Scouting for Growth
00:06:28 · Amrit Santhirasenan
[00:06:28] S1: Yeah, I love that question. That's a fantastic question I've never been asked before. I have an easy answer for this actually. I love insurance. I'm a huge insurance nerd. I think insurance is a societal pillar. Know, there are no rockets launched. There are no large buildings built. There are no planes that take off without insurance. And, you know, without the mechanisms by which we can absorb risk, we can't take it. And when we take risks, we reap the rewards of those risks and society moves on. So I'm a huge insurance nerd. I think it's important I tell everyone who joins Hx you need to learn to love insurance. Know, come for the fun opportunities, stay for the insurance, right, is what I would say. I think that insurance is one of these fascinating sectors that intersects technology, mathematics and human judgment in a way for people like me in a way that's very unusual. You have people who are tasked with working out how risky something is and you know if by definition if it's perfectly predictable, it's not risky. And so you have to analyse that risk. You do that in a variety of different ways. You use tools to do that, use techniques to do that. And I think it's a wonderful place for people who love solving hard problems in a variety of different ways. The actuaries and underwriters have to huddle together around the underwriting decision that's made is the essence of the problem that we're trying to solve. But I think this intersection of making human based judgments, combining it with interesting kind of mathematical analytical problems and using technology is like actually insurance is one of the best places to combine those things. And it's a heady cocktail for me. Again, I'm a huge nerd in this space, but I think it's an awesome place to spend your time.
00:08:09 · Sabine VanderLinden
[00:08:09] S0: You know, I really enjoyed my chat with Richard again recently. And we actually talked about for sure, Hyperexponential, but we decided to call the podcast, The Pricing Nerds, because he said, Sabine, we are nerds about insurance. We are nerds about pricing. I said, you know what? We found the title for this discussion. And so that, you know, it's doing very well over 3,000 downloads. But let's go to the decision intelligence. So decision intelligence is not a new concept you won't believe. You can see, you know, lines on my face, but I've been in the industry for quite a while. And I have a guest, I work with FICO and my job there was all about decision intelligence. And so I would love for you to decide... To actually define what decision intelligence is for Hyperexponential. And what does it mean when we actually talk about advanced pricing mechanism?
00:09:11 · Amrit Santhirasenan
[00:09:11] S1: Yeah. Yeah. It's a great question. You're absolutely right. Decision intelligence, the essence of what it is has been around for a really long time. I think it's really important to distinguish between a kind of decision intelligence as a mindset and a way of solving a problem and then the technology side. And I think probably the area where Hx has been very novel is in crafting a technology solution around actually the act of obtaining, if you will, or applying decision intelligence to the problems that you're solving. So that's probably where the novel aspect of Hx comes. But for us decision intelligence, you know, it's a straightforward concept which is where you go about the act of solving a pricing problem in such a way that you bring together the data, the insights and decisions together so that you're better equipped to make those decisions over time. Right? So every single time you do some work in the pricing as a pricing exercise or writing exercise where underwriters and actuaries work together to design and build the models and then to use them and to capture that data. Actually, just you set up a system such that the people who are doing that work automatically create a rich data asset that's coherent, that's evergreen, that's growing, and that the data asset can be fed back to make better decisions next time. And one of the things that I saw over time and, you know, people spend a huge amount of time and energy on technology. That's not new. That's happened for a really long time, but the technology didn't create this kind of virtuous feedback loop, right, of the data, the insights, the decisions over time growing. We would capture loads and loads of data. I remember this in the mid two thousands. I would watch as a software engineer us capture all this data and then throw it away, right, and never use it again. And it was hard, right? You know, one of the things about our space that people often underestimate, you know, I'm very fortunate to be the founder of a moderately to very successful enterprise technology estate company. But what that means is our clients have enterprise technology data landscapes. If you've a $100,000,000,000 balance sheet like some of our clients do, you've got data sets that are far and wide over multiple countries, territories.
00:11:17 · Amrit Santhirasenan
Some are still stored in green screen, like, underneath green screen systems. Some are in modern databases like Snowflake. Some are stored in PDFs and email systems. Getting this data together, operationalizing it so it can be used to make a decision, and then setting that process up so that actually that data gets get better over time is really hard. And that's where Hx came together... Came really onto the scene because we crafted the technology to actually take away all the boring act of kind of doing database manipulation or building algorithm containers or building web forms to let our clients focus on the problems that they actually cared about, you know, building the algorithms, using the data to make better decisions, surfacing that to the underwriters at the point of transaction. And that's where I think we really started to add value in a way that had never been seen before.
00:12:07 · Sabine VanderLinden
[00:12:07] S0: And what I hear here is we are going for dynamics. So we are moving through a business model, which is far more dynamic than static. When we actually start looking at where the market is going because of the complexity of the risk, we are starting underwriting. We need that dynamism. You won't believe that I was just last week presenting at the CCN Show Tech Hub in Zurich. And the enterprise application market is the largest one, the largest one which is still receiving investment, 1,600,000,000,000 investment to date. And when you start looking at, you know, some industry not doing that well, when you look at investment, your space is doing very well actually, Amrit. And I think it's important to highlight that. But then the second question is, you know, I get asked often by insurers is how do we mix external data sources when we start looking at this new world of decision intelligence with our historical data, we love our historical data, but we need to deal with those new emerging, right? So how do you address that problem around data to make better decisions?
00:13:21 · Amrit Santhirasenan
[00:13:21] S1: Yeah, another fantastic question. You know, something that I always found my job when I was an really fascinating, right? Because I often when I try to explain it, if you're an actuary, you have this really awkward situation at parties. If you're an actuary, don't go to that many parties, the ones I did go to, right? You have this awkward situation where you have to explain what you do. And one of the things that I think a lot about the way I did my job as a pricing actuary is really my task was to understand the world of the risk being priced better. Right? That was my job. Like, our job is to apply a framework to understand the risk and then set up the risk bearing structures on top of that that make the best sense for the customer, are the fairest, are the right price, are competitive in a marketplace. And so really, the job of a pricing actuary or the job of a pricing team, underwriter and actuary huddling together or data scientists in this day and age, you know, is really to understand that risk the best way possible. And I think when you think about pricing as that problem, the historical data is really important. Right? But you said something, Sabine, about the dynamism of the world. The world's never changed faster. Right? That's the thing about exponential growth. Right? It's never changed faster. The historical data is really important, but it's not the be all and end all. And if you look at risks like cyber, for example, the past isn't meaningfully a guide to the future for these emerging risks in the same way actually that it has been for maybe a more established, you know, if you think about like a fire risk or something, a classical fire risk for a more traditional line of business. One of our investors, we're very fortunate, the founder of Guidewire. Most people know who Guidewire is. Marcus Ryu is now a venture capital investor. He's at Battery. He used this really lovely phrase when he invested in us. It's like, actually pulling the signals out of the noise in a world that is never, you know, has meaningfully different from the past. The teacher is meaningfully different than the past is a hard challenge, and that's what we do. So when I look at these risks, actually, it doesn't phase me that we don't have a lot of historical data. In fact, the opposite is true for these very large risks.