practices, what we are seeing”.
So SAP needs to play a way more active part not only in the IT, but also on the business side because we of course, have these insights, we have the data. So also in the way how we get customers there, we need to play a way more active role. And trust me, Ben, that is the hardest part for many, many companies, to disrupt their own business model, to redesign the way how they work and then apply to technology. Migrating to new technology is actually by far more easier than really doing the hard work at the beginning with the business.
I want to zoom out on something you said elsewhere along these lines, where you’ve referred to SAP’s early days, which again, I am delighted, given that we’re the same age, you’re pointing to days before you were born, but when developers of SAP, they co-developed R/1, your first version in conjunction with customers in their data centers. You’ve talked about wanting to revive that spirit, is that sort of what you’re driving at?
CK: Absolutely. What we are doing is oftentimes with customers, we are actually saying, “Hey, before we start this project, let’s build a prototype, let’s build a prototype with real solutions on how you want to run your business” and then we send our product management there and say, “Okay, you know what? There is a point where I see some capabilities in the chain, which we should probably have in the standards, so let’s co-innovate together”.
When we think about AI, there’s a rule inside SAP that we don’t develop a new AI use case anywhere in our portfolio if there are not ten customers at least lined up early where we can co-innovate, because I don’t want to just code and count my number of AI use cases and play into a hype, I want to see where are the customers who see A) the value and B) also the feasibility to implement it in a pretty fast manner. That is the mentality what we have to have that we are getting our PM teams super close to the needs of our customers and also honestly sometimes getting them on-site and bringing our customers together in councils to better understand what they need all across our stack.
This is very interesting to me, and a couple of the other comments that you’ve dropped, this is actually one of my theses about enterprise AI adoption. You mentioned this before, but where this initial wave of enterprise AI has been about giving employee tools, copilots, whatever, that they don’t necessarily want to use, it’s kind of like early PCs in that regard. There were some people who really wanted a PC and they did amazing things with it, but it kind of took a generational change for everyone in the office to be using a PC, but that wasn’t the first wave, the first wave was SAP. Like in the 1970s, you go in top-down, a visionary at the top saying, “Actually we’re going to replace entire bookkeeping departments, people tracking inventory by hand, doing accounting, all these sorts of things”. It seems like to me, AI could be similar. The first real meaningful wave that actually moves the needle is more about the widespread replacement of certain functions. If you do that, like I said, someone at the top has to do it, it’s a top-down decision, they have to weigh the pros and cons, the error rate versus humans and the costs and all those sorts of things. But also, it’s going to be a long, extended implementation to do it, kind of like those old SAP implementation back in the 1970s. So what you’re saying makes a lot of sense to me.
CK: Two parts to that, two things. First, the journey you are now on with SAP to the cloud is your last upgrade because afterwards, we are keeping you on the latest and that is of course for a CIO, a dream coming true because this ERP implementation projects, they were heavy, coming back to the monolithic and now we have the modular, and so that is very rewarding, doing it for the last time and invest your IT money in innovation and not in upgrading of system.
The second part of it, I guess most of us have read the book Innovator’s Dilemma.
Yep.
CK: You wouldn’t imagine how often this plays into this SAP projects because think about it. I mean, has SAP seen the move to the cloud in 2005, 4, 3? Absolutely. Was SAP willing at the top to disrupt a highly profitable business model with double-digit growth rates? No. Then in 2012, we acquired, we started acquiring. Did we transform the company? The traditional business? No. And then in 2019 we said, “Okay, let’s do it before it’s over”, and we then pulled out a new strategy, but the strategy is nice on PowerPoint, then you have to inform the capital markets, share price dropped by 30% because you take billions of profit away, we’re moving from a traditional upfront license model to a subscription pay-as-you-go business model.
Then the hardest part comes, the execution. Now you can think through that happens in automotive, that happens in retail, that happens in travel, that happens in almost every industry and this is so important. What I always remind my peers about that when they’re starting a rise in ERP project with SAP, do it business-led, think about how do you want to run your company to not getting disruptive and then apply the new technology to that, it’s so important. And again, when you think about the copilot or our digital assistant, Joule, take the public sector in Germany. Still, we are here, Germany is not digital.
(laughing) I’ve heard some amazing stories about that.
CK: It’s very true. But the point is the technology there? Obviously, we are running in Australia digital services for millions of citizens. Why not in Germany? The biggest part is there’s someone at the top who really says, “Look, let’s digitize all our citizen services end-to-end on one platform”, and then doing the change management with the people who are used to work with paper, who are used to do their certain jobs for many, many years, I mean there is also, of course, resistance, which is pretty human. But you need to lead this change and that it starts, it’s not a lack of money, it’s definitely not a technology issue, it’s about change management, the willingness to disrupt the way how you do certain things for many, many years.
You talked about having to disrupt yourself and change from this upfront license maintenance contracts to a cloud subscription model and how difficult that was. It does feel like you almost caught the last train out of the station to do that in 2019 to get you ready for the AI moment of the last couple of years. Did COVID help with that where because you talked about some of these use cases that you got into and you were just better able to be flexible because you’d already started on that road?
CK: The move to the cloud, it was really a big change management effort because the way how you code modular cloud software, the way how you do DevOps is very much different compared to on-prem, and that was true. There’s so much change in all of the functions inside SAP for the people who work for SAP, also for the ecosystem by the way. We have great, great partners, but for they they’re going through a similar transformation supporting our software.
Now on the subscription and the maintenance, obviously yes, it has a financial impact, but the piece is what you can prove in the cloud and did COVID help? Of course it helped in a way that the customer said, “Hey, I have this legacy supply chain solutions from SAP, they are helping me to transact and run my supply chain”, but we talked about the Business Network, we talked about the resiliency, we talked about AI doing demand and supply chain planning, we are now talking about tariffs and how to ship certain products faster to place B or on how to change your production planning to not get so much hard hit by tariffs. That is, of course, in the cloud with our latest solutions way better because there’s way more intelligence embedded. You can do these things much faster, much more productive and much more intelligent. And then, of course, such crisis, such challenges on the geopolitical, on the business side, of course, help to justify spend into IT as long as you can show that you’re solving some real tough business challenges of our customers.
Competition
When you think about competition, we’ve sort of talked about three layers. There’s a data layer, then there’s the application layer and then there’s almost like a layer above you, call it a UX layer. For the data, before when I talked to Ali Ghodsi, he talked about, “Yeah, we can get all the data into our place, that makes us very well positioned”, well it turns out he realized actually SAP has the best data, so can we partner together? You seem to have that somewhat under control. You have the middle layer, which was kind of like the last decade, you have best-of-breeds that are peeling off specific functions. You’ve made a compelling case that you can pull that back in.
I guess that leaves the layer on top and I think about say a Palantir or whatever, it’s like, “Oh, we’re not going to rip out SAP, we’re just going to build middleware and everything’s going to tie into our semantic index and we are going to write all future applications using us on top and you can keep SAP, but it’s kind of a commodity underneath that just feeds us data”. Which of those is the most concerning about you and why does SAP stay on top from this layer perspective or does it even need to stay on top?
CK: Very good question, Ben. All our customers, what our customers appreciate so much about our platform in the cloud, it’s way more open, we have great partner solutions. We cannot code everything by ourselves. No matter if it’s about by industry or by line of business, we have really great partnerships where we are connecting third-party software out of the box.
Then second, talking about data. I mean take Databricks, they are world-class in data engineering, we are not playing in data engineering, but we have world-class data, we have embedded AI, we have tool for our end users. So it’s very complementary and so it’s a win-win-win situation. For customers they have a world-class data platform saving a ton of costs and finally have better data, for Databricks we build together this semantical data layer. For SAP, yes, we have our AI who has access to way better data, also non-SAP, so that is great.
Then moving to the app layer, take Palantir. Palantir is actually a great partner in certain ways because we do enterprise analytics, we do enterprise planning, this is not where Palantir sits. Palantir then comes on top for some very specific use cases in the public sector and retail, etc. So here we are even now discussing how we can even partner closer to give more benefits to our customers.
And then for competitors trying to replace SAP, I think it’s pretty easy. Five years ago that was tough for SAP, now we can say, “Okay, you’re running finance with SAP, you do financial planning, how about your workforce planning? Does this need to be connected to your financial plan? And when you, by the way, do payroll and you want to pay your employees in time, what about the costs? What about the P&L to also ensure you have a compliant quarter end closing?”. So you can really argue a lot about the best-of-suite when you are running commerce like Samsung does with SAP. If they want to ship or want to sell a mobile phone, “Is it in stock? Can we deliver on time? Okay, it’s better to run this with SAP because you have real time all of these processes connected”. You know what’s going on in the front office, you know what’s going on in your supply chain, like the elevator example I just explained to you, Ben. So there we feel in the meantime, we are way more competitive with the best-of-suite.
Last but not least, Ben, maybe one thing, when you are an end user and you are doing your job every day, would you love to use 30 digital assistants who don’t talk to each other? I’d rather have one assistant called Joule who can correlate HR with financial data, who can give me sourcing recommendations based on not only procurement data. I would love to do demand and supply chain planning, not only on demand data but also on inventory data and that is what SAP can do and that’s why I’m so convinced that the best-of-suite, the modular best-of-suite is way superior now in the age of AI than the best-of-breed. That’s why we are saying suite-first and AI-first because you need the data in order to deliver world-class AI.
The business data cloud and this partnership with Databricks is so compelling. Is there more you could do and more in the define the data and this idea you can carry permissions around it all knows who it is, where it goes, you get all that for free in a way that’s very hard to do, you’re tying stuff together. I mean, should Databricks be a part of SAP?
CK: (laughing) Yeah, you should ask Ali that question.
I should ask Ali that question!
CK: Just kidding. I mean look, I love this partnership because the portfolio is indeed very compelling.
I love it too. I love it so much I want to go even further! I think it’s amazing.
CK: Yeah, the two companies have a very compelling and a very complementary portfolio in bringing this together. What I actually see coming out of that, and by the way, we are going to see also more partnerships in that space, but what we are going to see is actually a marketplace of data products when, for example, now many customers are coming to us and said, “Okay, great. SAP you make sure, first of all, that in all countries in the world I have a compliant business because you can translate tariffs into all transactions what we are doing. That is a big benefit, but now help me to simulate the impact of tariffs on my business. How should I deal with price changes? How should I deal with inventory change?”.
Then with BDC and Databricks and SAP, you can build data products. You can say, “Okay, let’s use external data, let’s use some SAP data”, bring it together to simulate the impact of tariffs on your production plan, on your financials, on your pricing, etc. So what we will create is a marketplace of data products, this kind of semantical data layer where customers can pick and choose data products to have an immense value either on the steering, on how they run their business, on the decision-making, but first and foremost, and also of course on AI, because the agents will also have access to this data and coming back to my point, you need the business data, it’s not enough to only have non-structural data. You need this business data, you need it combined with the unstructured data and then the agents can become world-class.
One final question. You’re a German leading a German company in an industry that’s very defined by by-and-large American companies on the West Coast. Is that a strength? Is that a weakness? Is it something unique? How do you think about that as you’re dealing with everyone else in the industry?
CK: Ben, honestly, when you grow up in your career in SAP, you really don’t think about that. I mean, I’m a CEO for 110,000 people, most of them are in the U.S, most of my customers are in the U.S. But of course I have super-great customers in Southeast Asia in all parts of the world. So you think global and at the end we are also making use of this diversity because at the end I get some very smart database engineers in Korea in Seoul, I get some very smart people, data engineers, and yesterday we were in Hungary in Budapest, we have top universities there to source the best data scientists for some of our AI modules, so we are making use of that, and being such a global international company is great. Then in the United States for example, we have, of course, our lab in Palo Alto, but we do a ton of research, a ton of AI development in Texas when it comes to oil and gas and renewables so it’s actually a benefit and we want to use this global aspect of SAP to our favor and that’s what SAP is about. I mean, when you talk about a global company, we are definitely truly global.
Yeah, we’re in the middle of a trade war. If you’re a neutral third party, you’re going to help both sides to help everyone sort it out.
CK: This is very true. You would imagine in the last week so many customers reached out and said, “Hey, with this trade war and potential tariffs here and there, as a U.S. multinational, help me to run my business in China”, vice versa, same thing. That is, of course, an advantage. What we can play out and say, “Yes, we’re going to comply to local regulations and we’re going to make sure that you can run your business and even if supply chains are not that global anymore, we are still able to connect your supply chain”. That is, of course, something which I guess is super special about SAP.
Yeah. I mean it really does feel like the world is coming to you both on the political and economic perspective, but also the technological perspective. I buy your monolithic — not monolithic, don’t call it monolithic, call it suite-first — your suite-first vision, I think it’s very compelling. Christian Klein it was great to meet you and talk to you and hope to talk to you again.
CK: That was really great. Thanks a lot, Ben.
This Daily Update Interview is also available as a podcast. To receive it in your podcast player, visit Stratechery.
The Daily Update is intended for a single recipient, but occasional forwarding is totally fine! If you would like to order multiple subscriptions for your team with a group discount (minimum 5), please contact me directly.
Thanks for being a supporter, and have a great day!
Get notified about new Articles
Please verify your email address to proceed.
←Microsoft Earnings, Microsoft’s Core Capability, Amazon Earnings
English
Arabic
French
Spanish
Portuguese
Deutsch
Turkish
Dutch
Italiano
Russian
Romaian
Portuguese (Brazil)
Greek
