View all World of DaaS Episodes

Scott Stephenson (CEO of Verisk): How Data Sharing Transformed Insurance

July 22, 2021
Share this episode:
FacebookFacebookFacebook

About the episode

Scott Stephenson, CEO of Verisk (NASDAQ: VRSK), talks with World of DaaS host Auren Hoffman. Verisk Analytics is a $28 billion market cap data analytics company. Scott joined Verisk from BCG in 2001, was promoted to CEO in 2013, and doubled Verisk’s annual revenue to almost $3 billion since then. Auren and Scott discuss Verisk’s founding as a nonprofit, its role as one of the first InsureTech companies, how data sharing transformed the Insurance Industry, what went into building a successful data contributory model, and more.

Guests

Scott Stephenson

Scott Stephenson

CEO
Verisk

Episode Transcript

[Auren Hoffman] Welcome to World of DaaS, a show for data enthusiasts. I'm your host, Auren Hoffman, CEO of SafeGraph. For more conversations, videos, and transcripts, visit safegraph.com/podcasts. 

Hello, fellow data nerds. My guest today is Scott Stephenson. Scott is the CEO of Verisk Analytics, a $28 billion market cap data analytics company. Scott joined Verisk from BCG in 2001 and was promoted to CEO in 2013. And since then Verisk has doubled its annual revenue to almost $3 billion. Well, Scott, welcome to World of DaaS.

[Scott Stephenson] Thank you, Auren. It's a pleasure to be here with you.

[Auren Hoffman] Verisk was originally called the Insurance Services Office or ISO and was founded as a nonprofit, in 1971, to serve kind of a consortium of 100 or so various insurance companies, right? Why was it originally structured as a nonprofit?

[Scott Stephenson] So it actually was about 250 to 300 insurance companies that were a part of putting the organization together. Just a quick bit of context which will help kind of clarify how we got to where we got to, but also make an important point about data. One of the things that can stimulate data to flow is in fact regulation. And that is the point of origin for our companies. So even prior to 1971, the way that the insurance industry worked in the United States, and still does today is that regulation occurs at the state level. And each regulator says to the insurance companies being regulated, I want you to give me some very specific, fairly granular operating data. I want you to make it available to me, that's my requirement. They say two additional things around that. One is they say to the insurers, I don't want you to give the data to me directly. I'm going to name someone who stands between you and me. 

[Auren Hoffman] An intermediary that makes data easy for me to consume, etc. 

[Scott Stephenson] Exactly, exactly. So cause the report from Company A in form to be the same as Company B. That's the first. And the second is some data cleansing and assurance, some assurance that the data are accurate. So that's the first thing the regulator says. Don't give it to me directly. I will name somebody in the middle. And the second thing they say is you insurance company have to pay for this thing that's in the middle. I'm requiring it of you. But I'm also telling you, you have to pay for it. So that's been true for a very long time in the insurance industry. So the insurers were paying for this activity. It's called, you know, stat agency, stat reporting. They were paying for it forever. In 1971, I think it was around 280, actually was the precise number. These 280 insurance companies said, hey, you know what we're paying for this. Can't we do it more efficiently if we take each of these entities which are at work in these individual states and pull them together into one national organization? 

[Auren Hoffman] There was like a California organization, a Delaware organization, a Florida organization? 

[Scott Stephenson] Precisely. 

[Auren Hoffman] Okay, got it. There's probably a lot of overlap between them, you know, they have somewhat different requirements.

[Scott Stephenson] No, no, you're actually right on both counts. So the theory of the case was, well, rather than doing it 50 times, what if we just consolidate it? And, you know, kind of the guts of the data aggregation, the data cleansing, what if we just did that once? Wouldn't that be more efficient? And indeed, that has proven to be the case.

[Auren Hoffman] And so it was like a big ETL system in a way?

[Scott Stephenson] When it got started, right, and then yeah, I think that's pretty accurate. And then the other part of it is that the analysis is going to ultimately, why are these data being put forward? The regulators want to make sure that the insurers are solvent, and they also want to make sure that the public interest is being served and so a lot of the data is kind of at the level of what are you charging for this individual policy in this place for this kind of insurance coverage? And then what were your actual loss experiences against that same premium? So it's pretty intrusive, it gets to that level. The theory of the case was okay, well, so I'm trying to analyze the dynamics, profit and loss of personal automobile insurance in Nebraska. Well isn't it probably the case that the independent variables, the drivers of the outcome, the causal factors, isn't it likely that it's kind of the same in Kansas as it is in Nebraska? And maybe as it is in Missouri and as is Nebraska?

[Auren Hoffman] Also, if it's the same company that happens to be registered across states, you kind of want to know, they’re solvent across all three, right? Because if they go bankrupt, it's gonna mess you up in the other ones, right?

[Scott Stephenson] Right. But it's also the case that any one company's underwriting methods are probably consistent across states. And so the feeling was also that there would be gain analytically. You should start out understanding the dynamics of personal automobile insurance, or homeowners insurance or, you know, small business, general liability insurance. And then you can find the differences by locale, as they apply, and isn't that analytically efficient, and that also has proven to be the case. And so in 1971, the insurer said, let's not do it in all these separate places, let's do it in one place.

[Auren Hoffman] When all this data starts coming together for insurance companies, it seems like it would solve a lot of problems, like if you can imagine back in the day with fraud, you know, you may have someone submitting, like a fraudster submitting a similar claim to six different insurance companies. And, if you have a central place that's understanding the claims, it would be able to see that these claims look similar from the same person, and it would be able to flag it somehow, right? And so it seems like there's all these really great downstream effects when you have these, in some ways, kind of cooperative of insurance companies. Did that start to happen pretty quickly?

[Scott Stephenson] No, and you know, the pattern of thinking is the right pattern. But you have to add one other thing in here. And that is the data that we make use of, and this is an important thing for anybody that's a data oriented person as you are and I am, and anybody that wants to build a business around it. The data that we use are permissioned. And so the permissions tend to be specific about the use case that is at work. And so these data sets that I was just describing were specifically for the purpose of understanding underwriting practices. Are they in the public interest and the solvency of the company? 

[Auren Hoffman] Because there's a limit to how you can use them or what you could use them for. Ok.

[Scott Stephenson]  No, exactly right. So however, observations about the pattern of claims is a very useful thing to look at in the insurance industry, because one of the besetting issues in the insurance industry is fraud. So in this data set, the granularity on what is going on inside of the claims is not very great, because it doesn't need to be highly granular to serve the purpose for which that data set was put together. So we went and built a second data set only for the purpose of trying to root out fraud in the claims flows. And I'll add one other thing, which is, you have to pay a lot of attention if you're dealing with data, which crosses over into consumer space. You have to pay a lot of attention to things like the Fair Credit Reporting Act, which carries with it as you know, a requirement that you be able to respond to anyone whose data is in your data set, if they want to investigate. Well, what are you saying about me? And that's good public policy, right? But that first data set that I was just describing, the one that is kind of statistical by line of insurance does not need to be FCRA compliant. You want to think about segmenting your data sets according to the use case, permission, but also governing regulation as it relates to data privacy or your responsibilities as somebody who aggregates data. And so at the intersection of all those considerations, you can end up building two data sets that are 95% overlapping, but you want to do it twice because of these differences.

[Auren Hoffman] Yeah, that makes sense. Now, when you think about the history of Verisk, it reminds me a lot with the history of Visa, you know, like Verisk Visa kind of originally was like a not for profit, it was owned by the member banks. Are there other types of analogies if we look through history and stuff like that? 

[Scott Stephenson] Well, it's a really good question. I have spent the last 20 years searching for vertically oriented proprietary content. So content, which is very powerful in explaining what's going on in some specific domain, some specific industry vertical. The logic of our company is vertical. We serve in the insurance industry, in the banking industry, in the energy industry. And the data sets are very different. And in most cases, the most proprietary data that we've got is made available to us by our customers who then turn around and use the software and the analytics that we put together around all that to make it valuable for them. So that's the basic backdrop. There is at work, within the industry, a degree of data sharing among the participants in the industry. You've already pointed out the example that is the most prominent one that everybody is familiar with, and that would be in the world of credit cards. You know, there are the two primary networks, Visa and MasterCard. They're implicitly data consortia in that the banks that work with them agree that those two organizations will do some analytics. And then you've also got the credit bureaus. Now on the one hand, you could say that is contributory in that the providers of consumer credit solutions are making the data available to the three bureaus. But I don't think it's very much of a stretch to say that we, as consumers get so much speed and reach in terms of our own transactions, because you have these methods in the background, so that we can show up.

[Auren Hoffman] It's amazing how fast you can get credit today. 50 years ago getting credit took forever, and in many countries that don't have this kind of Experian, TransUnion, Equifax world, the only way to get credit is to know someone at a bank.

[Scott Stephenson] Yeah, although I will say almost every country in the world has acknowledged the value of what we're talking about here. And so, in some cases, you actually have governments that sponsor something along these lines, so, it definitely has pushed its way out. They may not be as developed as what we've got here in the United States. But, a lot of countries basically say, yeah, I want to be a part of this there, you know, this very high velocity, high reach facilitation of transactions. There are a couple of other industries that have a degree of data contribution on the part of the players.

[Auren Hoffman] Let's dive in because I think this idea, I think you use the word contributory data models, these are really economically powerful things. But they're a little bit different. Like, I don't know that most people study them in business school or something, even though they are so powerful. So I think it's worth diving in a bit.

[Scott Stephenson] They're also hard to achieve. That's why there's not a list of 30 industries that have big examples, the same way that Verisk stands as a big example in the three-industry. In other words, I've gone looking for the other industries, and there is no industry vertical on the globe, which is as comfortable with and as developed with respect to data contribution, the consortium data model, as the United States property and casualty insurance industry. There is no vertical in a locale, which is more developed, more evolved in this way than the United States property and casualty insurance industry.

[Auren Hoffman] Sometimes these contributory data models are the product. But sometimes they can be like a piece of a product. Like if you think of Gmail, like everyone's contributing spam signals into Gmail, and that makes the spam filter much, much better. So it's a contributory data model. Maybe the reason you use Gmail isn't because you want a good spam filter. That's like a nice feature. There's lots of other reasons why you use it, but that feature is really important. And it is solved by that contributory, and then they're doing a bunch of other things like anti-phishing, and other types of things that make the experience much better, because they have all these users. And in some ways, if you think of Facebook, Facebook is really just a data coop. It only works because everyone else is using it. And then you could start seeing LinkedIn in some ways as a data coop. Right? It benefits because everyone else is using it. So on the B2C side, it does seem like there's a ton of different things, and obviously very hard to get going on the B2B, much less so, maybe, because on the B2B, it's just a lot harder to get it moving or get it going.

[Scott Stephenson] I would actually characterize that one in both directions. You're absolutely right that in the consumer world, our demonstrated behavior is that we really value access. I'm saying, as individuals we value access to whatever it is. And we've demonstrated over and over again, that we're essentially willing to trade off the privacy of our data for whatever we get for that, whether it's I'm in the network, or it's simply being in the network, or some of the features that I can access because I'm in there. But we have shown a propensity to say “Yes, I lean into that”. For a business, I don't think people think deeply about it, when you're essentially giving permission in some digital environment. I mean, how closely are any of us reading the fine print? 

[Auren Hoffman] There is always some sort of implicit trade in ways you're giving your information about whether you're in traffic to get the collective information back and help you route and stuff like that. And you can imagine all these different scenarios where I want to know if I'm getting overcharged for my software subscriptions, or something like that, and you can imagine a collective thing where every company, opt in their general ledger into some common system, and then it comes back and it says, oh, you're getting overcharged or undercharged, or if I'm getting a new vendor or new a new customer, and they say they're gonna pay me in 60 days, like, do they pay on time? So you can imagine these like, general ledger coops happening. And there could be all these different types of things that don't exist today, that could potentially happen if companies were willing to auth their data and get some sort of trade back for it.

[Scott Stephenson] Yeah, exactly. And so then to kind of continue on with where you had us there. The fundamental basic point is that companies are not like your 13 year old, who finds some shiny objects out there in the digital world, and says “I'd like to have that” and thinks almost nothing about whatever monitoring tracking signals are being derived from their own journey.

[Auren Hoffman] Maybe the 13 year old thinks more about it than the 50 year old.

[Scott Stephenson] Well, let's hope so. But companies are not like that. Companies have been trained, I would say particularly over the last 10+ years, to know that data is an asset. Every company is at some stage of a digital journey and why are they on a digital journey? Well, partly for efficiency, but partly also for the precision of decision making that can come from having great data sets, which are well understood. So now,  general managers, commercial people, not just technically minded people, everybody knows data is an asset. And so it's going to be a longer and definitely deeper conversation with a business to say “here's what I would be able to do, if you and your peer companies made your data available. Here's what I can do, are you interested in doing that?”. That's going to be a considered decision on their part. And that's the fundamental difference.  Show me your technical environment, show me that you're going to keep the data secure, sign a contract, which guarantees my data will not end up in the hands of my competitors, a lot of stuff like that. And that's why it's much more deliberate getting the contributory data model put together, where businesses are the ones that are contributing to data.

[Auren Hoffman] These companies that have some sort of contributory data model or coop. In some ways they essentially become like a common good, right across the industry. And often, because of that, they get rewarded with some sort of, “winner take most” status. But there's an interesting relationship they have with this industry because they have a lot of power, but they also have to maintain their long term longevity within that industry. And they can't just raise their prices 10x or something because they have to serve that industry in a long term capacity. So how do they think about the long term, their constraints and how they work within their industry?

[Scott Stephenson] I can certainly report for us that our context is sustainability. First of all, I mean, just in every respect, you know, the sustainable performance of our company. And it is deeply in our DNA that we're here for our customers. I start with those two things. It's about the customer and the value they're getting. And we're trying to always render things in a long term perspective, and then to add one other point, and that is that with our customers in the verticals that we serve, we don't have just one solution, we have many solutions, many different data sets, many different solutions. And so if we were to take any one of those, and behave in a way that our customers would say “You're exploiting me, in this solution you are the center inside of our industry in this way. But man, you're squeezing me on pricing, and it seems to be all about you, not about me”. Well, first of all, the first order response from the customers would be “okay, well, on anything that you do, where there is someone that's competing with you, like, I'm now going to look at, I'm going to go look at them much more deeply. Because I just cannot put up with the way that you're behaving over here, where apparently you have a lot of market power, but you're abusing it”. So they have that particular way of responding, and then over long periods of time, they can say, “You know what, it would be really inconvenient, I don't want to do it. But actually, I'm willing to enable a competitor to you. And actually, I'm gonna go talk to six other leaders in this industry. And basically, we're going to sponsor the next data consortium. It's expensive, we don't want to do it, but you've been so irresponsible that we're going to consider other options”. And that would be one option.

[Auren Hoffman] One of the nice things about once you get to the position where you truly are serving industry from a coop perspective, is that your cost of acquiring a new customer is extremely tiny. If you think of like a typical SaaS company, like they have this pressure to always be raising prices because it's super competitive and their cost of acquiring customers is so high for a company like Verisk and maybe the few others that fit, your costs are so low that you can almost pass on your sales and marketing savings to the customer.

[Scott Stephenson] Part of the value proposition is that, first of all, no one customer would have anywhere close to the data asset that we're making use of, but in addition to that, because we do it for the whole industry, their costs to consume it from us, rather than try to do it themselves. And they couldn't do it the same way, because they wouldn't have the same data, but they could try to derive some of the same things, it would just cost them so much more. We're aware of how much we're saving them. And that's a very important part of the value proposition for sure.

[Auren Hoffman] These kind of coops, they start to be successful, usually in some sort of micro industry or like a limited geography like a country or something. And then once these like flywheel start to happen, I could see a company just getting trapped in that niche. And even though it's super profitable, how do you then expand or move or add services? Do you expand within the industry you're serving and add other flywheels? Or do you look for similar businesses and other industries that you could use the same technology for? How do you think about that strategically?

[Scott Stephenson] So there are three paths, and you mentioned two of them. So just to recap those first. So, a new solution to the same customer. Watching their emerging needs, and then asking the question “are there data assets, which are available to try to serve speaking to that emerging need? And if not, can we try to maybe develop them?” So that's one thing. The second thing, and I was on this point earlier, I've gone looking for other businesses, which are fundamentally this consortium model contributory data set, because we know the power of it. And when you see it, and it's established, you just know it. There aren't that many of them. That's my report, I'm not saying there are none, but there aren't that many of them. And of course, there is power in bringing your next solution to your existing customer, because you've already got distribution, etc. 

But let me give you the third one, which actually, among the three that we're talking about, is either one or two. It’s in many ways, maybe the most important idea, and that is: in the world of data analytics, I think of it as a spectrum, and the two ends of the spectrum are defined as on the one hand content richness. And on the other hand, software intensity. It's the SaaS model from the first day. I'm thinking about digital workflow, which may include analytics, that's what I'm doing, that's what the team is creating, basically, from day one. So that's the software intense end of the spectrum, aimed at making data kinetic, and then deriving value from the data. And it could be that the SaaS solution is more about the flow of data, which includes cataloguing and categorizing, and things like that, but it can also explicitly include analytics for dealing with the data flow. But at the other end of the spectrum, you've got content richness. So, I have specific signals that I have access to, and I'm mining those signals for meaning, basically. So the reason that I lay all of that out for you is to make two observations. The first is, whichever end of the spectrum you start life at, your second act is always going to be to move in the other direction. If you start out with content rich, you are almost inevitably going to want to become more software intense. Why? Because it's another source of value for your customers. 

[Auren Hoffman] Why is that? Because you can see a scenario where a company is really good at the content, let's say maybe not as good as the software or vice versa. And you know, in some ways, it does take a different DNA to do Experian than FICO.

[Scott Stephenson] But a content rich body does not have to reject software intensity. And you're right, they're different capabilities. But the attraction, if you can make it work, is very strong, because part of the sort of recurrence of the analytic that you're providing, so data boosted through analysis. Now, I can say things that I couldn't say before that hopefully have a precision to them, and a predictive quality to them, that you can make decisions around, all of that becomes just that much more valuable if I am also providing to my customer this fixed software, which is basically the conversion of those signals into their own world, basically. And that in and of itself represents value. And because you understand the content and the signals, you are in a position to do that, you're right. It's a different discipline to create digital workflow than it is to go crazy with the latest analytic method and a lot of data engineering. You're right: It's a different technical discipline, but it's not that they are incompatible. You're going to the same customer, right? And you're solving their same issue, you're doing it with content richness.

[Auren Hoffman] How do you know when you're at that point? is it just like, okay, we're, you're no longer growing super fast. 

[Scott Stephenson] And now what are we going to do? Right. And because you're so hungry to add value any way that you can to your customer. And that's really what gets you leaning into that. Let me add one other point, because we're here talking about data. Right. So if we had colleagues on the call with us, it's kind of all data analytic for them. Here's a very basic observation from the way the world is actually working. And that is that if you take my continuum one more time, if you start out with a very strong, dense, rich, content, deep business model, and again, I'm going to argue that in the B2B world, that all really works when you're vertically oriented. To go from there to adding the software intensity, and just amplifying your solution is a much more reliable way of getting an even bigger business than starting at the software intensity, and trying to cross over to content richness. I can explain why that is. But basically, in the verticals we serve, there's the “fill in the blank” tech company. So there's the insurer tech, and the FinTech and the energy tech. And a lot of what's being referred to there, it's a pure SaaS play, basically. We have to really watch out for confirmation bias when we look at the world of SaaS players. I know we're here talking about DaaS, but just let's talk about SaaS for a moment. You can find successful examples of companies that started life as SaaS, and they got a long way because they were the leaders, and they were facing a large addressable market, you know, all good. Right? All good. So you could say, “Well, isn't that so admirable, the growth rates are so high”, as opposed to if you're trying to start a contributory database from zero, that takes time, that's hard work. You don't start to hit the revenue cash register for a while, because first you have to assemble enough content to where it's even meaningful. But here's the point: on the SaaS end of the spectrum, for every company that you can point out that is a success,  there are 100 that died, they never made it across. So they wrote some interesting software, they got a couple of users, it's like, how much can I spend to distribute, but I don't really have distribution, the customer bases and spinning up fast enough? And those customers that I do have, will they trust me with the data? Will they actually permit me to use the data that's flowing through the platform that they gave me? That's very uncertain, you cannot really count on that. On the other hand, there's a time dependency to building the contributory consortium data set. But once you've achieved scale, the value is probably going to be there as long as you do your work well. And now the crossover to the greater software intensity, it's work.

[Auren Hoffman] If you're giving advice to a small company? Verisk is now 50 years old. Some company that just started a couple years ago, they may not want to wait 50 years.

[Scott Stephenson] Well, start, get your expectations in line. That's the very first thing. And then you really have to think about what you can sustain in the build phase. So you've got to scale your cost factors to what you can sustain and be realistic about that. You have to have the courage of your convictions, because it will take time. 

[Auren Hoffman] It's getting rich slowly rather than getting rich quickly, right?

[Scott Stephenson] I actually use that same phrase, but I say get rich steadily.

[Auren Hoffman] Steadily. Okay, like that. Now, Verisk is probably most famous for the insurance contributory data model. But like nine years ago, you bought this other company Argus. What's the big difference in that someone from the outside wouldn't see?

[Scott Stephenson] No regulatory mandate.

[Auren Hoffman] I got it. So it's a harder one to get going, because there isn't a regulator forcing everyone to work together. And so you really have to show that core benefit.

[Scott Stephenson] Right. And, at the end of the day, we hope we do a great job. But let's say we weren't doing a great job, the fact of the regulatory mandate is one more reason to keep any consortium member leaning in, even if they're thinking “Well, I'm not sure about this, I don't know” how but the absence of the regulatory mandate, all of the goodness you're providing could slip into the category of “nice to have”, as opposed to “must have”. 

[Auren Hoffman] Got it. So on the artist's business, there's almost a higher bar, where you have to make sure that the customer is really getting a high ROI, right? You're always trying to measure that ROI for that customer. Because there's nothing to fall back on if your ROI starts to dip below a certain level, they're just gonna leave.

[Scott Stephenson] Yeah, we try to be very clear about the ROI with everything we do. We're just manic about trying to understand what's good for our customers. It's just that all else equal, a customer would have somewhat more optionality on whether or not to work with you, when there isn't a regulatory mandate. And that's really the difference. The data assets over in the financial services, the banking vertical, are just remarkable. I mean, they're just amazing. They're granular, the data asset doesn't exist anywhere else.

[Auren Hoffman] Verisk is kind of like, in some ways, the original insuretech company, well before insuretech was cool. For a 28 billion like market cap company, it's not well known. I checked, Verisk has 3900 followers on Twitter. Obviously, everyone in insurance knows about it. But have you intentionally decided to keep the company out of the radar in a broader context?

[Scott Stephenson] Well, I guess the way I would put it is for us, it's so much about our customers that we're not really trying to, you know, say “Hey, look at me. Look at me.” I mean, there are some places where we actually do need to raise our profile. We have a great team. Our hiring managers tell me that when they're out trying to bring the next super talented person, like a cloud architect, the first 15 minutes is just explaining the company. And then after people get it, it's like, “oh, well, that's interesting, what's the job?” Like, “okay, let's talk about it”. So I get it from there. But definitely, the companies that are in the verticals that we're serving, they all know us, and you know where we are, and we work with all of them.

[Auren Hoffman] What advice would you give to say SafeGraph where I work? What advice would you give us if we're going to break into the insurance industry? Do we think about the insurance industry differently? Is there something like different quirks that we would think about, then breaking into some other industry that's out there, like retail or financial services broadly, etcetera?

[Scott Stephenson] I would add a couple of things, and they all go kind of the same direction. You have to think about what's your attachment point to the insurance company that you would like to have? 

[Auren Hoffman] What do you mean by attachment point? 

[Scott Stephenson] Yeah, so you can attach in several different ways. One way that you can attach is having really very capable and empowered human beings that are the front end of your business model, and send them into the companies that you want to work with. So the attachment point is a human being to human being, and maybe a fair amount of the value add is actually your human beings’ translation of what you got into the value proposition for the other company. And by the way, the things I'm going to lay out here, they're not mutually exclusive, so you kind of want to think about all of them. A second one is back to what I mentioned before, do you want to build software, which is the attachment point, at which you integrate into the customer's own digital platform. So now, when you come with all the goodness, which is your data and the signals that it provides, so be very easy to attach to: serve it up through API's that are sensitive to the workflows on the other side, and let that be your attachment point, or actually create the software gearbox that the customer also licenses, or maybe you give it away to them, but now you are completely in line. That platform that you just gave them is the environment in which they do the work related to the data and the signals that you're giving them. So there's lead time, you have to build that, you have to know what they want, you have to know how that attaches to the workflows inside of your customers, and you have to think about razors and razor blades. Maybe that attachment point, maybe I kind of give it away, and then “Okay, they start to use it”. And then the third one would be the partner. So go find somebody that's already well attached to that market. It could be that what you've got, is integrated in with what we've got, before we even present it to the customer. Yep. So like, make that tie.

[Auren Hoffman] Now, I know that we're both huge fans of the book “Zero to One” by Peter Thiel. And in some ways, it could have secretly been written about Verisk. The core central theme is competition is for losers, right? What lessons have you gotten from the book and how have you applied that to running Verisk?

[Scott Stephenson] As you say, I think it's a great book. And I'm really glad that Peter teases out the distinction between zero to one and one to n. Yeah, because they both actually represent innovation, but in different ways. And we actually need both. If you were to ask me, in the sweep of human history, how much of a difference has been made in the zero-to-one innovations and how much of a difference has been made and the one-to-n innovations? We want both of them off. If you go back to the year 0 AD, so 2021 years ago, the people who study these things tell us that there were maybe 150 million human beings on the planet.  So now in the span of only 2000 years, 150 million has become more than 7.5 billion. And lifespans have gone up: you can reasonably expect to live into your 80s if you live in a developed economy. This is spectacular success for our species, unreal, unbelievable. And if you were to ask me and I this is totally just informal, this is just my sense of it, I think one-to-n had a lot to do with it. Kind of going from being able to get three bushels of wheat out of that acre to four, and then four became five, five became six. But at some point, you have enough people that have increased the productivity of growing crops, enough that actually the whole community now has surplus. 

[Auren Hoffman] The way I think about it in company is that you've got these two things, one is going to be series of 1% improvements, which are really important, like how do you optimize your sales team, etcetera, obviously, 70 to 1%, improvements doubles, right? So that's a huge thing. And then you've got these step functions that also happen in companies and you do need both. You can't just have the step function, you can't just have the series of 1% improvements. So they're really important.

[Scott Stephenson] And the point is, you organize around them differently. You literally think differently, and you organize work differently.

[Auren Hoffman] A couple personal questions before we wrap up. You're really one of the few Fortune 500 CEOs that talk openly about their faith. You seem to be much more comfortable talking about it now than you were earlier in your career. I heard you say that you once passed up an opportunity to say Jesus was your hero at a BCG event and felt terrible grief about it, like how has that, in your own heart, evolved over time?

[Scott Stephenson] I did have that defining experience, and when you have something like that, you just don't forget it. I feel like I betrayed the thing that was one of my prime commitments, but I had a chance to demonstrate it publicly, and I didn't. And so the way that I felt after that is just a total cautionary tale to me that just simply states my commitments, and just be prepared for whatever effect that has in the conversation. The spirit of the cultural age that we're in right now is authenticity, and state who you are. The culture kind of invites it, the number of people just start talking about what they're passionate about. There's just a lot more of that than there was when I was a younger person. It sort of felt like you were just kind of thinking a little bit harder about how you presented yourself and, with time, I've just become extremely comfortable in my own skin. But the other thing that I would say is that on top of all that, I had a negative motivation, I feel invited by the culture around me in many ways to just state that I'm a Jesus follower, and I'm passionate about it. But I'll tell you, it’s interesting how frequently when I will state my commitments in that way, that creates room for the other person to do the same thing.

[Auren Hoffman] Especially since you’re the CEO and now as a manager level person, I have a belief that I might be worried about not conforming, and I might be more willing to be open about my feelings. 

[Scott Stephenson] And I take pains and I have to be thoughtful to not use it on people that are in the company that I'm a part of. I'm expressing myself to people I just meet, it's just in any walk of life, but I do have to be careful within the company to make it abundantly clear that your job here is not contingent upon agreeing or seeing it the same way that I do. Express your own commitments, that's what the inclusivity agenda that we talk about so much at companies should really be about: bring your whole self to work and you will be embraced by this community and be who you are, that's inclusivity. So one way of inviting inclusivity is to include myself.

[Auren Hoffman] You have to be a little more vulnerable to do that. Right. Okay, that makes sense. That is a big difference in CEOs today than it was totally 2030 years ago.

[Scott Stephenson] Completely. You basically better engage with your people a lot, you do not sit in the corner office and write memos, interact with your people a lot. And not only that, but when you get in these 10-15 person meetings and you're going into deep things that relate to the business, but also who you are, I I'd say the other difference with a few decades ago is, you better have some game, you better actually be a real human being, you better value the other human beings that you're in the room with. And these are good developments as opposed to, “hey, I'm the boss”. 

[Auren Hoffman] Also, people have a choice. They could go work for another company tomorrow, right? Everyone who works at Verisk is a super talented person.

[Scott Stephenson] For those of you who are a couple decades younger than I am, honestly, when my Qadri came through, and we were looking for our first jobs, the best job was the one that paid the most. I'm serious. Now, we all know, we're supposed to ask, well, what's the purpose of this company? 25 year olds today are much better and more sophisticated consumers of employment opportunities than my generation was, they really, really are.

[Auren Hoffman] One thing that I think is also a little bit different about you is that you're CEO of a huge company, yet you're insanely good at responding to email. How do you how do you manage all these? I mean, you must have crazy amounts of things coming out your inbox.

[Scott Stephenson] Well, I would say a couple things. One, it is helpful to anyone, if you know fundamentally what you're about. Thank you for noticing, by the way, because I take it very seriously. But, if you know what you're about, then there's something presented in this email, you can think about what you were just presented, but also kind of go inside and put it up against the things that you know, and the things that you've thought about, and you can respond quicker. So knowing who you are, promotes the efficiency of response, I would say. But then the second thing I would say is that, I've had a very stable and relatively uninterrupted career, but I did make a big change. A big pivot in my career was essentially a two year period, between when I left BCG as a senior partner, and when I entered, what is now Verisk, there was a two year period in there and I had a sense of the destination. I hadn't identified Verisk or what it was called back then. But I had a general sense. And, and my point here is, I remember in those two years, like you cast yourself off from something and, I was totally platformed to BCG, I could call my partner in Kuala Lumpur and ask her what's the latest because I think that's meaningful for my client over here, so I felt I was very close. And then all of a sudden, I felt like this just teeny little kayak bobbing in the middle of the Pacific Ocean. And I can remember so distinctly the feeling back then, you're trying to generate something, it's original, you're trying to get it going, basically, and so you're reaching out  in a lot of directions, but I wasn’t getting the inbound. 

[Auren Hoffman] Now, you know, it's like that for every founder entrepreneur.

[Scott Stephenson] Does anybody know or care? I'm out here. I just realized in that experience, one of the ways I can serve other people is to simply respond because it says the universe is not indifferent to you. The universe, at least this one little speck of the universe, you reached out, heard you and came back to you and I know how it made me feel when people did that when I was more that kayak bobbing in the middle of the ocean. I know the encouragement it was to me. I just want to do that for other people, because it's amazing. 

[Auren Hoffman] When I was in college in the 90s, I would occasionally email Steve Jobs. And he would always email me back. He knew I was a college student from my berkeley.edu email address, he would always email me back within like eight hours or something. Maybe just like one word, or something. But it was amazing. Like he was on top of it, I always remember that.

[Scott Stephenson] The universe answered you, you spoke to the universe and the universe answered. It was existentially affirmative. Well, we can do that for each other.

[Auren Hoffman] Our last question, we ask all of our guests, what would you have told yourself if you went back to yourself in high school or college? What would you tell yourself to save yourself? Either like time or money or emotional well being? What would you've done differently?

[Scott Stephenson] It's not so much sort of the efficiency of the journey, but it's something that I would say more I had to learn through experience, which if I had been told, and I could have believed that probably would have saved me, some cycles and anxiety. And so what do I mean by that? You know, today, the phrase “imposter syndrome” gets thrown around a lot. Yeah, you know, we hear that a lot. Right? I think it's fair to say that kind of the outcomes in my career are ones that a lot of people would be happy to have. I feel very fortunate and very privileged to get to do what I do. But my point here is that when I was 24, and when I was 29, and when I was 33, and when I was 38, I had experiences of what people are talking about when they refer to imposter syndrome. Another way of saying that is that the people who just gave me this new, bigger responsibility, do they realize that I'm not ready for this job? My experience base has not prepared me for this job. Do they realize where I am in my journey relative? It's kind of anxiety provoking, but you have that happen to you enough times. And you come to understand that's just the process. 

[Auren Hoffman] You can’t learn if you're not working on something where you don't have a high failure rate. If you're put into a new thing, where you have an extremely high likelihood of success, your growth is going to be capped.

[Scott Stephenson] And it is the source of growth, not only for people individually, but actually for the organizations or the teams with whom they work. It actually is the way that everything advances. It's the one-to-n form of growth. There's also a zero-to-one where you kind of rethink everything, but the one-to-n is multiple times getting put into situations that are just bigger than you are. But you go through it enough times and you get to the point where “Oh, it's not me”, this is the process. This is the way we want it to happen. But the issue is, you don't get told that, it's not necessarily highlighted. And that's why I want to highlight it for maybe some folks that are on our call today that are a little bit earlier in the journey. So when you have that experience, hopefully it won't have to be for you the way it was for me that I had to go through it three or four times before I got “Oh, it's not me, this is the process”. So if you can get it at the first one, then you can just save yourself the mental energy of “Am I just gonna keep ending up in places where I started out not feeling capable?” Well, maybe. But actually, you shouldn't resist that. You know, live into that. So if somebody had told me that when I was 24 I don't know if I would have believed it or not, but I think it's really true.

[Auren Hoffman] I love it, this is great. This is a perfect place to wrap. Thank you Scott very much for joining us World with DaaS. This has been really great. 

[Scott Stephenson] My pleasure. Always good to be with you Auren. I look forward to it, and may it be not too long until we can again break bread together. Last time we did such was a fun evening and I look forward

[Auren Hoffman] March 2020. It was one of the last few great dinner parties.

[Scott Stephenson] The curtain was descending even as we were finishing dessert. 

[Auren Hoffman] Exactly. Thank you. We'll see it. 

[Music playing]

[Auren Hoffman] Thanks for listening. If you enjoyed this show, consider rating this podcast and leaving a review. For more World of DaaS (DaaS is D-A-A-S), you can subscribe on Spotify or Apple Podcasts. Also check out YouTube for the videos. You can find me on Twitter at @auren (A-U-R-E-N). I’d love to hear from you.

Listen to more great episodes

July 14, 2021

Henry Schuck: Building A Billion Dollar Data Company

Henry Schuck, CEO of ZoomInfo (NASDAQ: ZI), talks with World of DaaS host Auren Hoffman. ZoomInfo is a $17 billion market cap B2B data company and one of the only billion dollar data companies that have been created in the last 20 years. Auren and Henry cover ZoomInfo’s origination, how it differentiated itself in an already crowded market, best practices on building a contributory network, ZoomInfo's unique acquisition strategy, and more.

July 1, 2021

Will Lansing: Power of Data Standards

Will Lansing, CEO of FICO (NYSE:FICO), talks with World of DaaS host Auren Hoffman. Will previously served as the CEO of InfoSpace, ValueVision Media, NBC Internet, and Fingerhut and a partner at General Atlantic Partners, a global private equity firm. He also co-authored the DaaS Bible 2.0. Auren and Will dive into data standards, the relationship between analytics and data companies, and how FICO successfully transformed their score into a currency.