There is no predicting of what may happen on the road. The best thing we can do is take precautionary measures. Taking that to heart, Carey Anne Nadeau founded ODN—a company that builds AI products that augments what insurance carriers know about who their customers are with information about where they drive and their risk exposures on the road. In this episode, Carey explains further to us how they do this and the risk rating methods they use to measure environmental hazards on the road. She also taps into the use of big data, shedding light on how far we’ve come in terms of the technologies we have to predict and protect ourselves. With technologies like this, the future is bright for the taking. All we need is to take advantage of services like that from Carey’s ODN.
—
Listen to the podcast here:
ODN: Measuring Risks For Safer Roads With Carey Anne Nadeau
I am excited to have Carey Anne Nadeau from ODNsure. Carey Anne was gracious enough to have me on Golfcour.se, her podcast, and we had a phenomenal time. I brought a wedge. I brought a seven iron and a driver and a putter. We had a great time together. We had so much fun. I had to bring her on because she is the power of ODNsure and we have to sit down and talk about this because what you do is incredible and you blow my mind. Welcome, Carey Anne. You guys are doing technological stuff. Tell me a little bit about yourself.
To fill in a couple of blanks, I founded a company called ODN. The website is ODNsure.com. You can go check out more there. I’m building a brand that builds AI products for insurance to help contextualize exposures for all different lines of business. Meaning where are you in? What does that mean for the likelihood that you might be involved in a bad thing, have a claim that you might want to file for your insurance? It’s a super technical subject. AI in and of itself, I think a lot of people go, “Statistics class in middle school. Please don’t make me do this anymore,” immediately. It’s an interesting subject to be able to talk to you about and share what we’re learning about translating something that is deeply technical and has that visceral negative response in most people. How to turn that into a positive, how to structure a conversation that gets people interested, excited and engaged about what you do. Don’t speak in words that are for college-educated folks but simplify and elucidate. Translate what is technical and things that are fun, relatable and help people understand what you do on an intuitive level.
Let’s get into this because what you do is massage big data at an enormous level. If you look at your website, from what I understand is you are able to predict where car accidents could happen or where the bad locations are anywhere in the United States, for the most part. Explain to me how this works. How did you get to this point? What brought you to sit there and go, “This is what I’m going to do the rest of my life.”
Hopefully, I do this and I do many more cool projects like this. How it came to be is if we think about how we drive around where we live, there’s always an unsafe road. One place to sit at the red light a little bit longer because you always see cars run through it. For me, there’s a stop sign on route nine south of Hartford on an unremarkable highway. That intersection bugs me because people expect the person in front of them to merge into 65-mile hour traffic. They don’t, for whatever reason. You just see a tremendous amount of fender benders there. Anybody who sat in traffic congestion and then the result of that congestion have been some traffic crash at the end, whether it’s a severe one that looks terrible and awful and you’re saying prayers for their families or whether one that’s like everybody is looking at while they drive-by and slowing down traffic.
We can all understand the experience of traffic crashes and we can all put together the reasons why they might happen, high speeds, narrow roads. I mentioned stop signs and merges, dangerous merges. We all intuitively know this is what causes crashes at this place. We said, “Can we take data and a wealth of data like brain-busting amounts of data and take what we know to be intuitively true and mathematically prove it to be true?” When you’re talking about questions like where a car crash is likely to happen or where claims associated with car crash are likely to happen, it feels like a human question.
Humans should be the only ones who are able to answer that question because it is complex. With the wealth of data that’s being produced about roads now that’s being generated from city and state governments first and foremost to keep track of the roads, their condition, physical characteristics and traffic volume, to police departments who are collecting information about where a traffic crash has happened, you start to layer on things like telemetry data devices plugged into your car to check where your hard brake or speed or devices in our traffic lights that are monitoring traffic volume or air quality. You start to realize quickly how much information is out there. We ask the question, is it enough data? Do we have enough experience and enough content to be able to get close to answering a question that to date only humans have been able to guess at the answer to?
Rectifying information to a person is a hard task. Share on XIt turns out we do. Unfortunately, there are hundreds of thousands of traffic crashes every year. To contextualize, about 40,000 people die every single year in a traffic crash. They’re awful. We shouldn’t be accepting that level of risk on our roads. For a statistician, that is helpful because there are many observations that we’re able to model them. With all of this data and advances in cloud computing that enable us to process that data quickly, efficiently and importantly for our company affordably, we’re able to generate these predictions that will only get better over time as more data is collected. I think AI as an industry, let’s talk about AI in general, is experiencing this transition from revolutions in the methodology itself. These are the pioneer statisticians who are creating new math techniques and methodologies.
Where we are now is this implementation phase. Where ODN sits in this implementation phase where we say, “The math is well-established. The data is there. The computing power is there.” Can we put it to work? Can we start answering these questions? This is one of many that I think we’re going to be able to answer and we’re going to be able to use, particularly in my industry, which tends to focus on insurance or managing risk with commercial fleets, city governments and all those who are driving on the road. Can we use it to inform these important questions about where we can keep people safer? Can we prevent crashes? Can we save lives? Can we reduce that 40,000 number to something that’s much more tolerable and much less accepted on a day-to-day basis? That’s what drives me. Where I think in terms of the movement of AI, we sit into this evolution that’s a much bigger movement than ODN.
It sounds like you’re at a nexus. For years, we’ve been collecting data. Everybody’s been knowing that. In a sophisticated way, some in paper and pen, but we’ve always been collecting data. We’ve always known that there are car crashes. We always know there are fatalities. We always know where the locations are of this thing. It’s being able to take all that information, synergize it, amalgamate it, process it layer on layer so that people know that it happens at 6:00 at dusk, on rainy days when the sun is facing a certain location that this becomes a certain situation. That’s where the magic occurs.
Too many people have all this data and they do nothing with it. People collect data because they said, “We need to fill up these hard drives.” You don’t need to clog data to fill up hard drives. You need to be able to make salient information out of it and be able to help people. My question to you is, first of all, how do you amalgamate all this stuff? Because you’re dealing with city government, state government, the federal government, highways departments, weather departments. I don’t even know where even to start, the insurance companies. How do you get all this information together and get everybody talking nice to each other, grab the information to be able to make sense out of it?
It’s not easy. A lot of folks who talk about doing data analytics or who are thinking about doing data science don’t give that part a lot of focus. How are you going to build the architecture so that you can do it once and you could do it nationwide? For us, it’s a fun little trick and one of the things that differentiate our business in the market is that we think of things not by a person. I don’t think, “Ben Baker, what’s your age? Where do you live? How much money do you make? What’s your financial history? Have you been in a crash?” I’m not collecting any information about you. What I want to know is where you are. If I can know that you are sitting in your podcast studio on the first floor of a building on a residential semi-commercial street where there aren’t a lot of traffic crashes.
If I can contextualize where you are, I can know a lot about you without asking you any questions or knowing anything about you personally. For us, that’s the trick that we join everything on a map. We put a map out and we say, “Here’s where you are. What does it look like based on your environment? What risk exposures might you be exposed to?” At the core, we start with where you are and location as part of the way that we join the data. Instead of worrying about having an identifier for Ben Baker, whether that be your name. No offense, there are probably a few Ben Bakers in the world.
I hope there’s more than one. I hope there’s another Ben Baker out there I can blame for everything. “It’s that Ben Baker, not this one.”
It’s a hard thing for insurance carriers to rectify because they may collect data about you, but they may get the wrong Ben Baker or more likely, they may look at Mrs. Baker and say, “I have information related to this address or this vehicle.” If Mrs. Baker takes your car to the store, now that car isn’t associated to Ben Baker anymore, it’s associated to Mrs. Baker. They have a hard time knowing whether that data is valid for you and should be used to price your insurance or if it’s your wife who’s the bad driver, whether she’s the one getting an accident using your car. Rectifying information to a person is a hard task. There are companies that tried to do this, particularly in financial services where loan servicing and fraud is an important thing to be keeping track of an individual’s experience.
There are businesses that focus on that exclusively and even may do it probabilistically, so they don’t have certainty that it’s always the real Ben Baker. They’re guessing and they’re saying with 90% certainty, “I’m sure that Ben Baker used his credit card here and this was the real Ben Baker.” What we say is that level of uncertainty, that difficult task of attaching and assigning data to you, forget about all of the expenses, time and difficulty of collecting all this information about you. What if we only knew where you are? You probably have already given away this information in 8 or 9 phone apps. If you open up your iPhone and look at where you’ve location-enabled, you’ve probably given it to Google Maps.
I’m giving it to everybody. Every time I click on a website, it says, “Can we track who you are, where you are,” and go, “Why not? Another person knows where I am. Okay, good. Come get me.”
It’s the most important piece of information that everyone gives up and they probably shouldn’t. Because it is infinitely powerful to associate data with you. If I know where you are, I can geospatially or use proximity attach to you. Ben Baker parked his car outside of a liquor store at 10:00 at night or a bar. He didn’t move his car for two hours and then he got on the road and he was speeding. Is Ben Baker a good risk to be insuring? Maybe not. I haven’t collected any information to answer that question about you personally or from you. I used where you are to answer that question.
It’s being able to sit there and say, “We’re able to track this person. We’re assuming their habits.” Whether I was in the bar having a cup of coffee or twelve cocktails, you have no idea. As an insurance company, you could make assumptions based on where my location was, how long I was there. That’s fascinating because it’s something that the average person would never think about.
Most people think about data collection about them personally. What the real trick of what we do is there’s no personally identifiable information at all. I don’t know who you are. You’re a black spot to me. I know that you’re on the map and where we apply it is in the auto space. I’ve given you some examples, but how it works for auto is where you drive. I’m tracking what roads you drive on. It turns out that knowing that you drive on unsafe roads helps determine whether you’re going to be in a crash. It’s intuitive. You may be the safest driver in the world but you’re driving around in an area that has had a historical record of a lot of crashes, has high speeds or cars parked on the right-hand side or some combination of all of these things. We put that together with a model to be able to control for all of these factors. You might be involved in a crash no matter if you’re the best driver in the world or not.
It is infinitely powerful to associate data with you. Share on XWhere this ultimately plays out though and I think the real magic of what we’re doing, the real opportunity of what we’re doing and the important future vision that I see for this technology. It’s important to know Ben Baker drives on unsafe roads. Where we see something magical is what happens when Ben Baker starts driving but doesn’t have his hands on the wheel? Now, Ben Baker’s in a car that’s driving itself. Does any of the data associated with Ben Baker matter anymore? Does your credit score matter if the car is driving itself or your hard braking and speeding matter if you’re not the one hard braking? Transitioning into semi-autonomy and full autonomy, we can debate about when it’s going to happen.
Hopefully, when I stop driving, then I’ll be happy with it.
There are arguments both ways. Some people say it makes it a lot safer. You might want to be on the road in a fully autonomous environment. You’re just not comfortable with it yet. It’s still a cultural transition there. When we get into that environment, now we have to think about what risks we can quantify because crashes are still going to happen, but it’s not going to be your fault anymore. It’s going to be the vehicle’s fault or something else’s fault. It’s something that affects you, something you’re exposed to. That’s the data that we provide. The example of the Tesla that was on cruise control and did get in a crash and had a fatality, it was an awful circumstance, but ultimately it was explained by a weird anomaly in the road infrastructure itself that the machine hadn’t learned to navigate yet.
It hit a median. It hit a Jersey barrier. If we can pre-program some of those exposures and let the car know where it’s unsafe to drive, now the car has a greater awareness of the environmental risk. It can sensitize its sensors to account for some of that, be a little bit safer in some places, have less or a faster reaction when it perceives outside exposures in areas where it knows it to be unsafe. There’s a lot we can do to prepare for the autonomous future if we understand and have a baseline measure of what risks that car is going to be exposed to.
I’m going to go back a little bit because all this is fascinating, but I’m not sure I quite got it. Is this because the data is, “I’m a black hole. I’m a black box. I don’t exist. It’s information. I’m a spot on the map. I’m part of an aggregate?” Is it that much easier for you to get the data from the different data sources because you’re not breaching my personal information?
I don’t need to know anything about you. I need to know where you are. As long as you’ve given me permission to know where you are, I can connect a world of data to you and measure all of your risk exposures.
Let’s take that forward. Now that you’ve explained all that to the Tesla examples, to road conditions, do you get hired mostly by the insurance agencies, by the car manufacturers, by the cities and highways? Who would be the biggest user of the amalgamation of the data to be able to sit there and say, “We’ve got problems here, here are some different ways we can fix it?” Whose ears are perking up when you’re speaking is probably the best way of putting it?
I hope your ears are perking up. You point out multiple markets. In our business strategy, we’ve defined a roadmap that starts with city and state governments where we’ve sold 38 cities around the United States. We’re deployed in Chicago and DC as a methodology to help the government best decide where to invest in traffic safety and vision zero initiatives. That was where we started. We have a roadmap that our next target is the insurance industry. We’ve been selling into insurance now. With insurance carriers, they care about a few things. One, they care about engaging their customers and acquiring customers. They want to provide this information to individuals like Ben Baker or fleet managers. You think about the Ubers, the Lyfts, the riders, even some major distributors like Anheuser-Busch that are on the road and have control over where their vehicles drive.
They also want to be able to price based on that information. It’s important that the math works, that we’re predicting where crashes are likely to happen. We’d hate to give you the advice to avoid an intersection and put you on another road that was worse. The math has got to work and it’s got to essentially be able to demonstrate to insurance carriers that they’re going to lower the actual risk, save money and more accurately price people. Folks that are driving on dangerous roads are going to be able to identify them and price them appropriately and make more money. It’s got to be profitable to implement new technology because of its costs. After that, we have built partnerships with telemetry providers. These are the plugin devices in your vehicle. The important thing about telemetry providers is they work across both insurance carriers and commercial fleets.
If you’re a risk manager of a commercial fleet, you are using these devices to track where cars are going, where they’re driving and in a way can benefit from more data from that telematics provider to contextualize not only how your drivers are driving, but where they’re driving. Next up on our roadmap is to capture a lot more of the telemetry industry by providing this additional value-added service to the data that they already provide to their customers. We don’t end there. This is a data business. This is a business that has the ability to capture many markets and be at a big business at the end of the day. These are the ones that we’ve done. You can see the direct applicability and going direct to fleets, direct to autonomous vehicle manufacturers and in general auto manufacturers.
Auto manufacturers like Tesla are increasingly looking to use data to help their drivers more safely navigate the roads. That’s part of their value proposition as a manufacturer that their car is going to be the safest car on the road. This technology built-in will help them achieve that. We’ll also help them provide insurance to their existing customers that’s more affordable and reflects the reality that our car is safer. Eventually, we may go directly to consumers as well. Where you’re receiving this information via Waze or Google Maps or any other mapping device, anybody that uses MapQuest anymore, TomTom, your Garmin, whatever it may be, that this information is given directly to consumers to help them make better choices and change their behaviors. There is a big market when it comes to auto and selling data into the auto industry. The auto market in aggregate is about 3% of GDP if you consider all of the players.
We have no problems finding buyers across multiple industries that if we are successful in selling into all of them, we’ll seed systems change. We will transform safety on our roads and be able to impact because we can sell to city governments meaningfully, but they can only do so much. They can only take it or change the roads. We can sell insurance and they can do their part to motivate behaviors. They can price your insurance a little bit differently, maybe change your behavior. We can sell to fleets because fleets are going to be able to route more safely and you won’t get hit by a UPS truck or whatever. Everybody has a stake in this and we see no problem selling our data to all of those stakeholders to achieve the end goal to get those crashes to go down.
My next thing with that is it’s fascinating. The one thing I can think of is how real-time or how much lag is there between the collection of data and being able to be able to put out throughput with AI. Because my thought process is you identify an area and you sit there and say, “Don’t go down this street. That’s a bad street.” Everybody reroutes themselves to a different habit. All of a sudden, where it used to be a bad street becomes a good street because no one’s there and the traffic accidents are happening three blocks to the left or three blocks to the right. My question is how quickly does the software update to be able to sit there and show either in real-time or within a week, a month, a year how up to date is that information? When you’re trying to make those adjustments and you’re trying to make those decisions, how quickly can you pivot when necessary?
What's transformative about AI is that it puts complex information in the palm of our hands. Share on XIt’s important to note how quickly you need to pivot. We immediately go to the, “I’m on the road at this moment. Should I turn left or should I turn right?” That’s an important decision in that second. In reality, the conditions on whether you turn left or whether you turn right, those roads don’t change all that frequently. If you turn left there in that minute or you turn left there an hour later or a day later, the road itself doesn’t change at that pace. There may be traffic congestion at the moment or there may be a pothole there that wasn’t there the day before. At a high level, unsafe roads stay unsafe.
For weeks or months or years at a time or until at least something is done about it.
What we see is seasonal change, to answer your question. It looks different in April than it does in June. It looks different in June than it does in December. We know all these fun facts about when it’s dangerous to drive. A few to keep your audience safer on the roads, if you can avoid it the day before, any holiday. Intuitively, we know that. We do see it manifest in the data two days before major holidays. Thanksgiving, Christmas, you want to be avoiding two days prior, not just the day before.
Last-minute Christmas shopping is out.
However, if you’re a last-minute shopper, the day of the holiday turns out to be the greatest day to be on the road. Fewer drivers, whatever it may be. It turns out, drive on the holiday if you can. The other thing that surprised me in the data is that it tends to be more dangerous to drive in the summer than the winter. It seems counterintuitive at first until you talk to insurance carriers who mostly confirm that this is true. They see more claims in the summer than in the winter. The few reasons people point to, definitely more traffic volume on the road. Summer traffic looks different than winter traffic. People tend to drive differently in the winter. They stay off the roads during weather, avoid unsafe areas or slow down significantly on tight corners or steep slopes. They don’t drive as aggressively in the winter. The physical road characteristics they’re more mindful of. I think also there’s got to be something to do with the college kids coming home from school. The type of driver on the road looks a little bit different as well. There are all sorts of reasons why, but some surprises in the data as well.
When you’re taking all this data to your potential customers, how do you turn around and communicate the value in a way that gets them to sit there and say, “I get this. This makes sense. I need to buy into this?” What you’re selling is above and beyond the thought process of a lot of people. It may not be in the industry, but for a lot of people, it’s above and beyond. When you’re dealing with C-suite people that are looking for this one-page dashboard, the people that are writing the checks, how do you sit there and say, “This is going to cost you X number of tens of thousands, hundreds of thousands to be able to implement this, but it’s going to save you millions?”
It’s something you’ve never looked at before that we all think you probably should. A few bits of advice, these are things we learned. I’m speaking from failing, falling on my face and getting back up to try it again. If others are in this journey as well of trying to figure out how to translate something that isn’t at first easy to understand something that is, try it a million and one times until you figure it out. It’s a circuitous path. As more people know about your brand, it becomes easier. The first thing I would say is we try to make it intuitive. We try to help people think about their own experience first. It’s to think about a road in your neighborhood that’s unsafe.
For me, I described the nature of a road that I’ve driven on. It’s funny because after I give speeches, presentations or sales pitches, we’ll be walking out and someone will stop me and say, “Do you know what road it is for me?” They’ll share that road. I know that it resonates with people in a way that they’re able to ground it. The second thing is to make your product intuitive. I think far too many AI and machine learning try to create these complex three-dimensional boxes and visualizations that are supposed to make you feel stupid. That’s the point. AI should be used and can be used and what’s transformative about it is it takes what we know to be true, but can only process in our brains, which are complex systems. It makes it that we can have that information in the palm of our hands and make decisions without having to think that deeply about should I turn left or should I turn right?
First, try to make it relatable. Second, make sure your technology itself is intuitive. Don’t try to do something complex for the sake of complexity. Third, I think the thing that we’re working on, if I was vulnerable about it, is to create a product that enables people to use it right away. To be able to enter their own address, see and validate that what you’re doing is grounded and makes sense to them, because that’s the key. That’s where folks buy-in because they not only understand your product, they understand the importance of your product. They can start using your product and start showing others and starting to spread the word and be your advocate for you when it comes to growing your audience. It can’t just be the people that you can physically talk to. You need to convince others that it’s worth talking to others about and make some virality to your concept. The way that you do that is by giving people the thing they need to go out and use to show to others. That’s where we’re iterating and trying to find the right fit for us too. We’re still learning.
That’s powerful to be able to put something in somebody’s hand and sit there and say, “Test it on streets that you know. Test it in your own neighborhood. Test it in the route that you take to work every single day.” If they’re looking at something that’s familiar, something that resonates with them, something that they know like the back of their hand, it’s intuitive and they can sit there and say, “I can see that this is right, based on the data that I’m going,” compared to say, “Here’s our demo from Los Angeles. I’ve never been in Los Angeles in my life,” or there or wherever, and be able to sit there going, “I don’t know if that’s true or not because I don’t know the location.”
I will push back on what I said too. Here’s where that goes wrong sometimes is that when people know their neighborhood well and they start getting into the minutiae of it and they’re like, “That’s not like that on Tuesday.” You’re like, “Okay.” Sometimes it makes sense to keep it in places that people don’t know because it also removes some of the personal nature to it. This is where we’re iterating because we’ve seen it go well and we’ve seen it fail for that reason. I’m letting people play, letting people use, letting people move around, letting people understand it, not just explain it. Put themselves in the shoes of the user and use it because if they can’t use it, they can’t sell it to somebody else to use.
That’s how I learned. I’m a hands-on learner. If you want to teach me something, put it in my hands and let me play with it. I’m not one of these people who can read a book and learn something. I’ve got to make it real to myself until I can make it real in one way, shape, or form. It doesn’t become ingrained into my psyche.
It’s important especially for data products. People who are selling data or machine learning, these are abstract concepts to begin with. It’s like selling a nebula. What is a nebula? Define the edges. Tell me how to understand its components conceptually, how to use it, how to make good use of it for your customers or for your company. That’s hard to figure out and it takes time and iteration.
Don't try to do something that's complex for the sake of complexity. Share on XYou had been an amazing guest. Two last questions. The first question is how do people get in touch with you? Is ODNsure.com the best way to get in touch with you?
Two ways, ODNsure.com is a great way. It’s our website. If you want to connect and look at information, there’s a map of the City of Chicago on our website that you can scroll around, zoom, click, and do all that fun stuff. If you do want to play, be my guest. Another great way to get ahold of me is LinkedIn. I consider that my platform of networking and connection. Not only to connect with me but to connect to my network because there are many people we’ve met in this journey from other startup founders to telemetry providers, industry players, thought leaders, executives. Folks like yourself that have been helpful on our journey, we connect to on LinkedIn. I always encourage folks to ask for introductions and network through me to others as well. Connect to me, connect to others. Check out Golfcour.se, which is a modern forum for business conversations and insurance. We invite folks who don’t typically get a platform or a place in the industry to share their experience, to explain a little bit of room to talk about what they do. We want to make sure that the insurance industry gets to know them a little bit, welcomes them into the family and help people make more introductions and get to know others in the industry as well.
The last thing I want to ask you, and this is something I ask everybody before they walk out the door, is when you get in your car and you drive away from a meeting and you’re no longer in the room, what do you want people to think about you when you’re not there and ODN?
For me as an executive, the CEO of ODN, they are one and the same. I want people to not only believe that we’re super freaking smart people. I think that’s an easy sell because we’re smart people. I want people to believe that we’re putting and manifesting our expensive educations and our creative brains to solve a problem that connects dots that have never been connected before. Because that to me inspires, that to me helps people feel like, “Not only are these kids bookish, they know the math, and know what they’re doing, but they see things that other people don’t see. They see the future and they’re dedicating their lives and what could be their time.” They could be earning millions of dollars for these incredible brains that we have. Instead, they are sacrificing family, friends, income, job security, family planning for a lot of women founders to delay or don’t start families because they are dedicated to their craft and what they do. I want people to believe that our vision is a vision that we’re trying to see the future that will leave people better off, which will powerfully make everyone in the room’s lives substantively better and safer.
For me, I want people to feel a part of it a little bit. I want people to feel like I might not be able to do that math. That sounds like complicated math, but it is. I trust that these people are taking their skills and applying them in a way that’s for good. It’s going to make the world a better, safer place to be. It’s what motivates. I know it motivates my team to show up to work every day. It’s what motivates us to work through the hard math problems. I hope that people appreciate that, but more importantly, support businesses like that. AI has gotten a bad rep from Facebook and from companies that have abused their privileges when it comes to your data. I want people to know that there are folks out there that care. Your data can not only be used for evil, but it can also be used for good and we’re dedicating their lives to seeing that future manifest. I want people to feel good about that. I want people to feel excited that people like me and ODN are out there and to support us. Tell others about what we’re doing because it is a positive and hopefully transformational thing.
We are all stronger together, Carey Anne. The work that we do together and we each have a singular part to play in this thing called life. You can do things that I can do. I can’t do it, and I can do things that you can’t do, but it doesn’t matter. If we all work together, we’re all going to come up with some amazing things. Thank you for doing everything you do to move the world forward. Thank you for all this insightful conversation, opening up my brain to new ideas and being a wonderful guest.
Ben, thank you. Part of the joy of being here is that you’re using your gifts to help folks like me spread the word. The thanks and appreciation is right back at you. Thanks, Ben.
You’re welcome.
Important Links:
About Carey Anne Nadeau
Carey Anne Nadeau is the Founder and CEO of ODN (ODNsure.com), which uses machine learning to measure where it is safe to drive and has established the first ever road risk rating score. ODN’s crash predictions and hazard maps help people more safely navigate the roads, cities to plan and prioritize traffic safety initiatives like Vision Zero, and insurance carriers to establish rating factors that are intuitive and improve profitability.
A rising leader in insurance thought leadership, Nadeau sits on the Advisory Board for Women in Insurance Leadership and hosts the podcast, the Golfcour.se, a modern forum for business conversations in insurance. She is a graduate of the Massachusetts Institute of Technology, with a Masters in City Planning and has previously worked on issues of urban poverty and economics at MIT’s Lab for Regional Innovation and Spatial Analysis where she remains a Research Affiliate, as well as the Urban Institute and Brookings Institution.
Love the show? Subscribe, rate, review, and share!
