Seize the Data with Carpe Data CEO Max Drucker

Seize the data!

In this episode, Max Drucker, the co-founder, and CEO of Carpe Data, joins Nick to discuss the revolution of data science and its impact on underwriting and claims.

Insurance has approached risk selection and pricing in a predominately experienced-based approach. “What is the loss history?” has been the driving foundation of our business model. But running a business with a loss history is like driving on a road using only your rear & side view mirrors. Without a forward-looking mechanism (your windshield!), you might be able to navigate some roads, but not all roads! This approach is by nature defensive; moving slow, playing it safe, because a wrong turn could lead you off a cliff. But with a windshield, the perspective of what you can do changes dramatically!

This is the promise of the data revolution! Combining the best of our collective actuarial science with forward-looking models that will create better risk selection, better underwriting, better pricing, and ultimately a better solution.

Watch here:

Connect:
Max Drucker (LinkedIn) – https://www.linkedin.com/in/maxdrucker/
Carpe Data (Homepage) – https://carpe.io/

Musical Credits:
Shadows by David Cutter Music:
https://davidcuttermusic.com
https://soundcloud.com/dcuttermusic
Free Download / Stream:
https://bit.ly/shadows-david-cutter
Music promoted by Audio Library:
https://youtu.be/qiBHOiEl9EI

Video Credits: Intro Stock Footage by Videvo

Transcript

Nick
And we're live.

We're back. This is the Coverager Podcast. My name is Nick Lamparelli. Today, the title of this one is going to be "Seize the Data". And I have the CEO and co-founder of Carpe Data, Max Drucker with me...Max, welcome.

Max D
It's great to be here Nick, thanks for having me.

Nick
We're gonna have an awesome conversation on data. And Max, I start all of these conversations by allowing my guests a brief time to do a proper elevator pitch and a proper introduction. So Max Drucker, who are you? What do you do? And why is that important?

Max D
Sure, well, those are a lot of questions. But the quick elevator pitch is that Carpe Data is an emerging and alternative data company focused solely on insurance. We look at the problems insurers face today, going after the really big issues and understanding that data has the opportunity to really move the needle for companies with hundreds of millions, billions of dollars at the scale of carriers. By focusing on two core areas, we focus on the data to drive automation. And we focus on data that improves insurance outcomes. And by keeping our focus, we can really dig in to what those major problems are today for insurers and provide them with new, unique data elements that they can't really get anywhere else.

Nick
That's a great introduction to this. I want to rewind the tape. Okay. There are a lot of companies out there that have "data" associated with them. So let's rewind the tape a little bit. Let's go back to the early days, and talk about what did you see? And what was the impetus for getting this off the ground and started, right, and then I think this is naturally going to, we're going to extend to I think a bunch of different areas on what you do and how you do it and, and why people should be paying attention. But let's start from the very beginning. How did it start?

Max D
Well, it goes back a little earlier than that back to that really the dinosaur ages. Because I consider myself to be an Insurtech dinosaur, because I was part of the founding team of e-coverage and e-coverage was the first online auto insurance carrier. And we sold the first auto insurance policy online, and that was in 1998, 1999 as we rolled out across the country. So that was then I had yellow hair, not white hair. And it really was an entirely new frontier. And that was a very exciting time and coverage blew up and in brilliant glory. But I got the experience and understanding how the world works in bringing new technology to insurance carriers. And from there was part of the team that created Steel Card, which was a policy admin company that was ultimately sold to Choicepoint and really the entity that you know, LexisNexis owns much of those assets today. And so that was really where I got that initial exposure to big data for insurance and understanding just how powerful and how critical these data elements are for insurers and how they do business. So fast forward to 2010. And I taken a little break at that point, from from the insurance space I had, I wasn't sure I was going to go back into the space and set out to approach a new frontier by going after thinking about my background in enterprise software, understanding insurance carriers but also really wanting something exciting and alternative data, new data, social data, data that's not really being looked at or leveraged by anyone else today. And those first few years, five years the company was frankly a huge struggle. It was, we really struggled to find our market fit, our product fit what exactly it was, what were the problems we were trying to achieve, until ultimately understanding that everybody, everybody wants to automate more, they want to have more consistency across the entire policy lifecycle.

Nick
Let me stop you for a second here. So let's look just let me pick it that just a tad In terms of the inability to get to product market fit, what what were your hypotheses that were failing? I'm, I think there's a learning element here for, you know, carriers and other Insurtechs as well, to sort of examine what, how you sort of thought about connecting data to a solution, and ultimately, where that where that kind of broke down, and then ultimately, when you pivoted, where that actually succeeded?

Max D
Sure. Okay. So that's a great question. I think it really broke down in two core areas. Number one, when we first started, we thought more broadly than insurance. And so we were focusing on the pre employment space, we had use cases in government and also an insurance fraud. And so as a company, we were spread very thin, and sort of really lacked exactly that right focus in what the right the core areas were. Our background was all having worked with insurance carriers. And we should have understood that at the very beginning. But sometimes you need to bang your head against the wall a few times to kind of realize this is what I know how to do. And this is where I've had experience and some of the other areas that were less experienced, and maybe let somebody else approach those challenges. But the second area, was understanding that carriers are not looking for an a small incremental improvement. Right. And you know, Henry Ford loves to tell the story that if you ask his his customers what they wanted, they would ask for a faster horse. And in many regards, I think that applies to a lot of technologies in that we talked to the potential customer and they understand what their problems are. And they're looking for greater efficiency. But understanding what's going to really changed that is is sometimes outside of the frame. And so we initially started in doing more dashboard focused areas, things that would create incremental improvements, but not really get the customer, the carrier all the way there until we really understood and learned and what they're really want, is being able to have that straight through processing, having systems, system communication, having the data elements, be robust, sufficiently robust that they can then have automated decisioning to whatever that given areas, again, across a policy lifecycle that can be in a claims process, whether it's about paying that claim more quickly, whether it's about paying the right amount for that claim, whether it's in a marketing use case, whether it's in an underwriting or pricing use case, all of these things need more consistency. And the only way to get there is by having that system of system communication, to pass it all the way through and being able to drive rules around what those underlying data elements are. And so we had sort of misstepped and thinking and believing that having dashboards and having better tools would make the process better. But that was really I think, an area that create a lot of learning for us enabled us to really dig into what the real challenges were, and what the customer ultimately was trying to achieve.

Nick
So how did you manage the trade off because there's, there's, there's a trade off of risk here that in the pivot, you're now creating something that's significant, that's not incremental, that could have, you know, significant impact to the bottom line. But you could also be building something that the industry is not quite ready for, from a technological standpoint. And, you know, I'll foreshadow a little bit a cultural standpoint! It could be too big to digest, how did you balance the trade off around that pivot, and potentially risking a marketplace that isn't quite ready for this real big solution that you're trying to offer?

Max D
By providing the best of both worlds? So what we'll do, we like to think of it as we provide new ways, and new methods and new data to solve existing problems. So that may be information that a customer has currently been using, to confirm claim facts to confirm information that's on an application, the traditional data elements, were finding new ways, using new methods and new sources to populate that data and provide that, that quality that veracity for those, again, those existing use cases. So that's really important. And effectively, any potential customer will digest that information, understanding it's coming from different sources, but the actual quality of the data is able to plug into what their existing framework is. But by providing that that also gives us an opportunity to develop what we like to call the rating factors of the future, the underwriting elements of the future. Areas that customers have not really looked at get are still predictive of insurance outcomes, and so they can experiment with those, while meanwhile, getting value on the existing promise today. So we're we're living in both worlds, we're providing information to be evaluated. And we understand it can take years for this to get fully implemented and up to speed, this stuff, the transition takes a lot of time. And we have to be thinking years in the future and understanding that that's where it's going. But meanwhile, we're able to provide extremely valuable and unique data elements that can get plugged into whatever that stage, that lifecycle needs to be for that customer.

Nick
So from from a data perspective, we're now you're now many years into this, what let I want I'm going to ask a dual question. First is in the companies you work with, what's the commonality of those companies that have had success? What have they done right, both in engaging with you, but also internally as a company? And also balancing? Hey, let's solve some problems. Now what let's look into the future. And then and then, and then I'm going to transition over to the industry and what we've potentially how we've potentially fallen. So let's start with the success. Let's start with companies that get this right. Right, they actually get that benefit from the data. What are they doing, right?

Max D
Sure. So maybe I can talk a little bit about our first really big successful customer, and that was with Allstate, and that was in 2017. Allstate understood that and the products we provide them or on the claim side, that their processes weren't as efficient as they would like them to be. And that the all of the other solutions in the marketplace that we're leveraging internet data, social data, all this giving to carriers, we're still creating that inefficiency. So maybe it could be outsourced, they might get a report of information. But a human being an investor still needs to wade through all of this data. And it was Allstate that really understood, we're really looking for the signals, we don't really care about the rest of it, we need to know what's relevant to help corroborate that claim, we need to know what's relevant to help us work that given claim, but everything else is just noise. And that was perfect for us. Because that's what we want to do is, is we want to be able to provide the carrier with actionable information and nothing else. So time is, is used far more efficiently and automation can be built into the process. And so they understood that, and that was very early on. And while some other customers that are smaller customers, were thinking that way, it was really Allstate saying listen, because this stuff moves the needle for us, right? Carriers of that scale, making these kinds of incremental improvements really have such a big impact, they wanted to have that big impact. And that was something that we were able to do for them. And it's been a tremendous and very successful & productive relationship over over the years as we've continued to develop new products, and work through that. And so in working with other customers that, that use our products that have happened along the way, they have those similar attributes, they are looking for a big impact. It's not about a 10 or 20% incremental improvement to reduce costs. It's about understanding, we have something that is a big cost center for us, or we understand that we're going to face adverse selection on some other type of data element. There are two types of different use cases. But we understand that this is coming. And we need to make a big move in order to avoid those situations from happening. And where Carpe Data is able to come in and provide those specific data data sets for those specific use cases.

Nick
Yeah, it sounds like a lot of the companies you work with are, are forward thinking already. Right there. They're eager

Max D
They have to be and I mean, we work with most of the top 25 carriers. And, you know, we talked about this industry being slow, and change is hard, and people are risk averse. And that is all true. Right. But on the other hand, most carriers are pretty aggressive. And at some level, carriers have been using data to predict outcomes for literally hundreds of years. Right? They are the original data scientists, they are the original companies thinking about predictive analytics. So what we do is not foreign at all. It's we come to them and saying we have a data element that will enable you to predict outcomes, specifically insurance outcomes, right and understanding that those specific problems for them and so I actually don't feel that this is a particularly stodgy industry that it's particularly difficult. It takes time, right but that's understandable and we're certainly patient and and work with customers in order to get that And obviously, we'd love it if it moved faster. But the scale and the impact is so important, it does need to take time. And, the talent and the expertise been one of our customers has more than 1000 people that fall into a category of being a data scientist, an actuary or a data engineer, or this is a scale of customers you work with, right? That just those skills. And so it's not that they don't have sophistication. Carriers, many of the carriers are very sophisticated. And so we appreciate that. And so we're not coming to the carrier and say we have a better way, or we know how to do something, or we have some special model and we know AI and you don't, that's not us at all, it's we have the building blocks for your process for your month, we will work with you, we understand what your problems are. And we understand the data that you don't currently have, we will provide those elements. So you can plug that into your given process, so you can achieve what those goals are. That's the profile the customer that we work with.

Nick
Okay, so, the carriers that might be struggling with this, let's talk about them. And let's talk about given how given how many carriers you've worked with, you have insights, you've seen success, right? You've seen carriers have success pulling this off, for the carriers that either haven't had success or haven't...they're, they're beginning that process to, you know, transition will, quote unquote, Titanic, right? Because it does take time. Can you give some insights as to from the companies that are successful for the, to the companies that are starting this journey? What's it going to take? What's it going to take internally, culture wise, technologically wise? How do they need to behave, even if they are eager to be able to be to have the same level of success?

Max D
I mean, I know this stuff, it can sound cliche. And you know, in the industry, we've talked about for a long time that that underwriting is it sort of two camps and underwriting is art and not a science, right. And then obviously, the technology providers, like companies like ours, make the case well, but perhaps in some circumstances, however, right, providing consistent data enables you to turn these processes and this is obviously how pricing is done. It's it is a science. And so we provide those consistent data elements to make that possible. So it's obviously first about identifying if it's a camp that feels that things need to be done manually. And there's only nuance and knowledge in order to do that, that's going to be an uphill battle. And that's going to take some cultural shift. And the company needs to make the compelling argument that look, we can do different things, we can be more efficient, if you can automate this that enables us to be better at other given areas and areas and in claims, for example, we may have been doing this manual process for a long time. Well, if we can avoid doing that manual process, we can now focus and dig deeper where it really matters and allocate our time or more efficiently. That's really how we think about on the underwriting side, the claim side, it's not about doing more with less necessarily, it's about spending your time better, and and being able to spend that time that you have on those areas that maybe will never be able to be automated perhaps.

Nick
Yeah, so I think culture is something that like a lot of us in the industry have focused on and you know, I think about underwriters and how they get trained, right, Max and so CPCU training, for instance, does not go aggressively into data science. I think the actuarial sciences do. I even believe like some of the certification designation, elements of the actuarial science, the Society of Actuaries in the Casualty Actuarial Society, I believe they have a data science component to what they do now. But a lot of the culture of the insurance companies is that actuaries do pricing and reserving. Right. And so that's an element that I'm concerned about is that a lot of the workforce, right, unless you're hiring pure data scientists, and even then if you bring them in, that could be a conflict with other elements. How do you think that should be navigated around? How do you think modern day insurers should be handling underwriters that may not have data science backgrounds and actuaries that do have exposure to data science, getting them more acclimated more involved in product development and other other things around underwriting? What should be done in that area? Because it's a lot of its workforce, right. A lot of it is what are the what are the resources within the company? How do we handle that?

Max D
I think it's a great question. I'm not, I'm not sure I have all the all the answers for that one. And what we do see does, it does vary wildly at companies, right? Some actuarial is very different than data science. Right. But but most carriers that we work with really do have a lot of sophistication and data science, and they understand and they're looking to build their bring these two concepts together, because they're you know predicting outcomes. It's obviously a some similar a similar discipline. I mean, of course, it just needs to be a campaign of demonstrating this is what works, this is what other people are doing. And this sophistication is the next evolution of what the carrier's trying to do at the end of day, which is predict outcomes, but then obviously, their regulatory confines. And but if I think you actually have some pretty good suggestions in there it is, it is a big, challenge. But I really do believe that the industry is moving in the right direction. And we look at industry conferences, we look at the focus, I think most people see this and actually embrace it, and really, really like the fact that we can use more and more methods to achieve the our insurance goals.

Nick
Yeah. So building blocks, right, so we have automation, right, which is a big component of what you do. And then you know, the big needle movers, right, the forward thinking stuff. Let me break it up into a couple other pieces, because you brought up claims a few times now. I want to try to differentiate how you think about that, how you've been doing it, how your product might be different. There, there are companies out there, again, with the words data that could be in their title that are, if I were listening to this, I would say I've heard this before, right? Like there are companies that will help you predict fraud. And so let's start on the claim side, how how would you label carpet copy data as different from others that might be in this space claiming to do the same thing?

Max D
Well, that the most important piece of the first piece is that the data that we're providing is data, that's just off the shelf. So we're not coming to the customer and saying, give us your loss history. And we're gonna build you a custom model for you based on your loss history, we obviously have those capabilities, right. But we also understand that most of the carriers that we work with also have those same capabilities. And so we are never saying we can build a better model than you can from the same data. So what we start with is saying this is data to predict a specific given area, whether this is somebody that's very unlikely to commit fraud, so you can now pay this claim more quickly, for example. We're getting to that from a broad array of data that the carrier is not working from. So we also also understand we know what carriers are using. And so what goes into our models or our data elements are, are intended to be additive and not duplicative it's really important if we know the carrier has basic demographic information about what that given claim it's about in most cases, so we're not putting basic things into this, the data elements come from places the carriers aren't working from. So they will see a unique impact of what these are. And these are lessons that we've learned over the years, as we understand what we do needs to be unique, it needs to be not something they can either do themselves, or they can get off the shelf somewhere else. This needs to be these unique elements. And I don't think in the in the claims marketplace, you're going to see a lot of other customers or other companies that are in this space saying yes, we don't need your data to predict you outcomes. I that's not something I think we see a lot of right now.

Nick
Yeah. It's kind of tangential question, what's the what's the craziest fraud case? That, that you've that you think that your your technology may have kind of red flagged? I mean, that's it because I hear I hear some crazy fraud cases. You know, so I'm always I'm always eager to like, kind of share how how these things work. Like what what's what's the technology, like in a real world example of something that at Red Flag that's like, Oh, my God, that's so interesting.

Max D
I mean, the story is, as you know, I mean, they come up like almost every day of you can't believe this, right? You couldn't believe it, right? If unless you were to see it firsthand, but one of the ones that I like is I think a pretty funny story is that there's a person that was without on Workers Comp, and they apparently caught a really big fish and I made a local paper. And so they were out, they caught some huge fish and there's a picture of them smiling holding this big fish, right? Well, you know, so in that, in that situation, that's likely an opportunity for that person to get back to work a little bit earlier than maybe they were potentially hoping for. So it's really more about situations that are exaggeration or abuse, people don't necessarily set out to commit fraud, right? Actual hard fraud is a pretty small component, but the the amount of opportunistic fraud or exaggeration abuse, you're comfortable with it, that's obviously a very, very big number. And that's where we really help our customers help their insurance kind of all right, you know, this happened here. There we go. All right, that was a great fish. Get back to work.

Nick
Because that's a that's a real world example. Right. And so it kind of shows what the technique where the how far the technology has come, that it can do those sorts of things. That's, that's, that would be a lot of manual work for a fraud unit to try to do alone.

Max D
Yeah, I mean, it's, it's virtually impossible. So what we really find with most customers that aren't using an automated solution like ours, they're either about half of the claims, they'll come in, they'll do their own manual searching. So they're Googling or they're, they're doing whatever that is, and that's obviously not consistent. And that can be very time consuming. And also subject to not necessarily fairness, but who gets searched and who doesn't. So half the people are doing manually, but the other half is entirely ignored. And so for our customers, either of those situations is not what you want to be doing, you want to have a consistent approach that scales. But you don't want to avoid this very valuable information that helps you with a given claim. And so that's where we step in is we provide that that outsource automated solution that that keeps people from having to spend hours or ignore data that could be very helpful in resolution, whether against corroborating that claim, closing it more quickly, or whether it's about helping work out some of the more complex details of it.

Nick
Okay, so let's transition over to the underwriting side. It'd be great to kind of go down the same path with what we just did with claims with like, what are some of the real world examples on the underwriting side that your technology is streamlining, automating or providing that lift, right, not the incremental thing. But hey, this by plugging this in, we're seeing a significant difference in, you know, loss ratio or outcome in some way. Talk about underwriting and Carpe Data.

Max D
So our focus today has been on the underwriting side has been in small commercial. So we built a data set of more than 45 million US businesses that we have that we make available in a sub second API that breaks into really three core buckets. The first bucket is for classification. We all know that carriers and I think this is somewhere 20 to 25% of all policies are misclassed, for whatever the given reason is they didn't know what that was, it came in that way from an agent, it may have been just entered, "lawyer and landscaper" popped in whatever that was, there's a lot of misclassification going on. So our data set using a very broad array of data sources that carriers aren't systematically looking at today, to understand what that business, what kind of business it actually is. So now they can automate that verification of what that of what the correct classification was. The second bucket component ofwhat we're doing is we have business characteristics about these businesses that are specifically what the insurers care about. And we built these out segment by segment. So it's a nail salon that they do waxing, tanning and the other given areas, some carriers that can be a knockout question other carriers that can be an area to provide additional coverage. Of course, areas like restaurants, do they have a balcony? Is there outdoor seating? Do they have happy hours? Is there a bouncer? Contractors...the kinds of things they do they do welding or other risky areas. Landscapers? Do, they trim limbs. So we have built these out business by business segment by segment, so that a carrier can have confidence in understanding that this indeed, is an attribute that they have that they can now automate around, because that's the name of the game. So an underwriter doesn't need to go to the website, or call that business or ignore this information, they can get that validation of these attributes that we've built that so that's the second part of this, of this data set. And again, this is these two components classification and the characteristics are really about driving that automation, right. These processes that are either are inconsistent, they are not necessarily accurate, but they're time consuming. We provide that ability to, as part of our mission to get small commercial as close to auto insurance as possible from an automation perspective, it's about chipping away at piece by piece, right? It's not going to be done overnight. There's no single data element that makes that possible. But I think most people in the industry see this is what the trend is, we're automating more and more and more small commercial process. And so this is how we get there are new ways to solve existing problems, classifications and legitimate these elements are. But the third bucket is we have built an array of indexes that are what I was referring to before as the new ways to solve existing problems, right, we're looking at things like the customer ratings, we're looking at their reputation, and we're looking at their online visibility, health and sanitation, property and safety. So we have built indexes scale one through five, based on elements that are that come from that business's website, they can come from the customer rating sites, they can come from a very good can come from reviews, and we extract that data put those into the indexes. So that carrier now has the selection ability to differentiate between Joe's cafe and Harry's cafe, because they might look the same on paper, but their risk profiles might be wildly different, but they have no way of getting to that outside of these new data elements. And and some of the things might might seem intuitive, but indeed a body shop that has terrible reviews, is indeed more likely to have a loss within 12 months, then a body shop that has great reviews. And so it's it's our mission to distill this into actual data points and plug into these models. And and that provides our customers that are currently using our capabilities, a short term, a clear competitive advantage, because they're avoiding the bad risks, they're able to avoid those bad risks and, effectively push those bad risks out into the marketplace to buy insurance from from their competitors. And longer term is becomes a very, very important part of selection. I now understand where the holes are, where the good risks are. So I can prioritize whether that's in pricing, whether I can pass more through and having this window is and frankly, again, I think it's very intuitive, and frankly, in many ways, obvious that this data is predictive. But the challenge has been how to how do we get to a cure, so they can consume it easily. And so now they can they can get that from our real time API. It's scaled one through five, they can see the progress of this business over time, and plug that in and and then issue that policy.

Nick
Yeah, it's my my favorite examples are the the roofers, when I was on the agent side, like the extent that they would go to, to say, Well, you know, 90% of our staff is in the office, only 10% of our business is roofing, but they're a roofing company, or they claim. Premium leakage is huge problem that kind of flies under the radar, it's it's almost another form of fraud, and in some cases can really be harmful in casualty lines, you know, Workers Comp and general liability. Because it could you could be taking on an account you don't want. And so it's better to it's better to know upfront.

Max D
I mean, it absolutely and, and it really to, we now live in a world which data is constantly changing, and we can access if we don't need to wait years to look at this data, right? data is continuous, right? And the idea of, of almost underwriting a look at something at a single point in time, right? At some level is an antiquated concept. Because clearly, we can now proactively and continuously underwrite now that stuff is not happening tomorrow. And and I'm not going to come to you and say, Oh, yes, our customers are continuously underwriting now and making changes in real time. That's not we're not quite there. But that will we will get there that, the industry is going there. There's no reason necessarily why they look at they re-underwriter business every five years. Right? That doesn't make any sense. But we've talked to a lot of customers, they say they arbitrarily it's this kind of company. We do it every three years. It happened here is how we do this. Those days, I think will ultimately be over with. The availability of new continuous real time data to give that carrier a window into what's really going on with that business.

Nick
Yeah, I almost think the word continuous underwriting is the wrong way for carriers or underwriters to think of this. I just kind of view it as, as much as the actuarial approach is highly valuable and proven. It's it has proven itself over hundreds of years. There's there's a flaw, and I wouldn't call it a flaw. There's something missing, right, which is you have to wait for claims to happen. And in that intervening moment, there's probably a lot of telltale signs that a claim is going to happen. It just hasn't happened yet. And to me, I sort of think of it as backwards looking and forwards looking, it's driving a car. It's Look your drive, viewing it through the windshield. But it is helpful having all of the mirrors around you, I think of the mirrors as like the actuarial element, the mirrors the rear view camera, like the sensor that says, hey, there's a car on your side. And then the forward looking view that, hey, that I see something happening up ahead, I'm going to start to break type of thing I, to me, it seems that like that way, like the the data will be coming at us. And it's up to the carrier to figure out how continuously they do it. But even in a modern underwriting book, where they're looking at a policy once a year, if they have that forward looking view, that's still to me, leaps and bounds better than a, an entirely or solely actuarial based approach.

Max D
I'll give the example that I mean, I couldn't agree with you more an example can be because we're we have this data set that's constantly updating and growing and evolving,

for example, starts to go down significantly. Well, I think it doesn't it reasonable expectation, but there's something going on with this business. And it may be ready for an intervention that might be able to help that business or that carrier, at least understand something is happening here. And so they need to be aware of it. And so this is I think what you're saying, and this is where it's going. And this is where again, the data is now available, we can now do that, we can see that Joe's Cafe is having great reviews, and they are expanding, right based on this finite data. And therefore maybe they might become a retention risk, because maybe they might be shopping because they're needing more insurance. And so this is an opportunity for them to reach out potentially, and make sure you can keep them as a customer or sell them additional products. As you're seeing that expansion. However, that is the date it is now possible. It doesn't need to be done at these arbitrary intervals.

Nick
Yeah, that's to me, Max that's the most exciting part because I think, what carriers are good at what, actuaries are good at and ultimately, what underwriters are good at is segmentation of market, you take the ability to take two different exposures and say, here are the difference between the two. Right? And the more that we bucket those into buckets, where it's like, I can't tell the difference between the two. That leaves you vulnerable to a competitor that can separate those two. Right? And I think adverse selection is that you know that very sly, it's like that bad pneumonia, that you have it you don't know you have it. But there's something wrong with you. To me that's like adverse selection, you know, it's company's inability to segment their data is, to me the ultimate killer.

Max D
That's right. And and what I love about the space, too, I think is exciting, as it carriers are always looking for new ways new insight into these given businesses. That, again, that's the history of insurance. And it's exciting to be looking to think creatively about what are these new elements that will enable them to segment in different ways than they have previously before, right, in better ways better than their competition, which is ultimately the name of the game, and continuously provide that in a easily accessible consumable way.

Nick
Yeah. This is going to seem new, like a lot of this is going to seem new or feels new to folks that are in the industry. But you know, my opinion about this is like, this is the way it should have been. And I don't see any path forward for companies that survive that are more data centric, that aren't, you know, doing this sort of thing on a more intense basis, versus the way that we've done it before.

Max D
Well, I sure hope so. Obviously, we're, you know, believers, as well. And also what I think is interesting and fun is that these are interesting challenges. I mean, in completely it's about solving new problems, and insurers understand that. And that's what I think makes it fun to being in these engagements. I think it's fun to be part of working through in any in the evaluation process. I think it's fun for my team, again, as we think creatively about whether what what data will work or what won't, let's see if it works. And then let's then put that into a product that then can be consumed.

Nick
Yeah, but Well, from a from a workforce challenge. You know, we insurance has a talent gap. It's not a primary area that graduates want to go to, they don't understand it. Right. And, you know, they see technology and finance and Data Science is sexy, right and exciting and stuff they want to get into. And that's to me, that's exciting for insurance. Because if we can rebrand ourselves as that sort of industry, then we will be a home for graduates again, like, they'll view us, like they view tech companies. You know, I already feel like Insurtech 1.0. And now Insurtech, 2.0, are starting to do stuff like that, you know, those, those tech companies like yourself, you know, can go to college grads, and just say, look at the problems we're solving isn't this interesting?

Max D
I mean, and, and for us that we're in based in Santa Barbara. And we have UC Santa Barbara, which has a great actuarial school, we had great success with graduates out of that program that worked for our team now that are really interested in these challenges. Absolutely.

Nick
So I'm going to lighten the mood a little bit. Before we we end our conversation, the bracket in Carpe Data, what does the bracket signify?

Max D
Well, it's the opening of an expression. And so it's an open bracket that opens up the expression and encompasses all that within it. And we liked it, because it was kind of, you know, we do a lot of data science. And we do a lot of a lot of math in what we do. And, and we write a lot of code. And code is starts with an open bracket for you know, most syntax do. And so it felt like that it felt like an expression, obviously with it in math, and also looks a little like a "C", so it was a really sort of it kind of came together and it was a total. We loved it. And we continue to embrace it. And more and more it's about that bracket. Because again, it's all encompassing, this is where you need to be within the bracket, we have all of those elements in there to, to solve the problem and to solve really big problems is what we're going for solving big problems, right, the big bracket.

Nick
And it sounds to me, what you're abstracting is that there's no close bracket. The process is never ending

Max D
No close bracket! That's right. It's not about the closed bracket, right.

Nick
The process never ends!

Max D
That's right. That's absolutely right.

Nick
Okay. So I'm gonna have my, I went through your website, there's an entity called bojack is Bojack, the muscle that makes sure the bills get paid?

Max D
That's right. He's, our chief adorable Officer. Bojack, the Rottweiler he comes to work on Wednesdays, he was named after the show Bojack horseman, he was the inspiration...

Nick
Is his last name, Drucker, it is okay. We, I will make sure we get a picture of Bojack up in the show notes because that was that was adorable. So fitting

Max D
Just don't tell my home ensure that I have a Rottweiler.

Nick
It's too late that they probably use your software. So they're gonna detect this, I'm sorry, to your point. And I believe, you know, given what, given what I talked about about training, I believe you have a fall event that's coming up. So why don't we give that a little publicity as well, because I'm all for training. And I'm all for what we you know, improving what we do. And if that means that underwriters and adjusters and others, insurance professionals need to become data scientist, or close to it, or at least trained in it, I'm all for it. Tell me about it.

Max D
So we just announced yesterday, so thank you, we're really excited to be conducting data in paradise in our annual customer conference. And we do this in Santa Barbara. We had to skip a year of course, and we're back at it of this October 24 and 25th. We have technology's most fear journalists Kara Swisher as a as a participant and speaker at the event, but it's really about our customers working together. And that was ultimately the idea of why we want to do this. We want our customers to come together to work together to see how they're working with us seeing what's working and create a community around the use of our products, but also create a place for potential prospects to come and understand a little more about what we do and talk to our existing customers about how they're were defining success and, and how they can better utilize the types of products that that we create. And so again, that's in Santa Barbara, October 24th/25th. It's at the Hotel California, and it's going to be a really fun and informative and in a great time when our first event in 2019 was tremendous and I expected Good to be there, Nick,

Nick
I was going to ask you, so we'll we'll take that offline. But it sounds like it's you can check in anytime you want. But you can never leave, or you can check out anytime you like. But you can never leave

Max D
something like that you're not going to want to leave it again, it's a good time, you learn a lot. It's a it's a right when you know, winter starts to come. And so it's it's pretty nice here that time of year. But it's really ultimately again about the community. It's about the users people. We love creating that collaboration of people that are currently using our our products, potential prospects, people that that are maybe are going to be using them shortly. We have our team there. We have demo bar there, we have partners there. And so it's just really again, creating this community.

Nick
Yeah, and what's really interesting is outside of I would think auto and homeowners. I think where as an industry, we were less competitive than we think we are, right? In a lot of cases, it's, we can collaborate more, and we should collaborate more, because some of these problems are gigantic, and no one insurer can solve them anyways. So I love that I think there should be more collaboration, because I think that will, collectively I think we'll get better economics, and society will get a better product.

Max D
I mean, I couldn't agree more. And our customers will often say fraud is an industry wide problem, right? We all need to work together on that it's not about we have a special advantage against fraud, that I mean, it's important to root out as much of it as we possibly can. But that's right consistency of data, right accuracy of data, this that's really important. We should, as an industry, we all want that. So we can all work together on that. And the secret sauce is what you do with the data, right and the secret sauce of the decisions that you make. And that's important that carriers have that but but but but but the big things it is about coming together to solve these problems together. So again, individually can apply your own know how your own algorithms, your own models, to create that competitive edge at the carrier level.

Nick
I've seen a great cross section of the insurance industry, it would be really difficult to find two companies where you would line them up side by side and they have all the same wishes and desires, the same risk thresholds, the same investors...like To me it's it's your culture, who you are what you're trying to accomplish. Everyone's trying to accomplish something just a little bit different. It's that's the secret sauce. Not this other stuff. I think there's plenty of room for collaboration. So I will put the information about the event in the show notes, of course with a photo of Bojack. And so Max Drucker...seize the data!

Max D
Thanks a lot.

Nick
Appreciate it.

Max D
Have a great day.