Transportation has changed dramatically in the past decade and a half, thanks to innovation from startups and legacy incumbents alike. From the surge in popularity of electric vehicles to the excitement of the possibility of autonomous driving, the sector has given rise to many new businesses and business models.
Jennifer Haroon witnessed it firsthand. She was on the early team at Google Self Driving that later became Waymo, and most recently worked as the CFO and COO of Nauto, a Greylock-partnered startup working on AI for driving safety. Jennifer recently joined Greylock as an executive in residence, and will be working with a wide range of companies. Prior to her work in transportation and mobility, Jennifer worked in health policy and technology. Her experience working in heavily regulated industries and identifying where technology can best be applied to increase efficiency has given her a unique perspective on the role of entrepreneurship in complex sectors.
Jennifer sat down with Greymatter to discuss the current state of the transportation and mobility sectors, including trends across go-to-market strategies and consumer behavior. She also shared her outlook on ways automation technology is impacting many different sectors, including productivity, security and saas.
You can listen to the podcast here.
EPISODE TRANSCRIPT
HM: In the last decade or so, we’ve seen a lot of major changes to the way humans, goods and services move around. You’ve been involved with some of the companies making the most impact here, from autonomous driving at Google to AI powered platforms for transportation safety at Nauto. Before we get into everything that’s going on in the industry, can you share your experience in the field?
JH: Yeah. Working backwards, I most recently was the COO and interim CFO at Nauto. Nauto uses computer vision and deep learning to help drivers and fleets drive smarter and safer. And before that, I was part of the Google self-driving car team, where I was really working on all sorts of early stage business aspects of the company – what’s the right business model, what are the partnerships with OEMs and other infrastructure and players in that market – and I helped spin them out to become Waymo.
HM: How did you get into that?
JH: It was actually a little bit random. I had already spent six years at Google, and one of the things I love about Google is that they really think about how to apply technology and new technologies to every space. So I had worked on looking at how we can apply Google technology to the healthcare field. That’s how I started my time at Google. And then I actually started a company with some other Googlers in the telecoms field. It was a B2B fiber optic network in Sub-Saharan Africa. And when I was feeling ready to move on from that project, I was just networking and a woman I know who worked in Google’s finance team said, “Hey, you have to talk to the self-driving car team. They’re looking for someone who can help, you know, build their partnerships, [someone who] has worked in regulated markets,” (and transportation is highly regulated). So yeah, it was not really because I sought out the transportation field.
HM: So what was going on at the time with transportation?
JH: It’s been really fascinating to look at transportation over these past couple of years. I never thought it would be such an innovative and amazing space. At Google, we were working on autonomy and it was this idea that – with the progress we’ve made on big data and AI and technologies like computer vision – we could apply those to something as complex as driving. That, for me, was a really interesting idea. And at the same time, there were a lot of other innovations happening in transportation, like with the shared economy and electrification. So all of those put together was making mobility and transportation a really interesting and very hot space right now.
HM: What was surprising about working in the sector?
JH: Well, certainly with autonomy, making predictions has been difficult. I mean, I was wrong. I thought we would be at a place where there was more fully autonomous driving available today than there is. That being said, in the last year we’ve seen Waymo – which used to be Google self-driving cars – offer fully autonomous rides now in the metropolitan Phoenix area. We’re seeing some sort of what I would call maybe “Closed Route” or “Set route” autonomous buses and shuttles both here in the U S and abroad like Singapore. So there is progress being made. It’s just slower than many of us expected. But at the same time, I think it’s going to happen, so you’re also seeing excitement.
HM: And tell us what you worked on at Nauto.
JH: I worked on a really big variety of things. So at Nauto, it’s this idea of using some of the same technologies using autonomous vehicles, but applying them today to help people and companies today. And what I like is this idea of the broad AI term – using AI technologies to help people do their jobs better, or more efficiently. And so my part of it was really just helping build the company and the operations; execute on everything from the people on the talent side, the finance side, to partnerships and how we just operate the company.
HM: And when you started, there were just a few organizations really working on a lot of these different problems (or solutions) in mobility. Where are we today? There’s been a lot more startups and there’ve been a lot more involvement from the big legacy companies as well.
JH: Definitely around the time that I left Waymo, there was a proliferation of startups, particularly in the autonomy space. And I think that’s just because there was a recognition that there really was a “there” there, and what we’ve seen in the last few years is a lot of consolidation in this space. I think that’s necessary because it’s going to cost a lot of money and require a lot of talent and time and expertise to get to the end state that we’re all looking for. And it’s very hard for that to be split amongst lots and lots of companies.
The other thing we’ve seen is traditional automakers over time have recognized that a number of these technologies from autonomous to electrification are important in the industry in general. So they’re also looking to both build it internally, but also to do so in terms of merging with other companies.
HM: Yeah. Let’s talk more about electrification. It’s hot right now.
JH: It’s almost bubbly. Yeah, it’s been amazing, right? I mean, Tesla has done an amazing job of showing both consumers and the industry that EVs can be great vehicles – in terms of great-looking vehicles, great performing vehicles, and that consumers want them. I think that’s been a really important part of helping EVs advance.
So, now we’re seeing all these SPACs. A lot of vehicle companies – not just electric car vehicles – but also other industry players around electrification – are SPACing. It’s interesting because similar to autonomy, it’s going to take a lot of time, money, and expertise to get there. So it’ll be interesting to see if public markets are patient enough for all these other companies to really make the progress that’s needed.
HM: Yeah, exactly. And Tesla’s unique in a lot of ways. Elon Musk has a huge following. A lot of fans and people have been really willing to embrace him and get on board with this vision that he has. You’ve had all these hardcore early adopters. Do you think that other companies will be able to have that sort of support from both consumers and investors?
JH: He has done an expert job on the marketing side, and he used a lot of techniques that you saw in tech businesses, like wait lists and signing up early. Teslas aren’t sold the way traditional vehicles are sold. And I think that was really smart, and it does remain to be seen if other companies can use some of those techniques to, as you said, build that fan base. So that’s where there’s that loyalty and demand ahead of building, because I think that helps a new manufacturer. It’s really expensive to put up a vehicle manufacturing line and get that done. And so I think getting that early demand signals really helped Tesla.
HM: Right. And even though there’s a lot of momentum on the electric vehicle front, there are a lot of issues still, even when lawmakers are putting policies in like you can’t have any gas powered vehicles in California starting in 2035. And I think both Joe Biden said this week that all government vehicles are going to be electric powered. What do we need to do to actually get there and to make that real?
JH: I do think the regulatory landscape has helped the EV space a lot, in addition to everything that’s already happened both in Europe. And now we’re seeing that more in the U.S. with, as you said, President Biden’s announcement – that helps a lot, obviously, in terms of creating demand and giving some level of certainty to makers in the space.
But one thing is that there does still need to be a lot of work, particularly in terms of charging. There are large swaths of the U.S. where there’s not a lot of charging available. I think for individuals, it can be a bit easier and that you can charge at home overnight when the cost of electricity is very low. It’s actually a very cost advantageous versus gas, even while gas is relatively low priced right now. But then when you start talking about range, there’s that issue, there’s anxiety about charging in the middle of the day during peak hours – where can you? So that means it’s still more expensive per mile than gasoline is right now. So charging is a big issue and there could be some creative business models. Actually, instead of “charging for charging,” I can link that to other things that people are doing. Can you charge your vehicle while you’re sitting at a restaurant, seeing a movie, things like that .
HM: In talking about interesting business models and business opportunities, this is still a hard industry to get into. It’s heavily regulated, as we talked about, there’s a lot of raw materials just to make batteries and parts. There’s a lot of manufacturing issues and all the things that come along with that, and supply chain logistics problems. Then, you need people who have really modern software engineering skills. How do you assemble these teams to keep building in this space?
JH: Yeah, it’s funny, when I first started in the Google self-driving car team, a lot of the press articles liked to pit Detroit versus Silicon Valley. That was a headline that I guess got a lot of clicks. In reality, it’s never so black and white. There needs to be a partnership because, as you said, the supply chain, the tier one suppliers, the manufacturing – that is not easy to just build from scratch. And so you do actually see a lot of these mobility players and a lot of the traditional automakers trying to hire from each other. We’ve seen that a lot, not just through M&A, but just in terms of trying to build talent. And I do think that’s really important. And there’s lots of ways to build talent and a team, whether it’s in your team directly. And then there’s also the role that board and advisors play,
HM: Also, people need to be patient working in these industries, right? It’s not going to be as fast as say, just pure software.
JH: That’s right. I remember the days when lots of people were building apps for Facebook. Just games and various apps for Facebook, you could do that for some engineer, they could seemingly do it overnight. That’s not the mobility [sector]. It needs really patient investors, patient capital, and even a patient employee base that really believes in the mission and what they’re doing and willing to build that for the long term.
HM: Building off what you were just saying about boards, we recently aired a two-part podcast series with Greylock general partner Reid Hoffman on the topic of building boards. And during that discussion, we talked about the importance of having people with a wide range of backgrounds, not just pure VCs or people you work with before, but a wide range of expertise and experience. How did you see this play out when you’re working in transportation?
JH: Yeah, that was a great two-part series. It was interesting because Stefan Heck, the CEO of Nauto, one of the things that I really admire about him is the way he very thoughtfully built out the board at Nauto from the very early days. So for example, he started adding independent board members as early as Series A. So he added one with a Series A raise, and then one with the Series B raise, you don’t see that with a lot of founders of startups. They worry that adding an independent board member just means another person that they have to manage rather than build the company. But the way I think Stefan saw it – and I saw it play out – is it was actually a way to add additional expertise.
So the first independent board member he added was a woman named Karen Francis who spent many years at Ford and then GM. She has a lot of connections in the automotive industry and is a really smart, amazing woman. And in my view, she brought a lot to the board that was different from what our investor board members brought. It was also a way to add diversity, as Reid says, realistically, these days your investor board members are most likely to be men. And the other thing I like about adding independent board members is they’re very likely to either be currently operating, or more recently been operating. What you see in a lot of VCs is, you know, people have been doing investing for a while. And so it’s been a while since they’ve been operating. And so I think that really provides another perspective and help for the CEO and the management team. Someone who’s either in it, or has very recently been in it in terms of solving operational and execution challenges.
HM: Yeah. Those are all really good points. And as we were talking a little bit about the different ways that people could get into this space, or different opportunities for innovation, a lot of this could be in not just cars people are driving, but in shared transportation or last mile type of city solutions. So in a lot of ways, we’ve seen the pandemic really affect pretty much every industry, some positively, most negatively. How did you see it play out with transportation and mobility and specifically public transportation and delivery?
JH: Yeah. The pandemic has been really tough on passenger transport broadly, for very obvious reasons. It is an area that I worry about. Ride hailing has obviously had a huge growth moment over the last many years, and they were obviously hit very hard. I’m a big believer in public transit. I used it a lot before the pandemic while we were still commuting to the office, and I’ve also lived and worked in other countries that had excellent public transit. So I see the benefits of it. But for good reason, people don’t feel good about using shared mobility right now. And so there’s a lot of questions about when we are going to have reached the point where we can call the pandemic over (or essentially over), will both ride hailing, as well as public transit, be able to rebuild trust for consumers?
And then it actually ties to the future of work, which a lot of people are talking about with regards to the pandemic. If there are fewer people going into offices that are consolidated in a central business district or an office park, how does that impact public transit? Because it’s not just the office workers, but there’s also all the businesses that exist around all the offices. And so if there’s less demand, it makes it really hard. Both public transit and shared mobility are in some ways a volume game. They benefit from having a good volume of users. So I really do worry about that as well.
HM: You have a note in here about the kind of data you would need from large shared passenger fleets. I want to hear what it’s like to get enough data if you want to improve these things.
JH: Yeah, I would not call myself a public transit expert by any means, but I do think one of the challenges I’ve seen – just as a consumer of public transit – is that customers expect more transparency and real time data and information. When I’m waiting for the bus and I’m on the app that tells me the bus is 10 minutes away, but then the bus comes in 30 minutes, that makes me upset. I don’t trust the system anymore. So I do think, particularly with public transit and governments who tend to run public transit, there is a need for better data on traffic, on traffic flows, on usage [of public transit]. And it’s not just the data, but the analysis of that data. The use of that data is so that they can both make routing schedules more efficient, to put in more routes, as well as provide that information to customers. So I think there is an opportunity there. And then on top of that, obviously the use of AI can help with prediction for routing and all of those things. So I think there’s an interesting opportunity.
The question always is, as you know, is there a big enough opportunity that it’s an interesting enough business that can then provide those services for the government?
HM: Right, and at the same time that there has been a decline in bus ridership and taxis and ride hail, there’s been – obviously to anyone who looks out the window – there’s delivery trucks constantly. So there’s lots of fleets and logistics services. What’s what’s going on there?
JH: Yeah, I think fleets are actually really interesting. It’s not an area I knew a lot about until I’d entered the transportation industry. And in the U.S. for example, in some ways fleets have been on the decline. Most people don’t know anyone who has a company car anymore. That used to be more of a thing. If you were an office worker, one of the parks might be a company car. That is less likely these days. So in some ways, fleets had been on the decline, but as you said, even before the pandemic, we started seeing more demand from consumers for things to be brought to them. One is obviously packages – e-commerce has grown enormously – but even services have grown. People want services brought to wherever they are. And so that has made a huge expansion in both long haul trucking, but also last mile service, logistics and delivery. And what we saw with the pandemic was that that just exploded. Over the holidays, most of the big name carriers that we all recognize had issues even getting extra vehicles. Their fleets were not big enough, and they couldn’t rent enough short-term vehicles to even meet demand. So I do think this trend of fleets will continue because a lot of autonomous vehicles – because of the restrictions in their use – that technology developed will also be in fleets rather than in individually-owned vehicles. I think fleets are having a growth moment and that of course means there’s opportunities all around fleets.
HM: What kind of ideas do you have there? What do you hear floating around?
JH: For one, we talked about electrification. I think there’s starting to be interest from fleets in EVs, and certainly President Biden saying all federal fleets should be EVs is going to help you push that even further. Then you’re talking about companies that own fleets – they’re gonna need charging, right? They’re gonna need to change their depots. If these are fleets of vehicles that people do take home, then are people at home in a place where they can charge those overnight? Not everyone has that ability. For example, I live in a place that doesn’t have a garage. So if I had an EV, I would need a place to charge it. Then, I think about fleets both around autonomy and EVs. You’re talking about different types of maintenance. When you’re talking about autonomous vehicles, there’s always the big question about cleaning and helping people who are using autonomous vehicles, because there’s no one in the car to immediately help you. So I think there are opportunities in that whole ecosystem that we’re just starting to find out about.
HM: Is this a hard industry to get into now – or to continue working in – with the need for so many people to keep their distance from one another?
JH: In some ways it can be. When you’re talking about a lot of these companies, whether it’s self-driving or autonomous vehicle companies, or whether it’s a company like Nauto, there’s a hardware aspect to the business. It’s not just software. And so when you have that hardware aspect, you can’t bring everything home with you. If you have a LIDAR lab or a thermal testing lab, you’re not going to bring that home in most cases. So even at Nauto, we had to really look at how we adjust during the pandemic to allow some to access the office in an obviously socially distanced, masked, safe manner. I do think these companies have a little bit more of a challenge than just a straight software company.
HM: Obviously there’s a bunch of companies in this space and the most valuable company in this space, Tesla, was once a venture-backed company. But I’m wondering, how do you make the case that this is an industry that should have VC behind it? It has these long, long, long lead times, it has very high overhead costs. How does that become a venture capital business?
JH: Well, it has become a venture capital business. I think it’s a good question because I wasn’t sure in the early days at Google. We weren’t sure if it would become one, because of the reasons that you said: there were really long timelines and we’re talking large, large amounts of money for both AVs and EVs.
And so I think we’ve seen a couple of things. One is some VCs like Greylock have made the jump because they see all the potential. Two, is that we’ve seen a lot of companies stay private for a lot longer anyways. Even just software-only companies have been staying private longer. There are some benefits that these companies have found in doing so. So that’s also meant that VCs are experiencing longer timelines regardless of whether it’s a hardware- or software-based business.
But I do think that these long timelines are part of the reason we’re seeing things like this surge in SPACs, because you have to look at other types of funding once you get past traditional VC-type amounts of money and timelines. So some companies have gone to funds, some funds have invested in companies even before they’ve been public companies. They’ve sort of built out a private equity (or as part of their growth equity) kind of business. But I think SPACs are another outcome of that – it’s a one way to raise a large amount of money. The question will be about investors. You build a PIPE with your SPAC. Well, those investors realize that they need to be so patient. And then when you’re public, there are other stresses on the business. And there is a question of whether that will interfere with the building of these really complex businesses.
HM: Well, I’m very happy that you have joined Greylock as our newest executive-in-residence, and while I’m sure you’ll still have a hand in the mobility and transportation space for a while, given you given your background. But now you’ll have an opportunity to apply your experience to a wide range of startups. Over the course of your career, what have you learned that you’re eager to take in other areas?
JH: I’m really excited about working with the team at Greylock to look at not just the mobility and transportation space, but as we talked about, I’ve been building companies in a lot of other spaces. Really, the theme has been about bringing technology to other industries. I really enjoy that. And then the other part of it is that I’m a company builder. I’m an operator, and as an operator, you think a lot about efficiency and productivity and how to help your employee base achieve those things. I’m really excited about the productivity space and the enterprise software space more generally, I think there’s opportunities, not just for the developers and for the engineers, but for folks in finance and HR and recruiting and legal. I think there’s an opportunity to add automation, to use AI, to help people – again, not to replace people – but to help them so that they have more time for the higher level activities. Certainly, as we’ve seen during the pandemic, there’s opportunity to help people collaborate more in new ways that they didn’t need before but that they need to do now. And then of course I’m always interested in the use of data and everything you do to create data. I think sometimes people think about [areas like] recruiting or legal, and they don’t think about the data aspects of it. But I actually think if you can find ways to get the right data from those activities – and figure out the best way to analyze it efficiently – you can find ways to make those teams even more productive. Obviously my dream as a COO-type person is to bring that all together, because those functions don’t work in a silo. That’s not how you’re going to build a company to work efficiently. So bringing that all together is why I’m also excited about the productivity space, in general.
HM: And you also had the opportunity to experience how to think about doing that in the most financially intelligent way possible, because as a CFO, you’re living in the best of both worlds.
JH: That’s true, because when you are really early in a startup, I think you just put everything in spreadsheets. I love spreadsheets – I very oddly use spreadsheets a lot in personal life for, for things that probably other people don’t.
HM: I could have guessed that.
JH: Ha, but there comes a point when your spreadsheets are starting to fail you. I mean, I had so many spreadsheets, it was hard to keep track of them. And at some point, they don’t all talk to each other efficiently. So as the company grows and becomes larger, both in terms of people and revenue and customers, but also more complex, that can be hard to avoid. The spreadsheets aren’t enough. So then you need to start adding really smart systems. And I don’t think all of the potential smart systems exist yet.
HM: Nor do I. Well, Jennifer, thank you so much for your time today. It’s been really interesting and I’m so glad that you’re here at Greylock. So thank you so much for joining as executive in residence, and thank you so much for joining Greymatter to talk with me today.
JH: Thanks, Heather. It’s been super fun and yeah, I’m looking forward to exploring all of these spaces with Greylock.
