In the rapidly evolving automotive landscape, we recognize that user experience (UX) has become a critical factor in shaping the next generation of vehicles. As we integrate electrification, connectivity, and autonomous technologies, we are transitioning cars from mere mechanical machines to connected digital devices. This shift compels us to prioritize UX, blending physical and digital elements to create seamless, intuitive user experiences. As cars increasingly resemble smartphones on wheels, our focus on usability, trust, and delight will define the future of mobility, making UX a key differentiator in the industry.

In our conversation with Casper Kessels, Chief Product Officer at Snapp Automotive, we gained valuable insights into the challenges and innovations in automotive UX design. He shared how the company is enhancing Android Automotive OS to ensure seamless user experiences across various vehicle types, with an emphasis on simplicity and safety. He also discussed the balance between screen use and physical buttons, the role of AI, and how SnappOS adapts to different branding needs. Sit back and enjoy the interview!

Hi Casper! Could you explain what Snapp Automotive does and the specific problem your company is aiming to solve?

Snapp Automotive is a company dedicated to Android Automotive, which is a relatively new platform being used in most future infotainment systems. We’re a spin-off from a larger development company that focused on mobile apps and related areas. We have a long history of working with Android and have collaborated with a few car makers. During that time, we identified the challenges they face with Android, which led us to spin off Snapp Automotive, a company entirely focused on Android Automotive. We continue to work with major OEMs on their future infotainment systems. Additionally, we have decided to develop our white-label infotainment system aimed at lower-volume car companies.

There are two problems we are trying to solve. First, there’s the service side of the company, where we help car makers get the most out of Android Automotive. As they transition to Android, many have no prior experience with it, and that’s where we come in. We assist them in the early stages of their projects, guiding them through the process of setting up their infotainment systems. The second aspect focuses on providing a product for car makers who lack the time or experience to build their own infotainment systems in-house. That’s why we offer a fully customizable infotainment system, allowing them to bring a completely bespoke solution to market in a matter of weeks instead of years.

Could you walk us through the unique features of SnappOS and how it differs from stock Android Automotive OS (AAOS) and Apple CarPlay?

What SnappOS does is take the Android open-source project and modify it or build on top of it. While Android Automotive is open source, the codebase can be quite messy and lacks many features. We take that and turn it into a full-fledged infotainment system. Our focus isn’t so much on individual features, but rather on building a coherent user experience. We consider how people interact with the car, how to make navigation easier, and what services we want to integrate.

Generally, when we work with a car maker, we do not include any third-party services in our product. We provide just the infotainment system layer and then collaborate with the carmaker to see which services they want. This could include a few native apps, such as a navigation app and a music app, or integration with an app store. At the very least, we handle all of the software on the vehicle. When it comes to media navigation, we can hand it off to Apple CarPlay or Android Auto. So, when comparing it to Apple CarPlay, we function as the layer below it. This means companies can still choose to have Apple CarPlay in their system if users want it.

What are the biggest challenges in creating a seamless and intuitive infotainment experience, and how does your team at Snapp Automotive tackle them?

SnappOS is a complex piece of software, and the biggest challenge we face is the many unknowns involved. Since we are developing a white-label product, it is not tailored to a specific car brand. Therefore, we must design it to accommodate all the different car brands and the various screen layouts that exist, as well as the different displays available inside cars and ensure that every car maker has the appropriate branding, among other considerations. This is purely a design challenge, and, of course, there are also numerous technical challenges.

As I mentioned before, integrating third-party services is always difficult. This arises not so much from a technical standpoint but from a business perspective. While there is no doubt that Android Automotive, with Google’s services included, is the best system out there, Google tends to overlook smaller brands, which makes it impossible for us to offer that solution. So, the question we face is how to create an experience that is just as good without those services.

There are major car manufacturers that use Android but choose not to include Google services, and these are often business decisions.  It could be that it’s too expensive, or they simply don’t want to rely on Google for everything. They might also have existing partnerships with other mapping or music providers that they want to incorporate into their cars. There can be many reasons behind those choices. But as a designer, I think that for a great user experience, Google Automotive Services is probably the way to go.

There is a well-known argument that having large screens in cars can be very distracting, and safety is paramount in automotive design. How do you ensure that your UX design enhances driver safety without compromising functionality and entertainment at the same time?

Yeah, it’s not the size of the screen that matters; it’s what you display on it, right? You can have a very big screen that isn’t distracting. From the start, we took a different path than most car makers with our infotainment system, focusing entirely on reducing distraction and simplicity. We don’t really care too much about having the latest features or new innovations; we care about providing a seamless user experience. 

For example, early on, when we were figuring out the first version of the product, we encountered the issue that there were no standards out there for button sizes and other elements. So, we basically created our own internal standards for touch sizes and how many clicks it takes to reach certain settings. It comes with its own set of challenges, like the fact big buttons aren’t very trendy, so making them visually appealing is an interesting problem. We focus a lot on simplicity. So, touch screens aren’t the problem per se, but if you don’t optimize for them, they can become very distracting very quickly.

That’s why we focus a lot on making everything as easy to reach as possible, keeping all functions within a certain number of clicks. For example, we’re currently experimenting with detaching the touch target from the button. Let’s say you reach into the top right corner; visually, it may look like you’re pressing a button, but you’re actually just at the edge. We make these touch targets bigger or move them so you’ll always hit the button accurately. Of course, there are limits when we work with a car maker who wants to do everything on a touchscreen, as we don’t get to influence that aspect. I would prefer if climate controls, volume, and similar functions were physical buttons, but we don’t have control over that. So, we try to make those digital controls as easy to use as possible.

Many people still prefer having physical buttons for some controls, but maybe not for all of them. We try to inform the automakers about this preference. Most of the car makers and major OEMs we work with are also aware of the issue, and it often comes down to a cost-benefit analysis. They recognize that physical buttons can be easier to use in certain cases, like a start button, but it’s just expensive to specify and maintain them. We also work a lot with startup car makers who are operating on very tight budgets and deadlines. For them, if they have the choice between spacing for one hardware display or putting everything on the software side, they often opt for the latter. Finding suppliers for buttons and coordinating with them is an additional job they would need to hire for, so the choice is obvious.

If screens are the future, in what ways do you see the automotive industry evolving in terms of user experience, and how is Snapp Automotive preparing for these changes?

Yeah, I hope screens are not the future. I think many carmakers have reached peak screen. I’m sure there are internal discussions about moving some of those controls back to physical buttons because they also realize that, while cost is one factor, customer satisfaction is also important. You can even see this at the highest level, where many hypercar and supercar companies are returning to analog gauges and buttons because it provides a much more luxurious feel. 

It’s similar to having a nice analogue watch versus a digital watch; they are different experiences. In the long term, this aspect will always remain in cars, and there will always be touchscreens in vehicles because, for many use cases, they offer the least problematic interaction. Touchscreens are also the best option for designing complex features like navigation, so it will be interesting to see what happens next.

The future seems to be more about platform competition between mobile mirroring systems like Android Auto and Apple CarPlay versus native experiences like Android Automotive. Design-wise, I think we are at a stage similar to the early days of Android phones, where many different manufacturers explored various crazy features, such as different sizes, form factors like foldable, and their skins on top of Android, etc. Now, there’s been some consolidation—of course, there are differences between a Samsung and a Pixel phone, but they all generally have the same gestures and interaction patterns. I believe we’ll get there eventually, especially when looking at Chinese brands, as their infotainment systems are mostly the same and share similar interaction patterns. Something similar will likely happen in the automotive industry.

For us, it’s not a threat at all. We are very agnostic. If you look at mobile mirroring, it will always need to run on top of an existing system, at least for the foreseeable future. For carmakers, it’s also a significant decision. They will likely have a solid Android system and then offer mobile mirroring as well. Our goal is to enhance the quality of both the Android side and our automotive side, bringing more simplicity and ease of use to the car industry.

Can you explain how emerging technologies, such as AI and machine learning, are shaping the future of automotive UX design at Snapp Automotive?

Yeah, we are more reserved in those kinds of things. The car industry often does a lot of innovation for the sake of innovation. In a very saturated market like China, where many infotainment systems look similar, it’s a real competitive advantage to bring these kinds of features and convey to journalists, "Wow, we can do this. It’s very exciting!” We do a lot of internal experimentation, but we always ask the question: What problem does this solve for the driver? We’re very careful about adding features to an infotainment system because there is a common belief that more features make a better product, but that’s not true. It’s the few features you have that must be very good.

For example, it's hard to overlook large language models because they are currently very hyped, and for good reason. Instead of just adding a web wrapper around ChatGPT in our product and saying, “Look, we integrated ChatGPT,” we ask the question: What can you do with a large language model that solves a problem for a driver or a passenger? We are conducting internal experiments on how to deeply integrate a large language model with the car's functions. Can it replace a typical voice assistant? How can you optimize route guidance? There are many exciting applications beyond just using it as a fact machine, which often doesn't provide the right information half the time.

As companies like Ferrari are phasing out their in-house navigation systems in favor of third-party solutions like Apple and Google, do you foresee this becoming a widespread trend in the automotive industry, and how is Snapp Automotive preparing to adapt to this shift?

Yeah, so this is basically a competition between the platforms I mentioned earlier. Ferrari said they would hand over all navigation and media to Apple CarPlay, so they don't have to pay for these services. I assume 95% of all Ferrari customers are iPhone users, so they can do that. Of course, for a larger brand with more budget vehicles, where you will have more Android customers, you cannot anticipate that. I think, in this slide, they’re talking about the next version of CarPlay, which can take over most of the car's functions. It will look like a much more integrated solution than what you can find in the current generation of Ferrari vehicles, where they just mirror Apple CarPlay in the instrument cluster. While that’s not a great experience, it aims toward a future where they can offer more.

As a carmaker, you have to ask yourself what you want to do next. Some brands will go with Apple CarPlay and build a fully integrated solution with it. Customers will only interact with the CarPlay software, but you always have Android customers, and you risk alienating them if you don’t have an alternative. Currently, Google's equivalent to Apple CarPlay is not Android Auto but Android Automotive with Google services. That’s how they earn money. It will be interesting to see whether Google stays with this native Android Automotive solution or upgrades Android Auto to a model similar to Apple CarPlay, where you use mobile mirroring and the underlying layer isn’t present anymore.

For us, we don’t really see it as a threat. We would be happy to integrate the new Apple CarPlay. We are not solely a software company for Android Automotive. We do full integration and we’ll have a hardware offering as well. Car companies come to us for infotainment needs that include hardware, software, and everything in between. If we can include Apple CarPlay, that would be great, but Apple CarPlay doesn’t currently cater to smaller players — maybe in the future. We’re betting on both horses and in the end, we just want to deliver the best user experience. Almost all carmakers are now switching to Android, and in a few years, only a handful of car companies will run on Linux or something else. 

The real question is what comes next: will they create a totally bespoke experience without any Google features, or will they give up the car's infotainment system to Google and run Google apps? Those are the two routes they can take. The route of not going with Google reflects what you saw in the past, where there were various platforms with totally different designs. However, there’s a lot more complexity at play now, so it’s hard to say for sure if it will have the same fate as mobile platforms. We’ll see.

White-labeling can be a great challenge when catering to a diverse range of clients with distinct branding. How do you build a personalized version of SnappOS for your client and meet branding needs?

Yes, this has been one of our main challenges over the past two years. At the core of our product is our theming engine. We designed a full infotainment system from scratch, taking into account all the different displays and layouts that exist. When we port our operating system to a completely new display layout, it automatically adapts without breaking the interaction flow. That's the first part. Anything after that can be fully customized. This includes the services you want in the car, but more importantly, it involves the user interface. We built our own theming engine that is closely tied to our Figma design system. That’s why, in a matter of days, we can go from a neutral SnapOS theme to a fully bespoke theme.

Our customization goes beyond just colours and fonts. Everything is attached to our theming engine and is customizable, ranging from map styles and specific logic to certain components that we can swap in and out. This flexibility allows us to create a fully bespoke system in mere days. We've done a lot of work on it, and it's been both exciting and challenging. We're also sharing insights on our company blog about how we built it and how we can use it to our advantage. For example, if we work with an off-road truck, we want a completely different UI than if we work with a small city car or a performance hypercar.

This adaptability applies to both single-screen and multiple-screen layouts. Our system can cover not just the main unit but also the instrument cluster, passenger displays, or any other displays within the vehicle. From luxury vehicles with displays in the back or sports cars with extra passenger displays to show speed, we touch both the hardware and software of the infotainment system. There are special cases where the responsibility is divided. For example, when a car maker is looking for a sound simulation, the design and playback of sounds through the speakers are typically not within our scope and fall under the OEM's responsibility. However, we do manage how to set it up within the infotainment system, including the desired settings and optimal design to match different driving modes and other things.

Wants and needs of users and automotive manufacturers can be different from each other. How do you decide which features to build and deploy? Do you collect feedback from your users?

Yes, when we started out, this was a significant challenge. On one hand, we didn’t know the size of the displays, and on the other, we were unsure who our customers would be. So, where do you start designing? We quickly discovered that when you’re driving, it doesn’t matter if you’re in a sports car, a city car, or a luxury vehicle. During driving, you only need three or four key features. Nine out of ten use cases revolve around navigation and media. Sometimes, you might interact with your phone, make phone calls, or adjust certain vehicle settings, like changing drive modes. Regardless of the vehicle type, this is where we began, focusing on integrating these core features as well as possible and building a platform that allows us to add more stuff without breaking the interaction model.

What we currently have is a base set of features in our infotainment system, including settings, charging apps, and more. These can be fully customized and themed. For example, if we work with an off-road vehicle company that wants an app for customizing drive settings, displaying the vehicle's angle of rotation, or creating a nice animation showing how the four-wheel drive system works, we can integrate that easily. In the end, it’s all about the power of Android. It’s essentially about developing an app and running it inside the infotainment system, whether it’s in the instrument cluster or anywhere else. Thanks to our extensive library of components, it is straightforward to create custom features depending on the specific customer we are working with.

We work with many startup car companies that have niche vehicles, such as dedicated off-road trucks or shared cars. These companies typically have a set of customers they are piloting their vehicles with. We run our software in their cars, allowing us to observe how customers interact with it and what custom features they desire. Currently, we don’t have a system to collect user feedback because we haven't had the need. However, if we were to work with a client that has a large fleet and less direct access to customers, gathering feedback would be essential. I think the difficulty for a large OEM is that, for example, if you're based in a city in Germany, how do you know how a customer in South America or China interacts with their car? Although we don’t have a general feedback collection method yet, our customers often have dedicated sets of users that we can interact with directly, making it easier to gather insights.

My last question comes from our student community: Is automotive UX design fun? What are your thoughts on automotive UX design as a career choice, especially considering its growing appeal among students?

Automotive UX design is super fun. I think designing interfaces for cars comes with a unique set of challenges that you won't find in many other fields. You’re dealing with a high level of technical complexity and designing an actual operating system rather than just an app. On one side, there are apps within the system that you design; on the other side, there are companion apps. Essentially, you cover all the different digital products you can. Plus, you're designing for a physical product that needs to go on the street, and you have to sit in it to interact with it, which adds another layer of complexity. The driver distraction and safety aspects are also quite unique, making automotive UX design different from other UX fields. That makes it very interesting to me.

Considering that automotive UX design is a growing focus for carmakers, it’s becoming increasingly important to find the right employees to design these interfaces. My advice to students would be to recognize that carmakers often have a certain approach to designing interfaces. They can be, without being offensive, somewhat immature in this area because they are manufacturing companies that may have never done anything like this before. Coming in with an outside perspective allows you to provide significantly more value to a carmaker than if you were to start directly with them. They need assistance with the right processes and a proper understanding of the customer experience.