The car is another device in the customer’s ecosystem
Remember infotainment from 2015? Static menus, slow response times, and maps that were outdated before you even left the dealership. Now look at today’s cars with Android Automotive OS. You download apps directly to your car, the voice assistant understands context, and the cloud remembers your preferences across devices.
This isn’t just a technological upgrade. It’s a completely different business model.
A software-defined vehicle (SDV) means the car’s functions, behavior, and user experience are primarily determined by software rather than by hardware. The car becomes a platform, like a smartphone. It can update, learn, and offer new services months or years after purchase.
For IT departments, this is an interesting architectural challenge. For businesses, however, it is something bigger. It is an opportunity to build digital services with recurring revenues around the vehicle. It is also a chance to differentiate themselves from the competition in a way that hardware alone cannot.
Three waves of digitization in cars
Let’s break it down chronologically. Digitization in the car cabin has gone through three distinct phases, each bringing different possibilities and limitations.
The first wave came with phone mirroring. Android Auto and Apple CarPlay made it possible to quickly get apps into the car without having to develop your own infotainment system. Users loved it because they had their favorite apps in their cars. Car manufacturers welcomed this because they didn’t have to invest in their own ecosystem. The catch? All the value remained with Google and Apple. The car manufacturer was just a display provider.
The second wave connected the car, mobile phone, and cloud. Native infotainment systems were complemented by mobile apps for remote car control and backend services. Drivers could lock their cars from their phones, check the battery status of their electric cars, or plan a route that was automatically transferred to the navigation system. This created the basis for collecting data and understanding how people actually use cars. At eMan, we worked on the ŠKODA OneApp project, which connects vehicles with mobile phones and cloud services.
Most traditional car manufacturers are currently somewhere between the first and second waves. They have a mobile app and collect some data, but digital services are more of an add-on than a strategic priority.
The third wave is SDVs powered by artificial intelligence. Here, infotainment and the digital cockpit become the brain of the personalized experience. The car recognizes you, adapts to your driving style, and offers relevant services at the right moment.
And who is furthest ahead in this? According to the 2025 Wards Intelligence ranking, Tesla still leads overall, but Chinese automakers NIO and Xiaomi have overtaken it in technological innovation. It is noteworthy that 62 percent of the automakers in the top SDV categories are from China or Chinese-owned. Traditional Western manufacturers have some catching up to do. BMW unveiled Operating System X with an intelligent assistant at CES 2025. Mercedes-Benz has integrated ChatGPT into more than 900,000 vehicles. However, the lead of companies that have had software in their DNA from the beginning is noticeable.
Why is this so important? The third wave is not just about a better user interface. It’s about a whole new economy of digital products in the car.
Where AI comes into play
Generative AI and machine learning are opening up possibilities in the car that were the stuff of science fiction just a few years ago. Let’s look at three areas where this really works.
A personalized in-cabin experience is the first of these. The car remembers that on Monday mornings, you drive to the office and listen to the news, while on Friday evenings, you head to the cottage and prefer music. The system adapts not only the content but also the interface. When it detects stress from the traffic situation, it automatically mutes notifications and adjusts the air conditioning. This is no longer a futuristic vision; these features are being tested in production cars.
But many companies don’t realize there is a lot of work behind such a system. It’s not just about the AI model. You have to collect data from various sensors in the car and process it. You have to decide what to process directly in the car and what to process in the cloud. And finally, the hardest part: designing a user interface that presents everything to the driver in a meaningful way without distracting them. At eMan, we have experience with real-time sensor data processing from industrial projects, and we know that this is not a trivial task.
Predictive maintenance and service is the second area. Combining sensor data (temperature, pressure, vibration, driving style) with service history enables predicting when a component is nearing the end of its service life. Instead of rigid service intervals, you receive notifications when it’s appropriate. For fleet customers, this is crucial because it reduces downtime and maintenance costs.
New revenue streams are the third area. Monthly subscriptions for advanced assistance features, dynamic service packages based on the current situation, and B2B data products for insurance companies or fleet managers. Tesla offers its Full Self-Driving system as a monthly subscription for $99 (as of February 2026, this feature is now available exclusively on a subscription basis). BMW experimented with a subscription for seat heating (and backed down after criticism, which is also a useful lesson in what customers will accept).
The economics of SDV: How to think about the return on digital features
The old model was simple. You deliver infotainment, collect a one-time payment, and then wait to see if the customer will buy a new car from you in five years. The new model is more complex, but also more interesting.
Are subscription-based models even worth it? The numbers show that they are, if you do it right. Analysts at IDTechEx predict that revenue from digital features in cars will grow at a 30-34% annual rate until 2035. The SDV platform market is expected to generate over $755 billion by 2029.
What metrics make sense to track? The feature activation rate tells you how many customers actually turn on digital services. Churn rate tells you whether the perceived value matches the price. If half of your customers cancel their subscription after three months, you have a problem. Average revenue per user (ARPU) measures how much you earn from a single user over a certain period. And release velocity tracks how quickly you can deliver improvements.
This is where the rubber meets the road. If your architecture prevents you from quickly deploying new features via over-the-air (OTA) updates, that’s bad. Competitors who can do this will overtake you. That’s why modular architecture is so important.
Connect this to what applies in other industries. Companies that treat software as a one-off project rather than an ongoing product incur a higher total cost of ownership (TCO) and take longer to bring new features to market (time-to-market). This is doubly true in the automotive industry, because a car’s life cycle is long and customers expect it not to be obsolete in three years.
What is typically underestimated in practice
Many manufacturers solve this by starting with the technology and only then thinking about the users. However, designing interfaces for different markets is more complicated than it seems. What works in Germany may not work in China or the US. Language variations, cultural preferences, and regulatory requirements. All of this needs to be addressed from the outset.
Response speed is another “popular” challenge. A voice assistant that responds with a three-second delay is useless. Drivers will not use it. Edge computing, i.e., processing data directly in the vehicle instead of sending it to the cloud, helps. However, this requires more powerful hardware in the car and a well-thought-out architecture. You have to decide what will run locally and what will run remotely.
Data flows are also often underestimated. Sensors generate huge amounts of data. Without a well-thought-out strategy for what to collect, how to aggregate it, and where to send it, your cloud infrastructure costs will quickly get out of hand. And it’s even worse when you find out that you’re collecting the wrong data and have to start over.
Security is a topic in itself. A car connected to the internet is a potential target for attack. Manufacturers must address not only data protection but also functional safety. What happens if someone attacks the braking system? Standards such as ISO 26262 for functional safety and ISO/SAE 21434 for cybersecurity set uniform rules of the game in this regard.
And then there is one decision that usually determines the success or failure of the entire project. The choice of architecture at the beginning. If you build a cumbersome system that prevents individual components from being updated independently, you will be paying off technical debt in 2 years. Investing in modularity will definitely pay off.
How to get started: three steps for automotive management
So what to do about it? Try the following procedure.
Start by auditing your existing digital map. Map all digital customer touchpoints. Mobile apps, web portals, and infotainment data flow between them. You will often find that different departments operate isolated systems that do not communicate with each other. This is the first opportunity for improvement.
Choose a pilot SDV or AI initiative with a measurable impact. You don’t have to redo the entire architecture right away. Start with a personalized interface for a specific customer segment or predictive maintenance for fleet clients. It is important to have clear goals and the ability to measure results.
Find a partner who understands the entire technology stack. SDV projects require a combination of competencies: mobile development, backend systems, cloud infrastructure, working with sensor data, and designing user interfaces for different markets. At eMan, we have experience with infotainment and enterprise systems for brands such as ŠKODA, Volkswagen, Porsche, and Bentley. We can help not only with implementation, but also with architecture and strategy.
The automotive industry is undergoing a transformation comparable in scope to the transition from combustion engines to electromobility. This time, however, it is not about the hardware under the hood, but about software and data. Companies that understand this and act quickly have a chance to define what mobility will look like in the coming years.
Frequently asked questions (FAQ)
What is a software-defined vehicle (SDV)?
A software-defined vehicle is one whose functions, behavior, and user experience are primarily controlled by software rather than hardware. This allows the car to receive remote updates, learn from data, and offer new services even years after purchase.
How is AI changing the user experience in cars?
Artificial intelligence enables personalization of the cabin environment, voice assistants that understand context and natural language, predictive maintenance based on sensor data, and adaptive interfaces that adjust to the driver’s driving style and mood.
What are OTA updates?
OTA (over-the-air) updates are software updates downloaded to the vehicle over the internet. They allow manufacturers to fix bugs, add new features, or improve performance without the need to visit a service center.
What are some examples of digital services in the automotive industry with a subscription model?
Examples include advanced assistance systems (Tesla Full Self-Driving), premium navigation services, extended warranties, in-vehicle internet connectivity, and unlocking the performance features of electric vehicles.
Why is modular architecture important for SDVs?
Modular architecture allows individual system components to be updated independently, new features to be introduced more quickly, and overall software maintenance and development costs to be reduced throughout the vehicle’s lifetime.
Which car manufacturers are furthest ahead in SDV?
According to the 2025 Wards Intelligence ranking, Tesla leads overall, but Chinese companies NIO and Xiaomi have overtaken it in technological innovation. More than 60 percent of SDV leaders are Chinese car manufacturers. Among traditional Western manufacturers, BMW and Volvo are among the most advanced.
