Scaling Industrial IoT: Cumulocity as Strategic Growth Partner 

Hey everybody! Today’s episode is one you’ve been waiting for, as many of you have asked about the future of the technology stack and Cumulocity after its separation from Software AG in January 2025 through the management buyout. How are they working with you and other partners? What’s the hyperscaler strategy for collaborating with AWS, Azure, and Google? Today, we have some great answers for you, whether you’re a customer, a partner, or exploring the IoT platform landscape. This episode will give you the full picture of what has changed, what’s new, and what’s now possible with Cumulocity, including concrete IoT and AI use cases and large-scale industry deployments. Joining me today is Dr. Jürgen Krämer, Chief Product Officer and Managing Director at Cumulocity GmbH. Formerly, he was also part of Software AG.
Let’s jump in! You can find all the implementation details and information for your use case on iotusecase.com. Let’s go!

Hi Jürgen, great to have you here. How’s your day going? How are you?

Jürgen

So far, so good, Madeleine. Thanks for having me here and for the invitation.

Thank you for your time. I believe you have over 20 years of international experience, from being a tech founder to a C-level executive, and of course, at Cumulocity. For those who may not know you, could you briefly explain your background and your current role at Cumulocity?

Jürgen

Sure, sure. First of all, it’s a pleasure to be here. Yes, I’m getting older, and 20 years in industrial IoT is quite a long time. We’ve learned a lot over the years. So, a bit of background: I haven’t been with Cumulocity for 20 years. Cumulocity was founded as a spin-off from Nokia Siemens in 2012, so they’ve been around for quite some time, but I wasn’t part of that spin-off. Bernd Gross, the other managing director, and Stefan Vaillant were part of the first spin-off. At that time, I was already at Software AG. I had founded a company in the data management space, which was acquired by Software AG in 2010. From there, I drove the acquisition of Cumulocity because, while we had great analytics, analytics is just the second step in IoT. First, you need to connect the devices and collect the data before you can do the analytics. That’s how I met Bernd and the others. This acquisition took place in 2017, and since then, Software AG has helped us mature the product, win major customers like Schindler, Hilti, ABB, and others, and expand our international presence. Since January, we’ve become independent again. We’ve now stepped out of Software AG and can accelerate our growth to the next level.

Nice. By the way, I looked it up. Bernd Gross was on the podcast twice, with the first appearance in November 2020, quite some time ago. Then he was also on episode 79. So it’s really nice to have you here and get an update on what’s new. And Cumulocity is out there with a new branding. I saw that at the Hannover Fair and everywhere else, like on LinkedIn.
What’s the new branding about?

Jürgen

We’ve upgraded, and now we’re shinier than ever! It’s a bold yellow branding, and it really stood out at SPS and the Hannover Trade Show. SPS last year was the first time we revealed the new brand, and the Hannover Trade Show just two weeks ago was another highlight with lots of positive feedback. Now, we have a clear focus. You know that Software AG had multiple product lines, and IoT was just one of them. Now, the whole team of around 300 people is fully focused on industrial IoT. We’re a young, dynamic, super agile, and flexible team, and I think we’ll accelerate customer success and deliver faster than ever before.

That’s cool. So, would you say the key new message is that you’re no longer in a “big tanker” or something? You’re now agile, young, and dynamic, helping customers in a new way?

Jürgen

Exactly. I think we’re much more agile and flexible, and we can deliver faster. Software AG helped us grow and mature the product and win big customers around the globe. We have customers like Sony, Hitachi, NTT in Japan, Telstra in Australia, and many others worldwide. We are strong in Europe and have customers in the Middle East like Saudi Aramco and in the US, like Baxter, Flexco, and Hillrom.
This global footprint was something Software AG helped us a lot with. Now, in times of market consolidation, where many startups are disappearing, we’re able to shape our go-to-market, mature the product, build important features, and reach the global scale we have today. We have around 1,000 active customers, with 25 million connected industrial devices—not home IoT, but real industrial IoT like elevators, compressors, hospital beds, wind turbines. Around 3 billion messages are ingested per day into Cumulocity. So, it’s already a sizable scale-up.

Right. And as you’re naming so many customer names, I’m also curious to learn more in a second about concrete use cases and what you’re doing. Just one personal question: You’ve been working in different markets, so how do you see the mechanical and plant engineering market, and what’s happening in your market? What’s your personal experience from the last five years?

Jürgen

We distinguish between smart connected products and smart connected operations. A part of smart connected operations is, of course, Industry 4.0—shop floor management and monitoring, or even in the chemical and process industries. That’s also part of smart connected operations. At Cumulocity, we’ve found our sweet spot with smart connected products. We connect everything—elevators, compressors, wind turbines, hospital beds, cranes, and so on. It’s all nicely built into our product. We also have these “tenant hierarchies”—the platform is designed for a B2B2X go-to-market model. This exactly matches the reality because companies that build powerful machines sell or rent them to the end customer. So, you need this indirect approach, and the platform supports this very well with its tenancy concept.
For smart connected operations, like Industry 4.0, we’ve tried to tackle this on our own, but let’s be honest, it’s a larger portion of the IIoT market. We’re software people—we understand software very well—but we lack deep domain expertise in verticals like oil and gas or manufacturing. You can’t just walk into Daimler as a shop-floor guy and say, “We’ll improve your overall equipment efficiency by 3%.” It’s not that simple. You need years of experience. That’s why we have an indirect go-to-market strategy in these areas. The platform has the necessary capabilities, but we don’t have the deep domain knowledge in these sectors. This is why we partner with others, and the primary partners take the lead—like SAP, IBM, Accenture, Wipro, TCS, or telcos. They lead, and they use our platform under the hood to power their solutions.

So what has changed with your new branding and Cumulocity’s go-to-market strategy? You mentioned a lot of big players, so what about smaller software or hardware companies? How do you partner with them?

Jürgen

Yeah, over the past years, we’ve come to understand the importance of the partner ecosystem. In the past, a lot of partnership messages in the IoT world were just lip service—nice marketing, but not much real movement or success behind it. I think we’ve moved past that stage. Now, with our partners, we want to deliver end-to-end solutions to our customers because that’s where the value is created. The platform is just the middle part, like the patty in a burger, but you need all the things around it. For that reason, we have certified device partners, technology partners, implementation partners, and solution partners, as well as a developer ecosystem for the techies. We’ve invested a lot in this and have a dedicated team for it. Now it’s open, so even smaller partners can easily certify their devices and gateways with Cumulocity. We also nurture the communities, organize hackathons, and have both big and small companies joining. It’s quite powerful. One demo at the Hannover Trade Show impressed me a lot: mfr, a smaller company from Leipzig specializing in field service management. They have a super tight integration with Cumulocity. It’s so seamless to connect the devices and trigger service tasks in mfr, and if you don’t get a dump from Cumulocity with problems related to your asset, they can take it directly to the field service engineer. If the engineer makes changes, it’s reflected back in Cumulocity. It was very impressive.
That’s an example of a smaller partner. But we also have SAP as a bigger partner for asset performance management. We work with both. The community is open for that.

Okay, that’s pretty cool. I’ll include links to these specific solutions in the show notes so listeners can check them out. So would you say that machine builders are also your partners, or would you consider them more as customers, given the B2B2X constellation you mentioned?
Where do you see machine builders? Do you help them as customers, or are they partners too?

Jürgen

They are also partners. It’s a partner-customer mix, I would say. We even partner with our customers, right? We don’t distinguish between “hey, you’re a customer” and “you’re a partner.” It’s all a learning journey. With use cases, in the beginning, for basic things like device management, it’s more of a vendor-customer relationship. But as you move on to things like data products or higher-value digital services, the relationship becomes more about being a trusted advisor. That’s where you really feel the partnership, rather than just the vendor-customer dynamic.

Okay. I want to ask a bit more about the use cases. You mentioned a few of them. Do you have one or two examples of your top use cases, where you’re implementing your solutions with customers?

Jürgen

Absolutely. We looked at our large customer base and categorized the use cases. In the smart connected product space, there are certain categories of use cases. The one that everyone starts with is remote monitoring and remote management. A good example here is Enercon, a large wind turbine manufacturer in Europe.
They’ve connected 30,000 wind turbines to Cumulocity, generating 55 gigawatts of power. Initially, they tried to build the solution on their own but failed twice, which they shared at one of our user groups. They then moved to Cumulocity and, in six months, onboarded the 10,000 turbines they had at the time. Now, they’ve scaled up to 30,000 turbines.
By using Cumulocity, their operations team is now 30% more efficient because they no longer need to switch between different tools. They have a single pane of glass, which allows them to focus on their work—improving wind turbine operations and solving issues.
The next level after remote management and monitoring is what we call field service management. This is when you need to fix things in the field and coordinate your field force. A customer in the US, Flexco, is an example. Flexco operates in the conveyor belt business, with belts in places like mines, airports, and other spaces. They had a custom solution beforehand, and we migrated them to a global rollout in four months. They retrofitted their conveyor belts with a device they developed and connected it to Cumulocity. If there’s an issue, for example, with the cleaners or a build-up at one of the critical points, Flexco can send a service technician immediately. Previously, they had a fixed maintenance schedule, sending people around at pre-scheduled times to check for potential defects. This system was less than optimal, as sometimes technicians would arrive to find everything was running fine, or too late, after a breakdown.
With this solution, they’ve shifted from a reactive to a proactive approach. They now receive live data, can see what’s happening, and apply machine learning to predict future issues. This has enabled them to save a significant amount of money. For example, for their customers, they’ve improved product output by up to $1.1 million per year. They have several big customers where they install these conveyor belt solutions. They also avoid lost production time—around $150k per incident—by saving materials and reallocating labor, among other things.

Okay. I imagine this savings comes from reduced service and maintenance costs, and other aspects as well. That’s a significant amount, especially considering the different conveyor belts and equipment in the field. I assume the business case is built around various cost factors, right?

Jürgen

Exactly. It’s a set of business cases depending on the type of conveyor belts, the countries, and so on. They have customers worldwide.

You were talking about how these customers initially tried to build the solutions themselves, and now they’re using Cumulocity. What would you say is the reason behind these companies now cooperating with you, and possibly with hyperscalers as well?

Jürgen

I think it’s because they’re engineers. They are equipment makers and machine builders, and by nature, engineers like to build things themselves. That’s why they often try to build the software part on their own. But many of them don’t have the industrial IoT experience that we have. They have good developers, don’t get me wrong.
What typically happens is they start with a pilot project. They play around with MQTT, maybe a time series database, a nice dashboard, and connect five devices. In the first pilot, this is easy, and it looks good. They present it to the board, and the board says, “Fine, go ahead and connect our fleet.” But the fleet is much bigger than five devices. It might be millions of devices, spread around the globe.
Then they need to think about all the enterprise capabilities: high availability, security, certificates, multi-tenancy, and global scalability. That requires more and more resources and time to implement. And then they realize that this is not just a project; it needs to run indefinitely and improve over time. It’s a product with a lifecycle that needs support and maintenance. That’s another big investment.
What happens then is that the team grows and grows, the investment increases, and the total cost of ownership (TCO) explodes. The board says, “Whoa, whoa, whoa. This isn’t what we expected.” At that point, they are in a good position to understand how Cumulocity can help. We do the heavy lifting. We provide the platform with all the enterprise capabilities, robustness, and scale they need globally.
This allows their developers to focus on the business logic and the intellectual property (IP) that really matters for their customers. If their team is just playing around with MQTT or Kubernetes, that might be fun, but it doesn’t add much value for their customers. They want to create value for their customers.
So, their R&D and development teams should focus on what matters on top of Cumulocity, because that’s where the value is created. This is what we call a “buy and build” approach, and this is how we help them. It takes time for them to learn this, and it’s a hard lesson, but we see these migration projects quite often.

Okay. And maybe to make that a bit concrete from your solution offering: When I look at the entire IoT tech stack, some parties in the market might compare you with the hyperscalers, but that’s not really the point here, right? Because you’re offering the IoT platform. So, as I understand it, you provide Platform as a Service (PaaS) or Software as a Service (SaaS) capabilities, and the infrastructure might not matter because I could set up on a hyperscaler like AWS or Azure and then build my services on top with you.
So, Cumulocity is offering not just the IoT platform, but also IT/OT integration components. For example, you have thin-edge.io, but also components for data visualization and data analysis, like streaming analytics and so on. Can you explain the tech stack a bit more? What do customers use from you, and what do they typically use from a hyperscaler? Just to make sure I understand it correctly from a practical standpoint.

Jürgen

So, we are hosted in the cloud. We’re running on hyperscaler tech, so we’re infrastructure-agnostic. We run on AWS, Azure, Alibaba, Tencent, and Google. We also have customers with global fleets who say, “Can you run on Azure in Europe and on Alibaba in China?” and they want to run the same application without redeveloping it. So, we sit on top of the hyperscalers. Cumulocity is really aimed at accelerating time to market for your digital and IoT solutions. We provide a lot of out-of-the-box features, so you don’t need to build everything using hyperscaler services. Of course, you can build everything on the hyperscalers, as they offer hundreds of services, but that would require a lot of development effort. You’d need to wire their services together and build many things yourself. With Cumulocity, for example, we have a device management application out of the box. It handles things like firmware and software management, configuration management, and remote access—all with just configuration, no coding required. You can start using it tomorrow. Regarding SynEdge, which you mentioned earlier, it’s an open-source framework running at the edge. It’s a lightweight and highly secure edge framework developed in Rust, allowing us to easily bring device data into the cloud. We also support edge deployments—what we call thick edge—if you want to run on an industrial PC or edge server, which is fully supported in combination with the cloud, even in air-gapped environments. In the cloud, as I mentioned, we start with device management because that’s the top use case. We also offer a cockpit for dashboarding and built-in streaming analytics. Additionally, we integrate with the hyperscalers. For example, we have a data hub that allows you to connect to hyperscaler data lakes, like S3 or ADLS. We can also utilize the BI and ML tools from the hyperscalers, or open-source tools, depending on what the customer prefers. Through our data app, it’s easy to integrate Power BI or a data scientist notebook for processing the data. We can even take machine learning models and bring them back to the edge. So, we also cover the MLOps part within the platform.  We focus on three key areas: device management, IoT data management, and operationalization of analytics. We sit between the OT world (your devices and assets) and the IT world, on top of the hyperscalers. We focus on preparing IoT data because it’s often heterogeneous and noisy. For any kind of analytics, you need high-quality data first. Once you find an anomaly using machine learning, just knowing that something strange happened doesn’t help. You need to take action to avoid it in the future. That’s why we have the link back to the device. The data scientist doesn’t have the connection back to the device, but we do. That’s why we have these three areas: device management, IoT data management, and operationalization of analytics.

Okay, I get that. And you mentioned earlier the project with Enercon. So let’s maybe take that as an example. They’re using their own hardware to connect 30,000 wind turbines, for instance—via some kind of gateway, I assume. Would you say that’s a good example of how customers work with you?

Jürgen

Sure, sure. I think Enercon is a great example, because these wind turbines are often in the field for 20 years or more, so it’s not always the latest generation. That’s why we also need to support retrofit solutions. Some of these devices you can’t even touch anymore, so we take a retrofit approach for those. We also provide server-side microservices that allow us to collect data from the turbines. For the newer ones, they use Phoenix Contact gateways to connect the turbines.
What Enercon did is set up a dedicated instance of Cumulocity. We ingest data from the different generations of wind turbines into Cumulocity, and from there, we integrate with their data lake in Azure, as well as various other IT systems—like their SAP Field Service Management, SAP ERP, a weather forecast system, and a custom-built parameter framework for remote parameter settings.
That’s what I meant earlier with “single pane of glass”—it empowers both the product owners who develop the wind turbines and the remote service teams. They all work within a single solution that brings together everything they need.
We have integrations with marketplaces—we’re on the AWS Marketplace, the Azure Marketplace, the Google Cloud Marketplace, and also on the SAP Store. What’s interesting with the hyperscaler marketplaces is that customers can use their commit-to-consume agreements. Typically, large enterprises have long-standing, multi-million-dollar agreements with hyperscalers that span several years. With these marketplaces, customers can now purchase Cumulocity as part of that already approved budget, which is really convenient.
So, for example, if they still have remaining volume on their commitment, they can buy Cumulocity through the marketplace and retire part of their hyperscaler commitment that way. That makes adoption easier.
With SAP, it’s a bit different. SAP has embedded Cumulocity as part of their Asset Performance Management solution. You might remember Leonardo, which SAP had in the past—that’s been deprecated. Now SAP has launched a new version of Asset Performance Management, powered by Cumulocity. We’re really proud of that. You can see it showcased at Sapphire, and there are also videos available—we can share a link.
In this case, we’re really embedded as an OEM, and we’re seeing strong traction in the SAP ecosystem for IoT solutions powered by Cumulocity.
And beyond that, many partners white-label our platform. You might’ve heard about this from Bernd in earlier conversations. Companies like Deutsche Telekom are white-labelling us in Germany, Orange in France, A1 in Austria, KPN in the Netherlands, NTT in Japan, or Telstra in Australia. That’s a really nice scaling lever for us, especially with big partners like SAP who have massive sales forces and take us to market under the hood.

Yeah, that’s really nice. I’ll include the link to the Asset Performance Management solution in the show notes. You also mentioned earlier that you have a user group or a community you’re working with. So for anyone listening who wants to go deeper into the Cumulocity service offering, is there a way to join this user group, get in touch with other users, and learn more?

Jürgen

Definitely. You can find it on our website—there’s a tech community. Just join that, and we also use that channel to send out invites for hackathons and other community events.

Yeah, nice. So please get in touch there—and also with Jürgen. You already talked a bit about the different solutions Enercon is using, but could you tell us more about the building blocks behind your offering? Just to explain the solutions and products a bit more in depth.

Jürgen

Let’s use the three areas I mentioned earlier. On the device connectivity and device management side, we have our open-source thin edge—thin-edge.io. You can read all about that framework online. It’s designed to bring data into the cloud, and not just to Cumulocity—it also works with AWS and Azure. It’s an open framework, but of course, it works best with Cumulocity. On the cloud side, Cumulocity’s device management is plug and play.
And yes, I can confidently say we have the most advanced device management on the planet. I know how many features we have, and we invest a lot to stay ahead—things like A/B firmware updates, delta updates, and so on. We now also have embedded certificate lifecycle management directly in the platform.
It’s a true SaaS platform, multi-tenant, with tenant hierarchies and white labeling supported out of the box. It also works in combination with the edge. In addition to thin edge, we also have what we call thick edge—that’s a single-node Cumulocity setup. So you can develop an application once and deploy it anywhere—in the cloud (on any cloud, since we’re cloud-agnostic), or at the edge. Some of our customers in sectors like defense or manufacturing run these edge deployments even air-gapped, with no connectivity to the cloud.
We support a wide range of protocols, as you’d expect: OPC UA, MQTT, and for metering use cases, things like LwM2M, LoRa, Modbus, CAN bus, and others. Internally, we have a messaging service for intra-platform communication and routing, and a data broker that allows you to share data across tenants and platforms.
On the data storage and data management side, we have an operational store that’s highly scalable, fast, and flexible. Then there’s our DataHub component that enables integration with data lakes, BI tools, and ML frameworks. Cumulocity is an API-first platform, meaning everything is accessible and automatable via APIs—and they’re extremely stable. If you look back over the last ten years, we’ve barely changed them, which is important—both southbound and northbound.
When it comes to data processing and analytics, you already mentioned streaming analytics. We support a low-code approach with drag-and-drop interfaces, but also allow full coding. We’re investing heavily in data preparation—so, data mapping, normalization, and contextualization. That’s becoming a really hot topic. You want to model assets—like a wind turbine, which is a fairly complex system—or even a building, and then flexibly map sensor and device data to those assets.
We do that with SAP, for example, in their Asset Performance Management solution. That’s for mission-critical, high-value equipment like gas turbines. SAP models those assets in their ERP system. But guess what? What’s modeled in SAP doesn’t always match the real-world setup—and that’s totally normal. So, you need a way to map the physical world to the logical model. For that, we have a component called the Digital Twin Manager.
It lets you either take existing models, like those from SAP, or create your own. Then you can map real-world data to those modeled assets. And that’s incredibly powerful—because for analytics, you need to create the right views on your data.

Right, cool. So a really wide range of different services and end-to-end solutions that can be used for specific use cases or to roll out a project. I’d happily invite everyone to check out your website and see more details there. Maybe one last question for today—I want to focus a bit on what you’re curious to see in the future. I’m not sure if you’re already working on things like data preparation with AI based on IoT data. So, what’s your internal roadmap when it comes to AI?

Jürgen

Yeah, AI is definitely a super important topic. You could already see that reflected in our branding at Hannover Messe—we called it AIoT, meaning AI meets IoT. We have several projects already running. For example, we’re working with Sony on Vision AI. In these cases, we run models at the edge. The process involves training the models in the cloud, bringing them to the edge, and running them locally. One example is using a light sensor in a factory—if the lighting conditions are bad, we use a different model than when the light is good. So, we dynamically switch between models depending on the environment.
We also have exciting projects using GenAI. For example, we trained a model on product documentation for knitting machines together with our customer and partner Mayer & Cie.
That really helps service technicians, because nobody wants to read thick manuals. If you get an error code like 7411 on a knitting machine, GenAI can help by explaining what it means, guiding you through the root cause, and even suggesting which spare parts you’ll need.
So, AI is definitely high on our agenda, and we’re investing a lot in it. We’ve already seen some strong validation in early use cases. But what’s also becoming clearer is just how crucial data preparation is. If your data is poor quality, even the best model won’t help—garbage in, garbage out. You need clean, high-quality data and, importantly, contextual knowledge. You need to know, for example, that this wind turbine is part of the same wind farm as that one. That kind of relationship isn’t obvious in raw data.
That’s why I believe the combination of data preparation, contextualization, and AI is what really makes the difference.

Okay, that’s cool. I was going to ask about your best practices with edge AI and AIoT, and you’ve already mentioned that it’s all about the data. But do you have more advice for someone setting up edge AI use cases in the coming years? Any key learnings from your projects so far? Or would you say things are still mostly at the proof-of-concept stage?

Jürgen

I don’t think it’s still in the proof-of-concept stage. Of course, the super fancy things like agentic AI still need to be evaluated. But the general direction is clear—AI and IoT are incredibly important toolsets to enhance use cases for our customers. At Cumulocity, we follow a co-innovation approach, and we already have mature machine learning use cases that are deployed and delivering real value. These are not pilots anymore.
If you want to dive deeper—because you know these IoT use cases very well, Madeleine—they are super heterogeneous, and it’s the same with AI. You always need to fine-tune your models to the specific use case. That has to be done in collaboration with the customer. So if you’re interested in AIoT use cases, our co-innovation approach means we say, “Let’s explore this together.” We bring our software expertise, and you bring your domain knowledge, and then we see what’s possible.
Most customers right now are focused on efficiency improvements. If you look at the business case side of things, the majority of use cases currently being deployed are about operational efficiency. I believe it’s going to move toward data products and maybe even equipment-as-a-service or other high-value digital services—but only a few percent are there yet. Still, AI already plays a significant role in improving operational efficiency.

Okay, that’s cool. So are you saying that these co-innovation projects are kind of co-funded? That you’re also investing into the future and working together with partners on these cases?

Jürgen

Exactly. That’s how we run them. Because we’re learning what’s needed in the platform versus what’s specific to the customer. At the same time, the customer gains an innovation advantage, because they’re the first to have this new capability in their product—and that helps them differentiate. So it’s a win-win. That’s why we usually approach it with a co-funded methodology.

Okay, that’s cool. So thanks so far—I have so many more questions in mind, but I guess that could be something for a follow-up in another podcast episode. And if you have any questions, feel free to reach out to the Cumulocity team. It’s also really cool that you’re co-funding these new projects—that’s a great approach.
So, for this session: thank you for your time! It was really nice to learn more about your offering, and also to dive into the business cases you’re focusing on—from smart connected devices to different customer types and the broad market you’re working in. I really got a clear picture of what you offer.
From my side, just: very grateful to have had you here. Thanks again for your time!

Jürgen

Thank you, Madeleine. Maybe just a quick summary from my side, if I may. What you should take away from this episode is that Cumulocity is now an independent company. We are one of the leading IIoT platforms, with many years of experience in industrial IoT projects—new projects as well as many migrations we’ve seen over the past decade.
We have a strong partner ecosystem to deliver end-to-end solutions. And you can count on us not just for our SaaS offering or platform capabilities, but also for bringing in the right partners—whether it’s for retrofit or greenfield projects. We work with partners to get the data in and to build or extend the solution, because we can’t and don’t want to do everything ourselves. We’re not the best AI company—there are better ones out there. We’re not the best field service company—again, others do that better. But integration is where our strength lies.
If this sparked your interest, check out our website, start a free trial—it’s fully automated and up in two minutes so you can explore the product. Or just reach out to us, or join our tech community.
Thanks so much, Madeleine, for having me. And thanks to everyone for listening—I hope to see you again soon.

Thank you so much—and have a nice rest of the week. Bye!