Key takeaways:
- Use data led, digital & integrated solutions
- How to properly analyze and report data for solutions and answers to business challenges
- Ways to capture, not only create value
- Define and Foster a Data Culture
Peilin Corbanese (00:10):
We're going to have a high energy session to show you what we can really accomplish with a fantastic panel here. So I am Peilin Corbanese. I'm the moderator for using data to drive high value growth. It could be a misnomer. We always talk about high value growth. We always want to grow. Everybody wants to grow. You have the KPIs, you have revenue, you have sales, you have premium, you have profitability, you have happiness. As I said yesterday is one of my key KPIs. So how do you try high value growth? Because this is something everybody talked about, but nobody really truly knows how to do it because it's different from company to company, from individual to individual. And the important part is we are going to learn from these esteemed panelists, care and steward about what they do. And specifically, I promised that we're to give you use cases so you can learn from it and there'll be numbers and then you can ask questions. So here we are. We're going to do the self-introduction. Kara, please tell us a little bit about who you are.
Kara Hoogensen (01:28):
All right, well thank you for joining us. I guess technically this, oh, I can say now this afternoon. So happy to be here with you. Kara Hoogensen. I am the Senior Vice President of Specialty Benefits at Principal Financial Group. And those words, specialty benefits don't often resonate with most people. So to be more descriptive, I actually lead the employee benefits are the group insurance business and the individual disability insurance business. For principal, we serve about 4 million individuals and 175,000 small and mid-size businesses in that business. Do you want me to tell my story now or do you want to wait?
Peilin Corbanese (02:11):
We promise a story, an intimate story where you can actually connect with Kara. So I was on a panel with Kara last year and I love her, but I don't remember what her story's about except I remember who Kara is and her toilet paper story. So she promised a new one this year.
Kara Hoogensen (02:30):
Yes, I did. I couldn't subject Peilin to this, the same story. I call that one my tears and toilet paper story. So if you're interested and that's intriguing, happy to share afterwards. But we were asked to think of kind of a defining moment in our careers, maybe a leadership experience. So I'll try and make this close, and I don't share this necessarily to say this is the right thing to do, but I share this as maybe an opportunity to remember as sometimes what we do in our roles impacts not only the employee but their family. And so I'm going to take you back a few years to a point in my career where I was leading the retail broker dealer for a principal financial group. And there's certain parts of leading a broker dealer where there are individuals that as they make decisions in their roles, there actually can be personal liability with those decisions. And we as an organization found ourselves where we had identified an issue and we needed to report it to our regulator. And in this situation, while based on the data and the facts that we had at our disposal at the time the decision was made, it was absolutely the right decision. In hindsight, we probably could have done something differently. And so I ended up in a situation where I had about three employees that were really, really nervous and nervous to the point of not only about the potential impact on the business and the registered representatives that this could impact, but also from a personal perspective because absolute worst case scenario, there could be again, this personal liability. So as a leader, what I ended up doing is I felt so compelled they were acting in, they were doing good work based on what they knew at the time. And so I had all the appropriate conversations that I needed to have with key stakeholders at principal to say, if the worst case scenario ensued, we would stand behind these employees and make sure that they landed softly and in a good spot. Well, thankfully, nothing close to the worst case scenario happened. But by having those conversations, by sitting down with each of those employees and sharing with them that I as a leader and principal as an organization was going to stand behind them, irrespective of the outcome from this particular set of circumstances. I got in the mail, believe it or not, snail mail. And again, this was a few years ago, but I got in the mail a few weeks letter, a handwritten note from the husband of one of those people. And he was just expressing his appreciation and thankfulness for the organization and for the conversation that I had had with his wife. So again, I just share that in the spirit of sometimes we don't recognize the full impact on not only the individual, but also their family members from the decisions and the actions that we're taking as leaders.
Peilin Corbanese (05:30):
So think about high value growth from individuals that you have impacted like that. Thank you. Thank you for sharing your story. Stuart, I understand there's a todo of France story in the making.
Stuart Mathews (05:43):
Yeah, so we're going to go in a different direction with me. So mine was about personal suffering and how you can push through barriers. So I decided after my basketball career and volleyball was starting to wane that I needed to somehow learn how to be an endurance athlete. So I took a cycling competed, raced, if you could see me. I'm not the typical cyclist. I'm six three. The typical cyclist is about a foot shorter than me. But I did stages in the Tour de France following the professionals along as an amateur, climbing up Alta and other places to suffer and realize that they really truly are superhuman. And the joy was in the finishing, not in the time.
Peilin Corbanese (06:25):
And tell us what you do for a living.
Stuart Mathews (06:30):
Yes, So I'm at UC, Irvine, which is one of a hundred R one research universities in this country. The University of California itself is 10, and we get massive dollars coming through the grant process, massive research taking place on campus. And the reason I went there is because I'm a kid in the candy store, I'm working with industry to take invention and intellectual property that's gets created, could be on a topic from clean hydrogen power as an alternative to solar nano technologies for EV batteries, health medicine, urology, et cetera. No one person can be equipped for that. But my excitement is around what we're doing with generative AI. As most of you probably know, AI is all about the math. And the researchers at our campus are doing some incredible things with deep learning and machine learning math. So I'm here to collaborate with people that want to work with these academics.
Peilin Corbanese (07:30):
So you did use the word generative AI.
Stuart Mathews (07:34):
I used it. Do. Is there a cookie jar? I got to put money in.
Peilin Corbanese (07:39):
So can we then say that data equals generative AI?
Stuart Mathews (07:46):
Oh, I don't know about that. I'm not going to step in that. I'll, we'll let her answer that one.
Kara Hoogensen (07:50):
Nope, I'm not touching that one.
Peilin Corbanese (07:53):
In that case, Kara, tell us a little bit about a use case, how you define high value growth and how you drive outcome. And give us some real numbers please.
Kara Hoogensen (08:02):
Alright, so to share one of the ways that we are using the combination of our talent and technology, because at our organization we believe it's the combination of the two talent and technology that ultimately is going to help drive growth. So the part of the business that I want to speak to first is our disability, our group insurance disability. And these are people, this is the claims world, individuals who have had an illness or some sort of medical condition that is keeping them unable to work for a prolonged period of time. And we have started deploying a technology from a third party firm that is helping us manage that part of our business much more effectively. So these are individuals, our claimants are people that have been on long-term disability claim for at least two years or more. And sometimes they're on claim for five years, sometimes they're on claim for decades. It really depends on the situation. But at the end of the day, we're keeping people out of poverty because they're now able to make mortgage payments to pay for their children's college tuition, whatever the case may be, because that income is coming in during that period of time when they're unable to work. What happens after we reach that two year mark for a long-term disability claim is historically we have a set of professionals that on kind of a time-based approach every year or two years that claims analyst is reaching back out to that claimant to inquire how they are doing. Has anything changed in their health condition such that maybe they are able to do some type of work or maybe they're actually able to go back to the role that they had previously or that same type of work. So very methodical, time-based approach. It's also very, very inefficient because the data would tell you that certain conditions most people are never going to recover from. So our claims analysts reaching out to those individuals every year, 18 months, 24 months, is actually wasting everybody's time and is a complete nuisance to that claimant who is probably not, they're struggling, they're not able to work, they're dealing with a significant health condition. So this technology that we have deployed has put us in a position to move away from that time bound approach to identifying those claims where the pattern of the claim is an outlier. So there's something about that claim that would say our claims analyst should actually reach out to that claimant and have a conversation because that claim is behaving, if you will, differently than similarly situated claims with the same condition. And we are still at the early stages. We've deployed this technology against our long-term disability claims and we feel confident that we're headed on a path that is similar to another firm that has already deployed this same technology. And that particular firm is seeing a seven to ten seven to ten times r o i in their first year. And they're actually have seen a greater, greater than 25% reduction in what we call in intrusive claim activities. So reaching out to those claimants where it's just a complete waste of time because nothing has actually changed. So you ultimately think about the impact on that claimant experience by having interactions with people where it really makes sense to do so, the employee experience because we're not wasting their time. And ultimately then the impact that that can have on claims reserve. It's a win win win. We feel so confident in this that not only have we deployed it against our long-term disability block, but we're actually implementing modules that will support our individual disability block of business as well as our short-term disability block. And we invested in the company that actually is making this technology available. So that would just be one way where I'd say the power of technology to allow our talent to be more effective and ultimately improve the claimant experience is having a real benefit to our business.
Peilin Corbanese (12:20):
So should I say to summarize what you said is looking at the outlier of your data, understand the pattern and identify what that means for your organization. And once you learn from that, you scaled it across different lines of product?
Kara Hoogensen (12:35):
Well, yes. So it came to us in a way that we could scale it across our entire long-term disability maintenance block, which is a few thousand, call it five or 6,000 claims. And then there are different modules because nuances to the different types of income protection or disability insurance. And so those nuances are embedded in the other modules from the same organization.
Peilin Corbanese (12:59):
So you mentioned it's seven to ten times, right? So is that based on the time save for the people or is that based on, what is that based on?
Kara Hoogensen (13:09):
So it really is, the calculation is based on the fees that we're paying to use this technology relative to the time saving of the employees, the claims analysts, and then the reserves that we're able to release from individuals that are no longer on claim. So they've been able to return to work.
Peilin Corbanese (13:29):
And that's what you would consider as high value growth?
Kara Hoogensen (13:32):
I think. So ultimately that has an impact on the bottom line.
Peilin Corbanese (13:36):
Fantastic. Thank you so much for sharing. Stuart. I understand you want to talk about cybersecurity.
Stuart Mathews (13:43):
Yeah, why not? Let's talk about cyber. So as we looked at this topic, I'm going to take an angle from my past role as a VP of insurance for Experian. I'll speak to a product that we developed. Ultimately it was my team and the work that we did in the data lab, a hundred data scientists, physicists, software engineers, and others coming together to build new products. We typically wouldn't work on something unless it fit this moniker, high growth, not going to take this team and have them work on something small or that's going to be in the business and waterfall. So when we looked at our data, one thing that jumped out at me was the amount of dark web data that we have and how good Experian is, was at connecting identity information with random bits of data that's across the universe we all live in. So the initial hypothesis was I wonder if our data would inform cyber risk such that someone would pay for a model. Let's go create a model. We don't have one based on cyber risk. We have identity flags that if you use Experian will tell your identity got exposed, but nothing that would be used by a cyber underwriter to be able to say that's a good company or a bad company. And how do we look at it? So without giving away the secrets of what we did, essentially what we said is how can we do it for free as best as possible, i.e. with our own data without having to license from someone else. And what's the ultimate goal? So the approach we took was the employees are the greatest risk to the business. Anyone that's in cyber in this space knows that it's true. The people are the weakest link. So with the data that we had available, we were able to see first of all, who are all the employees of a business really hard to do, right? Background checks and other things point to that. But then what can we see about the people in their personal lives as well as their business life that tells us something about their cyber risk and their behavior as an employee of this company. So it turns out people are not very good with passwords. People reuse them, they don't understand password managers. And as you look at the exposures of John Smith or whoever across the dark web and the internet, you see very distinct patterns and weaknesses. So we built this model that would pick in the worst scoring decile, the way that anyone that's done modeling those, you look at these deciles of pieces of scored information and the entities that come across the worst scoring decile was 44% of all the claims dollars for the customers that we worked with. And there was many millions and millions of policies, which ultimately points to people are the weakest link and the biggest problem that businesses are trying to protect against, you know, can do all this stuff around the infrastructure and it doesn't have to be mission impossible and Tom Cruise popping something and jumping over the firewall, it's someone sitting at their desk clicking on something that they didn't think about, but the data proved it and it was a very profitable model. So high growth for the company.
Peilin Corbanese (17:01):
So I hear that people are one of our most valuable resource. I know Bruce likes that and I also hear that people are our weakest link. So from your perspective, how would you leverage people in the most effective way to generate high value growth for your company? Are you looking at Stuart or at me? I'm looking at both of you. She's
Stuart Mathews (17:25):
She's looking at both of us.
Peilin Corbanese (17:27):
I'm going to both of you.
Kara Hoogensen (17:27):
No, you go first.
Peilin Corbanese (17:28):
You have very different perspectives, right? Yeah. One is collaboration and you, Kara, you manage this large team and you have a lot of people. So perspective will be very different. Okay, why don't you go ahead pick.
Stuart Mathews (17:40):
Yeah. So mine, staying on that toping.
Peilin Corbanese (17:42):
This wasn't rehearsed. So they're both shocks. I'm asking the question.
Stuart Mathews (17:46):
Yeah, staying on that topic, I would say one of the things that the carriers wanted to know that the model gave back was 16 different attributes, some of which could be used to educate the people on how their cyber life is bad for their own personal life as well as what's happening from a business standpoint. So that's a quick answer to your question.
Kara Hoogensen (18:11):
So I'm actually going to piggyback off what you shared, Stuart. And this is about, so my business focuses almost entirely on serving the needs of small and mid-size businesses. So think of employers that have less than 500 employees. That's who we primarily work with. Well, if you also think about those employers, and I will say our average customer has somewhere between 30 and 35 employees. So we're even focused more so on the smaller end of the small and mid-size business segment. Those employers, most of them don't have a dedicated chief technology officer. They don't have a dedicated CIO, they don't have a dedicated CHRO. And so it is an individual that is wearing multiple hats back to Stewart's point that from a cyber risk perspective, employees being the weakest linked, most of these small businesses don't even know what to do. They aren't prepared for a cyber risk event. And I am getting to the high growth part, but there's organizations out there that can help there's, and so principal is actually partnered with an organization called the Cyber Readiness Institute. We're a member of that organization and their sole mission is to help small businesses address cyber risk. They have modules available free of charge to go out and help educate someone at each small business on strong passwords. The appropriate use of USB drives the importance of making sure you're regularly patching the technology that you're leveraging for the growth of your business. And I'm remember, I'm not remembering the fourth one, you're nodding a little bit. Do you remember?
Stuart Mathews (19:54):
You're on track, those are good.
Kara Hoogensen (19:56):
Anyway, there's four key elements and they're evolving into some cloud-based aspects of security as well. But why I share that as an example of here is really important information that's our target market, our customers need to know, but maybe aren't necessarily aware of or equipped to pursue on their own. And so we have engaged in trying to support our small business customers and prospects by making sure they're aware of these free tools and resources that are at their disposal. Because ultimately if those small businesses are better protected, we as an organization in doing business with them are also going to be better protected. And so that ultimately is just one example of the approach we take to value adds in the way that we go about doing business so that we can ultimately drive the growth that we're looking to drive for our shareholders.
Peilin Corbanese (20:52):
So that's another layer because we started by having talking about use cases for this session. So you have your use case and you have yours, but now we are pivoting onto in the next level is where you drive value across partnership and investing in people to drive that high value growth. So at this moment, I would like to ask the audience to get some feedback. One is how does your company or your organization value growth? And there are a couple possibilities. One is sales, so that's one profitability people. So I see same number of similar number of hands on people versus sales most in profitability. So Kara and Stewart, can you opine on that and where do you think we should go as an industry to really generate this future of insurance that would create a mission and drive the next generation to participate in it?
Kara Hoogensen (22:12):
So maybe to clarify, Peilin, are you thinking about driving talent to the industry?
Peilin Corbanese (22:18):
Driving talent to the industry, knowing that most of the organizations value profitability first and then sales and people?
Kara Hoogensen (22:30):
So my two sense is that we're not going to generate sales nor ultimately profitability without people. So I don't know that to me it can be an either or. It is, it's a Venn diagram that we have to be constantly cognizant of in order to ultimately drive growth. So I'll go back to one of the comments that I made earlier in that our belief as an organization is technology is going to continue to change at an ever increasing pace, but technology is only one part of our success. It really is the success is going to be defined by the combination of that technology and the talent that is using that technology or leveraging it in different capabilities or in different ways to ultimately serve the needs of our customers. And that is ultimately what is going to drive growth. If I think about insurance at the end of the day, whether we're talking about property and casualty, whether we're talking about health insurance, whether we're talking about ultimately retirement savings in some regards you could think of as insurance. It's about building trust with people so that they can have peace of mind. That's a very human activity. So I don't think that at any point, at least during my working lifetime or in lifetime in general, that will technology completely replace the role of our talent. So back to this Venn diagram and making sure that we're creating the right combination of technology and talent to serve our customers' needs and ultimately deliver that growth.
Peilin Corbanese (24:17):
We did not rehearse this, and I love this because the reason we started with this track and with this topic using data to create high value growth was because I recently learned that in order to create high value growth, all you need to do is build trust. It's not all you need to do, but that's almost step one. You need to build trust with people and once you build trust with people, everything else will come. But I'm going to turn it to Stuart, who is our innovation person. Tell us a little bit about how you build trust within your innovation.
Stuart Mathews (24:53):
Well, I mean that's pretty simple, right? Be authentic, right? With the people that you work with and the partners that you work with. As we were thinking about the problem and the setup, that was a great answer I'm hating having to go after her answer. That was great. Instead of saying yeah to what you said, I agree with it. I'll say on the people side of what people raised their hands to you touched on it a little bit yesterday in your session, but I think that there's a lot of growth that can come from investing in our people more. Unfortunately, there's just not enough time in the day to create systems that can learn what our people are good at and not so good at and help them get better. But I would suggest that we need to excite the next generation of people to come into the insurance industry. I think I've heard a lot of people talking about the importance for new blood. We definitely need that. And so I believe technology can help grow and retain and excite the people that are on staff to one better understand the business. And as we know when everyone's marching to the same beat, we're going to be more profitable and revenue will be higher.
Kara Hoogensen (26:05):
So maybe if I can add on to that, I am blessed to be the mother of a 21 year old and a 19 year old and my 21 year old is studying to be a high school math teacher and which just makes me super proud. And as I think about just her as one example of that generation, when you ask her why she's choosing that profession, it is because of the potential impact that she believes she can have on young people's lives. And it's partly to pay it forward. It's because of the impact that many of her math teachers who also happen to be her cross country coaches as a runner have had on her life. And she's just one example of that generation that is ultimately fueled by a passion and doing good. When I think about what we get to do, irrespective of what part of the insurance industry we are in, we trying to make people's lives better. We are trying to take risk out of their lives, back to income replacement or disability insurance. We're trying to make people able to provide for their families or themselves during periods of time when they can't work. We're trying to provide a financial benefit to a family that maybe just lost their primary breadwinner due to a death. There are so many noble reasons that we exist and I think if we as an industry can do a better job of conveying that to the younger generation, that is ultimately going to be key to us getting new talent, new blood into this business.
Peilin Corbanese (27:44):
So I would like to say this, using data to drive high value growth, you can have partners, you can have service providers, you can have bosses, you can have colleagues, you can come up to any one of us, we'll listen to your story, we'll listen to your problem statement and give you multiple use cases that you can possibly take it back to your organization and deliver high value growth, whatever the definition is. But the meaning of the session is really about investing in people, build that trust and that's where you generate your ultimate high value growth. So we have a minute and a half left. Any questions? Well, we can do a secret dance. Yesterday. Yesterday, okay. Hold on. Yesterday when I said that we'll do a secret dance so we can retain our talent and because the industry needs to transform, there were quite a few of you afterwards told me that you'll really come up here and do the dance with us, but nobody else did. So can we do that today, Bruce? So there is always a question.
Audience Member 1 (29:00):
So you mentioned that you were working with a vendor to essentially identify as a backbone of their capability, was that leveraging AI to do that identification?
Kara Hoogensen (29:15):
Yes. I can repeat the question and give me the head nod or the shake if I get it wrong. So the question is, with respect to the vendor that we're working with on our long-term disability claims, are they ultimately using artificial intelligence as kind of the backbone of their capability? And the answer to that is yes. I'm happy to share the name. It's Evolution IQ is the name of the organization. They have focused their business far. They're about a four year old company. They focus their business so far on the disability space and are looking to expand into the workers' compensation space as well. There's a natural point of connectivity there. So yes, they actually are using artificial intelligence. They have one of the individuals as their CTO who was formerly at Google and involved in developing some of the machine learning and artificial intelligence models that Google has in place.
Audience Member 1 (30:08):
Okay, very cool. Thank you. And then Stuart, the second question was for you. So you get the opportunity to work with different industries and including insurance. Is there anything that you see others doing with their data that you wish insurers were doing more?
Stuart Mathews (30:25):
Oh, Great question. I think the biggest thing is relates to entity resolution. So I think everything that people try to do ultimately is pointing back to understanding that entity that you're talking about so that you can accurately model that, whether it's a person or a business. Doing that better. Whether you want to call that master data management, et cetera, is important. There's been so many acquisitions of different systems still trying to get them to talk, but underlying that bed needs to be this who, how are we going to refer to all the entities so that all the assumptions we make are accurate and point back to that same one.
Kara Hoogensen (31:02):
No easy feat.
Stuart Mathews (31:04):
Super hard. Yeah. Thanks guys.
Kara Hoogensen (31:09):
Thank you. Thank you. Enjoy the rest of the conference. Secret dance time, right over there.