The future of transportation and mobility is autonomous. In this episode, we explore value chains and digital transformation in the transportation industry with investor and author Evangelos Simoudis.
The future of transportation and mobility is autonomous. In this episode, we explore digital transformation in the transportation industry with investor and author Evangelos Simoudis.
Evangelos Simoudis is a seasoned venture investor and senior advisor to global corporations. The author of two books, Evangelos is the founder and managing director of Synapse Partners, a firm that invests in early-stage companies developing applications that combine artificial intelligence with big data. Evangelos earned a Ph.D. in computer science from Brandeis University and a B.S. in electrical engineering from Caltech. His most recent book is Transportation Transformation: How Autonomous Mobility Will Fuel New Value Chains.
- What are Autonomous Mobility Solutions?
- Role of Smart Cities
- Public Transportation and Autonomous Mobility
- Role of Data and Platforms in Autonomous Mobility
- New Business Models for Autonomous Mobility
- Data Collection and Personal Privacy with Autonomous Vehicles
- Advice to Cities in Digital Transformation
This transcript has been lightly edited for length and clarity.
Michael Krigsman: Autonomous vehicles, autonomous transportation, that's our topic today on CXOTalk. Evangelos Simoudis is one of the world's experts on this topic. His new book is called Transportation Transformation. Evangelos, tell us about your background briefly and about your relationship to autonomous mobility.
Evangelos Simoudis: For the past 20 years, I've been a venture investor. The last four of those, my firm Synapse Partners is investing in companies, early-stage companies, that develop enterprise software AI applications. We see mobility as an important area to focus, so a number of our investments are in mobility but not exclusively.
I started writing and the writing went from blogs to books, including the most recent effort that I launched last week with the book Transportation Transformation.
Michael Krigsman: When you talk about autonomous mobility, what are you referring to?
Evangelos Simoudis: I'm referring to the transportation of both passengers but also the delivery of goods using a variety of autonomous vehicles. It may be speed of traffic vehicles like so-called robotaxis that we're seeing from companies like Waymo or Cruise. It may be slower-moving vehicle shuttles like what we're seeing from companies like May Mobility. Or it may also be various forms of robots that deliver goods such as the robots of Nuro that is delivering groceries or those from Starship Technologies. Those are vehicles that can operate either on sidewalks or on the street.
Michael Krigsman: What is the underlying set of factors that are going to drive this? In a way, your book is painting a vision of the future but it's also very much grounded in the realities of what's going on today from a technology standpoint, from economics, and so forth.
Evangelos Simoudis: Over the past few years, we have seen the very strong emergence of on-demand mobility services. Those services have been embraced by large groups of people, particularly in urban areas. They have created, of course, other problems.
We have also seen a lot of efforts by many automakers, mobility services companies, and startups to develop autonomous vehicles that would be used for these types of applications. Finally, we're starting to see cities instrumenting themselves in order to be able to accommodate this future of mobility. So far, these efforts have been very siloed and the future that I am both envisioning and advocating for will require for these efforts to start being integrated a lot more than they are today if we are to address some of the megatrends that are starting to plague us.
Michael Krigsman: When you talk about these megatrends, what do you have in mind? What are the megatrends that are relevant to this?
Evangelos Simoudis: The book is very much about urban mobility as opposed to mobility as a whole. First and foremost, as we look at cities, we look at congestion and pollution in inner cities. If you think of cities like Mumbai, Rio de Janeiro, Sao Paulo, Istanbul, those cities have tremendous traffic problems because of congestion. The average congestion in a city like Sao Paulo is one and hours each way, so about three hours a day you spend on the road just commuting to and from work.
Climate change is also a big megatrend that needs to be addressed, again in the context of urban transportation. You're seeing already today many governments starting to impose more restrictions on the type of vehicles that it will allow on the road and OEMs, automakers are starting to work on that.
Finally, in many places, in many countries, you have aging populations. Japan is a very good example that's often cited. These populations will need a lot more help with their mobility than we are able to provide today through mere public transportation.
Michael Krigsman: What are the pieces that need to be put into place in order for this future to be realized?
Evangelos Simoudis: The efforts that we are seeing are well-intentioned but, because they are siloed, they are starting to produce, in many respects today, the opposite results of what they intended to do. First, there has to be a realization that these companies need to start coming together in collaborating in a meaningful way.
The second thing is that cities need to take the lead in this effort. In taking the lead, some of their efforts will need to deal with instrumenting their infrastructure as opposed to just expanding their transportation infrastructure, as it was the case especially here in the U.S.
I think we need to, in a sense, balance technology with some of the other issues that need to be addressed such as regulation, policy, data rights, and the like.
Michael Krigsman: Why do cities need to lead the way? Why does the burden fall upon cities?
Evangelos Simoudis: I chose urban transportation because, to me, cities are an important space. First of all, it is a small enough space where we can start controlling and making changes in. They present a magnet for population. There are many studies from world bodies that indicate that this urbanization move will continue. Finally, this is where automakers and mobility services companies come together in order to serve the public.
Now, I should also say that when I talk about cities, I don't mean a city in a very restrictive manner. I talk more about metropolitan areas. Here in San Francisco, for example, we have the city of San Francisco, but we have the metropolitan area, the Bay Area. The same thing goes in Los Angeles. The same thing goes to, as I said, many world cities. When we think about urban transportation, I'm implying the systems, processes, and operations that need to be put in place in these broader metropolitan areas as opposed to just a specific city.
Michael Krigsman: What are the challenges that cities will face as they try to implement these systems?
Evangelos Simoudis: In the book, I paint three scenarios for how this can emerge and, obviously, in the short-term, we're also dealing with the aftermath of the pandemic. But the cities will need to understand what type of role they want to play going forward.
In the United States where the public transportation systems are not particularly good and we have much more of a private vehicle ownership culture, this future of transportation may evolve in a way that cities will have to coexist or the public transportation and other systems will need to coexist with privately owned vehicles.
On the middle case, I see Europe where we have strong support for public transportation but, because of its nature, regardless of how much funding a lot of times cities provide to their public transportation system, these scheduled rides provide gaps and inconvenience for the public.
Then the final scenario I see in many cities in Asia, starting with Singapore and several cities in China, that I think have taken a very strong position on transforming their systems and establishing partnerships with mobility service providers in order to provide a more seamless experience that offers mobility as a service.
Michael Krigsman: We have a question from Twitter. Arsalan Khan is asking, "It's obvious that collaboration among cities is going to be essential to this. Is there a central place that cities can collaborate and how should cities go about creating that ecosystem aspect?"
Evangelos Simoudis: First of all, I should clarify that I'm not necessarily advocating that all of the cities need to collaborate together, but within a metropolitan area. Again, if you were to think of the Bay Area as an example, particularly here in the United States, there are over 140 agencies that deal with transportation issues. This reality makes it extremely difficult to try to achieve the type of future that I describe in the book.
The first thing that needs to be simplified is the jurisdictional structure of cities to make it to the point, a single body, a single person, a single organization responsible for bringing all of these together.
The second is areas of competency. Cities will need to acquire certain competencies, particularly around running specific platforms and being more flexible with regards to how their public transportation systems are enhanced through the addition of these types of mobility services that we are describing.
There is a lot of know-how, I believe, that cities can obtain from the industry. That's why I've been talking a lot about public/private collaborations.
I'll give you, actually, a recent example. A couple of days ago, Uber announced that they will be licensing their software platform to the City of Marin, just north of San Francisco, to enable it to better manage how public transportation and mobility services, offered by Uber in this case, will be interwoven better together.
We've seen examples, very successful examples, in Helsinki. We're seeing very good efforts in Berlin. I would actually put Berlin as one of the top cities on how they're starting to approach and implement mobility as a service.
Michael Krigsman: How is it going so far?
Evangelos Simoudis: The first results have been quite encouraging in terms of how people are starting to use these types of applications. But it is not only an issue of using a mobile application. A lot of the efforts, initial efforts, particularly by startups, have been to provide an application that is informational. Then these applications were augmented with payment platforms. Not only do you get information, but you can also have a common place where you can buy tickets or reserve rides.
I think that the next step and really where all of this will come together nicely is when cities are starting to play more the role of the orchestrator, right? You're able to preplan how your mobility options are going to be paired with the daily needs that you have, whether it is commuting to and from work and dropping off the kids at school or dealing with your daily meetings and having to move from place-to-place in order to address them.
I think we need to start thinking much more about how this advanced planning has to be utilized more successfully and then added, too, is the more opportunistic transportation. I have an emergency meeting somewhere that was not planned. Now I can go and have a transaction to pick up transportation means to get me to my destination. I think this orchestration, which will start happening and needs to start happening more and more, will give us the right mobility as a service more.
Michael Krigsman: Given the challenges of achieving this vision that you're describing, what should the various players start doing now?
Evangelos Simoudis: Three basic players that I see here are the automakers and their ecosystem, the mobility services companies, and the cities, the metropolitan areas. We are seeing already different approaches, so we have some automakers taking a very technology-centric approach. We have others that are starting to experiment with different models of providing both vehicles and services, so GM is a good example. Daimler, Mercedes is another good example. Toyota in Asia is another good example.
In the case of mobility services, we're starting also to see much more interest in expanding the type of business models that they're willing to consider adding to their transactional models that have been primarily used today. Also, making technology investments that will start transforming them from ride coordinators to mobility providers that are using fees.
Then, finally, in the cities, as I mentioned before, particularly outside the U.S., more so than within the U.S., we are seeing very active efforts to embrace mobility as a service and start offering it to their citizens. Here in the United States, I would say New York is doing a lot of work. Los Angeles is doing a lot of work. But by and large, we are lacking. There is so much ground to cover on the public transportation side that there's a lot more emphasis that needs to be put there.
Michael Krigsman: Where does public transportation fit into this picture?
Evangelos Simoudis: Public transportation will be one of the components of this picture. I think the pandemic actually has brought us some very interesting examples whereas the ridership has fallen and, obviously, the corresponding revenue from the cities has decreased. Cities have started to reevaluate what they should be offering through public transportation. In other words, what routes they should support through their own fleet and modalities because public transportation is also subways and light rail, buses, and the like where they should start partnering with mobility services companies in order to supplement the shortcomings of the public transportation system but also achieve better economics. I think that this is going to start a dialog which my hope is that these entities, these constituencies start transforming, this dialog will intensify, and will lead them to the type of convergence that I think will be necessary.
The other thing that, again, the financial factor is always a great incentive for serious thinking. As I said, as the ridership in public transportation systems has fallen, as driving patterns have changed and I think they will change again as corporations are rethinking how employees should be coming to the office—in other words, the so-called work from home effort—I think that will require cities to think about how they use their transportation infrastructure, how they charge for it, and how they reconfigure it.
Michael Krigsman: We have another question also from Arsalan Khan on Twitter, a very interesting one. He says, "What is the role of the state and federal governments in having a sandbox and the budgets for cities to start experimenting with?" It's a really interesting question.
Evangelos Simoudis: This varies by geography. If you look at in Asia and in Europe where we work quite a bit with my firm, there's a lot stronger emphasis by the central government, the country government, to support transportation initiatives at the city level. In Europe, for example, in most cities, we see over 25% of their budget, of their public transportation budget, from the central country budget.
Here in the United States, we have taken a different approach. A lot more decisions have been pushed down to the cities. From that perspective, I think the cities will need to rely much more on their own initiative and their own effort.
As we talk about the introduction of various technologies, particularly those autonomous systems that we're advocating, as well as the issues that relate to the data because these are data-heavy systems. I wrote my previous book called The Big Data Opportunity in Our Driverless Future. I talk a lot about the role of data in driverless mobility, which includes both autonomous vehicles and mobility services.
In this book, I also present data as an extremely important component, both as a value creator but also as an important component for making all these things work. Both because of the autonomous systems and because of the data issues, I think that the state and federal government will need to become much more involved because it deals with issues that go beyond just transportation.
Michael Krigsman: This issue of data, the role of data, is obviously crucially important. Tell us about that.
Evangelos Simoudis: There are really different types of data that can be utilized here. There is the data that is coming from the vehicle itself but the so-called telematics data.
There is data that is coming from the rise in the transportation that is being provided through this vehicle. There's data that is created by the passengers or the item being transported. Finally, there may be data that's generated by the various content, digital content providers, which play a huge role in making all this work seamlessly, be it the weather data, be it traffic data.
Each of these types of data have different owners. I think, today, we're seeing a very big effort for these large constituencies to claim that they own big chunks of this data, but we need to recognize the four different owners of data: automakers, mobility services companies, cities, and consumers. Again, transportation infrastructure also creates a lot of data which is extremely useful.
Going forward, we need to recognize who is the rightful owner of that data and what permissions each of these constituencies will provide to the overall system in order to have certain conveniences that can be provided by the system. In a certain way, it is a very similar conversation that we've had over the past 15, 20 years about Internet data. It has taken us a while to both understand the value of that data and how to protect consumers and how to capitalize on that data, how corporations can capitalize on that data in order to provide value to consumers.
I think a similar conversation will need to take place with regard to the mobility-related data. My hope is, because of the experience and the lessons that we've learned through our Internet journey, we will not have to spend the next 20 years trying to analyze and discuss who owns what, who has rights to what, and how to protect it, but we will be able to converge to certain decisions a lot faster.
Michael Krigsman: What about the platform aspect of this? If you're generating data from all of these different sources at this kind of volume that you've been describing, it needs to be aggregated someplace.
Evangelos Simoudis: We've actually founded a couple of startups that have been looking at both the data management side or the aggregation and orchestration side, as well as the analysis side of this. As you say, what adds complexity to all this is that we're not going to have a single platform where all of this data will be aggregated and, in fact, will have to contend with federated environments, cloud-based federated environments.
But it will require us to think about data in even newer ways than we have to date. I would say newer ways because, to date, dealing with Internet-based generated data has created challenges that few companies have been able to address, companies like Google, like Facebook, like Apple, and maybe Microsoft and a few others.
A similar thing, plus-plus, will have to take place on the mobility side. If you think about, again, as we started this conversation, we were talking about autonomous systems. If you look at an autonomous robotaxi, at scale and not during testing but during deployment, we were expecting such vehicles to be generating probably about one to three terabytes of data per day. If you assume that you have a fleet of, let's say, 100,000 of these vehicles in various cities, let's say you're Waymo and you have 100,000 vehicles running in a variety of cities and you want to start aggregating that data, we're talking about thousands of petabytes per day that will need to be processed. A subset of that will need to be analyzed so that you can be making the decisions that you need to be making, whether it is decisions about how your vehicles operate properly or decisions about how your customers are serviced properly.
Michael Krigsman: What about the business model dimensions of this? How does this change or what are the types of companies that will emerge and who are going to be the losers as business models evolve to reflect these realities?
Evangelos Simoudis: I spent a lot of time in the book talking about both the value chains that will be created out of these new mobility models as well as the business models that will need to be used across these value chains in order to monetize them. To date, if you look at mobility, whether it is in terms of buying a vehicle or getting a ride, the primary model that we have used has been transactional. I buy a vehicle. Maybe I pay it in installments or I lease a vehicle, but I'm making a long-term commitment towards that item. It's one transaction that may last over time. Similarly, when I rent a scooter or when I call for a ridesharing ride, I execute a transaction.
These models have worked, more or less, to date. But, particularly when it comes to mobility services, they have not shown to be particularly profitable because of customer acquisition, again very much like what we've learned with software. Customer acquisition is very expensive. Customer retention is not very high. Customer loyalty, in other words. Reactivating a customer can also become quite expensive.
We need to start thinking, first of all, about different models. Subscription is a very good example. But as I say in the book, I think we also need to have the imagination and maybe even fortitude to start combining different models. For example, combining a subscription model with an advertising-based model or a loyalty-based model with a transactional component.
We need to learn from other industries. The mobility industry, the new mobility industry will need to learn from other industries.
Now, who are winners and losers? That's a very good question. I think that companies with startup roots, such as the mobility services companies, have a higher probability of success not for any other reason but because startups are all about business model experimentation. I think that those companies, whether it is companies as large as Uber and DiDi or whether there are smaller competitors, they will be able to iterate, to test and iterate, through business models a lot faster and find the ones that work for them best.
The automakers, in general, larger corporations that have established and scaled business models, I think, will have a harder time testing that. To me, some early examples of that difficulty is what has happened with the mobility services such as Maven or Chariot that GM and Ford offered respectively. In both of those cases, they had to retrench. In the case of Chariot, obviously, Ford closed it down. In the case of Maven, GM retrenched significantly from the initial goal. Daimler with its now mobility service also has significantly curtailed the deployment of an expansion of that service. Again, to me, these are indications that these companies have a harder time working around these newer models than their competitors that have more startup roots.
Michael Krigsman: We have a couple of questions from Twitter relating to data privacy. Because of the amount of data that's being aggregated, whether it's technical data coming from cars and telemetry data, as well as tracking data and the risks associated with bringing together that kind of very personal data, where will that be an inhibitor and to what extent can we think that companies will respect personal privacy? Will it be any different from the way things are today?
Evangelos Simoudis: I think the experiences that we've had with Internet-related data, browsing data, will provide some guidance on how to proceed here. The truth of the matter is that, as it happens with the Internet, we are exchanging privacy for convenience. Here again, the public will need to decide how to deal with that with the privacy and the convenience.
I think that the cities, in particular, will have to impose certain restrictions on what is being tracked and how it is being tracked. I think even with conversation that you have today and maybe that have evolved into lawsuits with the City of Los Angeles, you're seeing some of that tension between mobility services companies and cities in terms of what data is shared and how that data is used.
For me, this is why I talked about the four different constituencies that have data rights. I think those need to be involved in this conversation and we will need to create those guidelines and the roadmap of how the data will be used. But again, there's no question that you're giving up certain privacy if you want to have the convenience of certain types of transportation.
Michael Krigsman: In effect, it's hard to see that this is any different from what has come before Internet data, as you described.
Evangelos Simoudis: It's no different, in my mind; it's only bigger. With the case of the Internet, it deals with your transportation within the Web. In the case of what we're talking about, it deals with your physical transportation, what you do on a daily basis. From my perspective, it is no different than that. Again, in both cases, you're dealing with certain privacy issues and certain breadcrumbs that we leave behind.
Michael Krigsman: The traditional automakers, what happens to them in this scenario?
Evangelos Simoudis: Automakers have some very tough choices to make going forward. As I say in the book, perhaps they are the group, the constituency that will have the hardest time undertaking and executing the transformations that I prescribe.
Like every other large corporation that has a scaled business model, they will have to determine whether to abandon or to lessen their reliance on that model and adopt something else and what that something else is going to be. In the book, I talk about automakers, going forward, being organized into four different categories depending on what type of transportation they want to undertake and how far they take it.
I think that automakers that decide not to embrace new mobility, they run the risk of being consolidated. We're already seeing certain consolidations in the auto industry here in the United States and in Europe. I think you will also see the closing down of the number of models that they offer, a shrinking of the number of models that they offer.
Also, you will see them determining who they compete with. I imagine that, in certain cases, they may be competing with their suppliers who have manufacturing capabilities. You have companies like Magna, for example, that is able to produce complete cars now very much like at Foxconn, and even Foxconn can come into the play.
Michael Krigsman: What advice would you offer to cities? You started talking about that a little bit earlier.
Evangelos Simoudis: My advice to cities is that they should really embrace this transformation because it will allow them to provide better services to their citizens. It will allow them to address many of their long-term problems as well as generate new forms of revenue.
In the process of doing that, they need to think themselves about how to instrument their infrastructure, what type of instrumentation will be necessary if they're going to provide these services, and what will be the best forms to collaborate with both automakers and mobility services companies.
They need to become far more confident in themselves to become the next generation, in a sense, coordinators and even air traffic controllers of this transportation because, again, it's not only about movement of people. It's also about movement of goods. As we are changing our preferences on where we work from, how we work, and how we live, relying a lot more on e-commerce, the movement of goods will continue to become a very important part of the overall transportation puzzle.
Today, over 30% of the trips that we take are to do shopping. More and more, this will have to lessen if we're going to address some of these issues.
Michael Krigsman: How should cities think about paying for this?
Evangelos Simoudis: Cities will need to start understanding business models themselves and they will need to understand how to best leverage their infrastructure. Their infrastructure is not only the road. It is the sidewalk. It is the curb, which plays a very important role.
What you will see going forward is, whether it is for bike lanes and scooter lanes or vehicle lanes, you will see cities able to monetize those infrastructures according to usage and according to demand. In order to do that, of course, it will require to, as I said, instrument them but also have the platforms to manage the data that the instrumentation provides.
More importantly, technology aside, they will need to start understanding that this is what can give them a very bright future. Of course, certain cities have started already realizing that. You see it in Singapore. You're starting to see it in certain cities in Europe like London. New York, here in the United States, has already started moving in that direction. Again, I'm encouraged by what I'm seeing and I hope that there is going to be accelerating momentum by the cities in doing that.
Michael Krigsman: Evangelos Simoudis, thank you very much for taking time to talk with us today about the very interesting vision of the future, one that, in one way or another, is very most likely to show up sooner or later.
Evangelos Simoudis: My pleasure.
Michael Krigsman: Everybody, thank you for watching, especially to those folks who contributed comments. Check out CXOTalk.com, subscribe to our newsletter, and we will see you again next time. Thanks a lot, everybody, and I hope you have a great day. Bye-bye.
Published Date: Jun 19, 2020
Author: Michael Krigsman
Episode ID: 658