Root, Inc. (ROOT) Earnings Call Transcript & Summary

Hey, good afternoon and evening to those in Europe. My name is Mike Zaremski. I'm super excited to have the opportunity to introduce Dan Rosenthal, Chief Financial Officer of Root with us. His colleague, Joe, is also with us, Director of Investor Relations. So for those of you who know Root and know me, for years now, we here at Crédit Suisse have been talking about the myriad of benefits, telematics-based auto insurance offers consumers. And having -- and Root having a direct-to-consumer-based distribution model sets itself up very well to be able to save millions of customers over the coming decade, who may drive less, may not use their phone a lot while driving, but basically, given the opportunity to save a lot of money on their auto insurance bills. So with that, I did survey a number of you attending right now and ask you how much you knew about Root. And I'd say 1/4 of you said you didn't know much. So I think with that, Dan, maybe we'll kick off our fireside chat with just a couple of minutes on what Root does, what -- why it was founded, what's its strategy?

Yes. Great, and good afternoon, good morning, good evening to those of you around the world. I'm Dan Rosenthal. I've been the CFO at Root for about 1.5 years, but been a member of the Board of Directors since the early years about 4 years ago. So thrilled to be with everyone today and talking about Root. We took the company public in October and starting on December 1 post our Q3 earnings, got out on the road and met with investors. We did about 70 One-on-Ones and met with over 100 investors, and we were thrilled to announce our quarterlies, the way 2020 finished just last night. Root, in a nutshell, is a mobile-based car insurance business that is the largest InsurTech in the United States with more than $600 million of premium in 2020 and guiding to 30% to 39% growth in that premium in 2021. Frankly, 3 things matter, I think, if you're trying to learn about Root. One, massive market opportunity, $266 billion market in the United States for auto insurance. It's a required purchase for most consumers, and consumers generally don't like their car insurance. We said last night in the shareholder letter, it's kind of like going to the dentist. We -- if you go to Google and type in car insurance, you get much stronger responses if your safe search is off. So that's the disruption opportunity that we saw. The second thing is our telematics. We think they're differentiated. Our first talent in telematics, our Chief Data Scientist was the head of telematics at Progressive. Our CEO and Co-Founder is an actuarial fellow himself. So we really do have the insurer in InsurTech. This is not a group of technologists who've decided to found an insurance firm, really focused on insurance principles. And that has impacted the development of our telematics. They are mobile-based. So this is not plugging the device into your car or the integrated telematics with the car. And mobile-based has multiple advantages. It's just much easier for the consumer. It's much more accurate in terms of just tracking distracted driving and hard breaking. We can tell who's driving. So we really apply telematics to the individual policyholder and not just the car itself and whoever is driving it at that specific time. So the second differentiator is around our telematics. And then the third thing to emphasize is that the disclosure last night is very clear that in our seasoned states where the regulators have allowed us to put in market our telematics model and the latest generation of our pricing algorithm. And for us, that's about 2/3 of our states, the business model is working. The loss ratio is 15 points better than in the unseasoned states just in the second half of last year, trying to remove any COVID impact that we saw from earlier in the year last year. So we can see the model working. And in the guidance that we put forth last night, we made clear that our goal is to extend into another 7 states this year to be at 85% of the U.S. addressable market by year-end 2021, but with improving unit economics. And we showed how we expect those unit economics to improve over the course of 2021 and beyond. So thrilled to be here, really limitless opportunity and excited to take your questions, Mike.

Okay, great. That was a thorough interim you touched at some points, I think we'll dig into. From last night's call, I think in the short run, the market is probably focusing a little bit on the loss ratio and maybe less growth potential, at least pulling back in the states. But I thought there were a lot of good tidbits of disclosure. And maybe let's focus on the first one you mentioned, which was seasoned states. So you said 15 points better in seasoned states. Maybe you can kind of define to people what that means? And then should we expect that just to continue to play into the portfolio given growth is not going to go back to 100% per year levels, as you guys saw a couple of years ago. So how can that kind of play into the portfolio?

It's fair. We had a year -- 2018, we had $106 million in premium and $450 million in '19. So we certainly did see those levels of growth. Obviously, we're on a larger premium base now. But I'll tell you, we don't feel like we're sacrificing growth. We're guiding to 30% to 39% growth this year, and we feel really good about the ability to do that at improved unit economics. So I think the key for us is as we -- we're very data-driven. That's in our DNA. And the thing about our business is we rolled our first policy 4 years ago. So we're learning along the way. And we now have learned that our pricing algorithm and telematics models are developed enough to show differentiation. To give you an idea of how fast this is happening, at the beginning of 2020, we only had 3 seasoned states. And we define a seasoned state as where our pricing model has been adopted by the regulator, and we've been in market for a year and been able to get at least 2 pricing changes in because that's what's happening. We're iterating so much faster on the pricing side than any of the legacy car insurance companies that we need the regulator to work with us and to allow us to put the pricing in the market. If they're slower, then you can't really evaluate how Root is performing. So the good news is, our regulatory relationships are in a wonderful place. We hired the former President of the National Association of Insurance Commissioners, who is a commissioner herself in Tennessee and Kentucky for more than a decade, to come in and run this aspect for us, and it's made a huge difference. So I feel really good about where we are with the regulators, but it's still fair to say that some move faster others. So for us, we really saw the data show that in the second half of 2020, as you alluded to Mike, our loss ratio was really materially stronger in the seasoned states. And so for us, that's a sustainable, repeatable growth. That's where we know the model is working, and that's where we're going to devote the majority of our marketing dollars, the vast majority of our marketing dollars as we enter 2021 and beyond. And then we'll also spend time investing in those 10 states that we're in today that are not seasoned, getting our pricing model in place in those 10 states, so we can begin pushing the gas pedal down on growth even further and then also expanding into the other states where we're not licensed today. And as we said, we expect to be licensed in an additional 7 states by year-end 2021, which will put us in 85% of the addressable market.

So Dan, I'll go back and do the math and help people. But I think if you want to mention it now, you said about 60% of the portfolio is seasoned. Based on your outlook and kind of where you expect to push on the gas, will that ratio change? This is before we talked about UBI for and other things, but will more of the book be seasoned? Or will the growth offset that?

No. We expect that more of the book will be seasoned. It's really done -- the seasoning's on a per policy basis, if you think about it. So 60% of the earned premium that we wrote in the second half of last year in the fourth quarter was in seasoned states. And we do expect that to grow. There will be new writings, certainly in some of the newer states, especially as we enter newer states. So you'll see maybe some short-term blips in that. But this is not a question of it. At the end of the day, we're going to have 30 seasoned states and 20 unseasoned. At the end of the day, we expect all of the states to be seasoned. It's just they'll come along at different points in time. For us, we have the pricing model. It's working with the regulator in that specific state to get it adopted. And as I said, those are very fruitful conversations. So over a relatively short period of time, we expect to have the vast majority of our stage B season.

Okay. I thought one of the other comments that was bullish, provocative everyone termed it was about UBI 4.0. So maybe you can explain to people what UBI 4.0 is? Why do you feel -- what are you seeing that can make it kind of 30% more predictive? I think that, that would mean right a better algorithm in terms of pricing and segmentation. And also, it's a long-winded question, how can we think about that being deployed? Is it just new business and then solely to existing customers? Is it -- and there's nuances in certain states, as you alluded to earlier. Yes, I can repeat them too later.

No. I think I got the gist of your answer, and if I missed something, Dan can help me here. But we're really excited about 4.0. I think as we discussed in the letter, we are a data-driven company, and we are consistently finding ways to improve our pricing. And I think 4.0 to 3.14, Alex disclosed last night that we're seeing significant improvements in the predictability between the 2 UBI models. And a lot of that is just off of more data. We can lock more data. We can find ways and more interactions between the data to really drive in the predictability of the data. But I would say what's really exciting about 4.0. And as we collect more data, is we can do more with less. And I think by saying that as in the past, maybe we had some data on a certain person as they were coming into our funnel, and we were going through the test drive, and we weren't as comfortable with pricing that. And these new models allow us to price people with less data and still achieve that same level of predictability, which is super important. It starts to eliminate some of that friction as people come across.

And -- Joe, that's helpful. So -- go ahead, Dan.

You asked the question, too, Mike, how will we roll it out? And so some of that will be a little bit state-dependent. Again, making sure that the regulator approves that new model. And then we haven't yet made a decision exactly how we will apply it to renewal customers, but certainly will impact new writings from the moment that the regulator lets us use it.

Okay. So I want to remind people because a lot of -- not a lot, but I think certain people listening are aware of Progressive Snapshot, which is mostly a monitoring basis of just a certain time frame. Can you remind us, are most of your routes, clients being monitored on an ongoing basis?

Yes. And I think that -- I get asked this question all the time, Mike, and it's a really important question because it can be confusing. You've got companies that do telematics on their own. You've got the Progressives of the world that partner with like a TrueMotion in their case. Certainly, AllState has acquired their own solutions. Some companies aren't doing anything with telematics. You've got this mix. The OEMs are -- the automobile manufacturers are starting to do things. So what is making Root's telematics differentiated within that is really important. I touched on earlier some of the hard breaking and distracted driving elements that we feel like we're able to measure, and that's a significant difference. I think the other difference is how do companies use telematics. For us, we -- and this is really important. We use telematics to drive pricing. We underwrite out 10% to 15% of our customers because we believe their driving score means that we cannot underwrite the risk. We just -- it's not a matter of charge them $50 more or $100 more. We just underwrite out the rest. And for us, that's important because 10% to 15% of drivers end up making up about half of car insurance claims costs. That's a big deal. So if we are successful in underwriting out the worst of the worst drivers, then it allows us to offer better pricing to everyone else. What's really unique, as you talk to other insurance companies who are using telematics, is nobody else is doing what I just said. In fact, most companies only use telematics to offer a small discount to their existing consumers. They don't surcharge. Progressive will surcharge customers based upon a telematics score, but that surcharge level is capped. It's not limitless. So they're not underwriting out the 10% to 15%. And again, it's that innovator's dilemma. They've already got the customers. They don't want to cause massive churn. They feel like they're managing the overall book. So that is really the thing -- if I leave everybody with one thing about our telematics, it's that what sets us apart is how we're utilizing the telematics. We think they are differentiated. We think mobile base is the right solution. But we are also utilizing them to drive pricing and underwriting in a way that no one else is doing.

And I would even add to that, like not only are we -- not only are we using telematics for pricing and underwriting, we're using that data across the company. So we're using it to help improve first notice of loss when an accident occurs. We're helping it to identify fraud better. That data is so rich and we're able just to leverage it across so many different aspects of the business that it really gives us an advantage, and that's why Dan says, we are a data company because we think it's important and it will really give us an advantage.

Dan mentioned those 10% to 15% of drivers that cause a disproportionate amount of loss. I think we've all seen those drivers on the road, and that kind of hits home.

I'm not judging you, but yes.

I definitely was one of those when I was 15 and 16 in Illinois, and I learned it the -- I learned my lesson the hard -- well, my mom learned lesson the hard way. So talk about it later. You mentioned, Joe, friction. And Dan, you mentioned also kind of this 10% to 15% drivers, you don't think there's a price to offer. When I thought of friction, I kind of thought that as you roll out more sophisticated pricing, maybe you -- what do you mean by reducing the friction? Because I think some people look at the retention ratio, currently, it is improving, but people think like if the friction is lower, will that allow you to quote more drivers or maybe shorten the duration of the quote process. Maybe you can kind of elaborate.

Yes. I think it's a little of both. With more data, we can price more drivers accurately. And I think by -- our goal is always to try to find ways to reduce the amount of time before we give them a quote off their drive because that is some people -- they want insurance today. We do have a product that allows them to do so. But there's also people that are willing to do the test drive, understand what their premium will look like and then quote. And we'd love to find ways to shorten that time because it only helps reduce that lag between the quote and when they sign up for the app.

Let's talk about marketing. Are you -- you guys have raised a good amount of money. You guys have already experienced significant growth, but any changes in the marketing strategy? Is there kind of a long-term also kind of ratio you're -- we should be thinking about, too, in terms of marketing spend?

Yes. We haven't disclosed a specific long-term ratio, but very clearly, we market based upon LTV to CAC, and that is the governing principle for us within Root in 2 ways. LTV to CAC drives who we bring into the funnel at the top of the funnel, and it drives our underwriting. Our underwriting is done also on an LTV to CAC basis. So we're highly conscious. Meaning, if we have a customer that we believe is going to retain longer and is going to bundle with our homeowners offering or our renters' offering and be a stickier customer, we can pay more to acquire that customer. We use a variety of differentiated digital marketing channels, again driving in and identifying the characteristics of the better drivers. We also use partnerships that are in place with Carvana, Chime, so many other companies that we've disclosed previously that are really important to driving conversion at, frankly, lower customer acquisition cost than even the marketing channels. And obviously, referral matters as well from our existing companies is a hugely important piece as we go forward. The other thing I'd emphasize, which was in our shareholder letter yesterday, and this may seem like a really basic concept, Mike, but it's core to us is that you can see that retention is better, the higher the telematics score the customer gets. So the key here is get the best drivers, they're going to retain longer. Why are they retaining longer? We'd like to think it's our customer experience and everything else. But frankly, price is the biggest driver for that. And the better drivers we're able to give the better price to. So this is really the key for us, is continue to home our marketing work around identifying the characteristics that make up the best drivers even before we get their telematics score. And the better we are at that, the better we're going to be able to drive LTV to CAC going forward. And we're really excited about some of the strategies that we have in place from a technology standpoint to accomplish that.

And Dan, right before this conference, we disseminated a survey we did of your customers, only 100 of your customers, but it wasn't easy to find a lot and also limited customers. To your point, there was a significant percentage of customers receiving a 30%-plus discounts from their prior insurer, which means, right, it shows that...

If you want to help identify more of our more than 350,000 customers, I'm happy to help with that. We did notice that. And I thought it was really interesting data to see even given the small sample size, price matters. And for us, we have initiated a Drop the Score campaign because fundamentally, we believe that a lot of the traditional variables don't accurately measure how someone drives. Credit score, education, occupation, are those really markers of the best drivers in the technological world we live in today, is that really the best we can do, whether you're single or married, male or female, age, credit score and the like? We think we can do better. And our customers respond to that. Not only do the best drivers get the better prices, and to your point, Mike, it's of material difference. They also really like being aligned with a company that is about fairness and about authenticity. You see the Bubba Wallace car. He's driving the 23 car for Michael Jordan and Denny Hamlin in NASCAR, and we have joined DoorDash and McDonald's, Dr Pepper and Columbia to sponsor Bubba. It's our first sponsorship. I don't see us doing a ton of sponsorships. I don't see us having talking animals and the like. I think we're really -- the fit with Bubba was it was about fairness and authenticity. And I do think you'll see us doing more of that going forward. Again, brand spend for us is going to be a much smaller percentage. We're not going to try and compete with billions of dollars a year. We don't think we have to, but we are going to focus in on the messages that are core to the Root culture.

Some of the -- your comments, Dan, remind me of kind of -- I believe it's your company's goal to wean off credit scores by 2025. Maybe you can kind of talk to you how tough that's going to be and how reliant the company currently is because I believe most insurers are very reliant today, auto insurers, especially on credit score. So maybe you can kind of talk through, is that when you get the UBI 7.0? And why is that important to Root?