Insurance Talks #4 – The Holistic View of Data
For the Insurance FM podcast, Keylane interviewed Christof Mascher, COO and member of the Board of Management of Allianz SE (until December 2020), and Benedikt Kalteier, a member of the Generali Board of Management in Germany and CEO of Dialog Versicherungen and CosmosDirekt, on the subject of the holistic view of data as the key to understanding customers, risks and partners in the insurance industry.
The full interview and podcast can be found here (only available in German).
The changing role of IT in insurance
Michael Carl: Welcome to the show. Over the past few months in these podcasts, we’ve discussed a bunch of big topics in insurance and the insurance industry. We’ve talked about agility and security, debated the customer interface and how it’s being digitised, and we’ve talked about internal processes. All of these topics are related to IT, to digitalisation, and are therefore under pressure to change, and at the same time facing opportunities for change and development. Today it’s time for the big picture. So let’s talk about the role of IT in insurance companies. Let’s talk about the role of technology in insurance and insurance companies. I am extremely pleased that we have two really prominent figures from the insurance industry here in this discussion. Christof, on your LinkedIn profile, your position description says “Reimagining Insurance”. Can you tell us what you are working on?
Christof Mascher: Oh, at the moment I’m trying to get to grips with my new status, but essentially I’m already dealing with all the issues of data architecture, data structure, in particular, on the side. So that’s one of my core topics: How can you translate insurance expertise into data, so to speak? And then, of course, the further pursuit of technologies such as AI and data analytics has always been a concern of mine, and I am now also using my time to educate myself again.
Michael Carl: Benedikt Kalteier joins us for the interview. He’s been a member of the Generali Board of Management for quite a while now, and since around the turn of the year he’s been Chief Business Officer Digital at Generali, with responsibility for CosmosDirekt and Dialog.
Benedikt Kalteier: Hello, thank you very much for having me.
Insurtechs bring new opportunities
Michael Carl: Let’s approach the whole insurance complex, and I thought maybe the easiest way to get started is to approach the topic from the outside, step by step, and look at what’s developing around the traditional companies in the insurance industry. So if we go into the whole area of insurtechs and startups, you both have a link in that regard. You, Mr. Kalteier are a board member of the Insurtech Hub Munich. You, Mr. Mascher, are very closely linked to the idea of expanding the IT of Allianz as a platform that is not only available to Allianz, but also to others, thus creating a completely new enabling function, business options, etc. that you can define at different levels. How do you both assess the role and the opportunities of the topic of insurtech?
Benedikt Kalteier: Then I’ll be happy to start. I think what we will see in the end is that insurtechs will lead to the insurance industry no longer being the only industry that believes that it organises the whole value chain itself. The production depth in insurance is extremely high compared to almost all industries, and I think insurtechs will lead to that going down. I am also very curious about the platform approach in the discussion with Allianz. I think the whole topic of technology and data is becoming incredibly important. We will increasingly have to be able to bundle different players, insurtechs, traditional insurers, and other partners together, in order to cover the value chain. We also see that insurtechs are eating their way through the value chain. They started at the customer interface, then there was a lot of deep tech and such topics, and now entire insurers. So I’m very positive about the whole thing. I think it can only help an industry to get a bit of traction from innovative forces.
Michael Carl: Mr. Mascher, what is your perspective?
Christof Mascher: I have a similar positive view of the insurtech side. I think the central thing is, does an insurtech succeed in linking the options of new technology with improvements or new qualities of customer service? So sometimes you get the impression that insurtech or digitalisation is just the use of some new tools. It’s certainly true, but it falls far too short. So wherever we succeed in using new options or new technologies, in such a way that we create better value, new value, and perhaps more convenient access for customers, I believe that there is potential for success, and then also the need for modernisation in existing insurance environments. I can imagine that being very positive. I am actually looking forward to the competition. I have never been worried that they would take the butter off our sandwiches. I don’t think so. On the contrary. When I think about areas like claims, for example, when I think about claims performance, I can definitely imagine tremendous new developments there. But it has to be simple, it has to be convenient. And that’s how the technology has to be used.
Michael Carl: Yes. We’ll come back to the topic of production depth and its reduction in a moment. Let us briefly share your positive ideas. You say that you can imagine a great deal in the area of claims. Could you be a little more specific?
IT drives smarter data gathering
Christof Mascher: Yes, insurance is essentially concerned with financial protection for assets that are important for the individual customer, target company, etc. The claims area then comes into play when the customer has to pay a premium. The indemnity area comes into play when this protection becomes serious. The asset replaces the car, the home and personal life, illness or even value chains within the companies that are damaged. Then the business processes damage, or processes damage performance, and there I believe that modern information technology offers a variety of communication and information processing possibilities, which allows us to modernise this antiquated way in which we currently manage, or perhaps have managed, damage. That it becomes much easier to quickly get a car repaired, much more convenient to understand what coverage I have in a particular claim, for example, in health insurance, and that’s all much faster, less bureaucratic. In my opinion, there is unimagined potential there, and that is indeed interesting competition between insurtech and traditional insurers. The traditional insurers have also fully understood this. So I can only say, from my time at Allianz, we have invested enormously there, and that can only benefit the customer.
Benedikt Kalteier: I would support that. After all, this is also the area where the most money ultimately flows, where, so to speak, premium income is suddenly added to expenses from an insurance perspective. And here, of course, all the customer-focused issues are very relevant. But of course there are also high financial opportunities. In my opinion, this has a technical component, a very exciting technical component, but I also think it has a component of how far do you want to go and how far should you go to price individually, and then to move away a bit from the collective, which is actually also a cornerstone of the insurance industry. So from that point of view, I think it’s also a very exciting topic to think about technology on the pricing side, in addition to claims.
We are an information-processing industry. It is to a certain extent already anachronistic that we have not actually pursued an ecosystem approach earlier in the area of the basis of risks.
Michael Carl: The word “understand” has come up often. In my view, it is precisely the possibilities offered by large volumes of data and their processing that help us to understand claims more precisely, to understand what the problem actually is, and what is still perceived as a problem on the part of the customer. Do you share this view that there are still significant steps to be taken at this communication level?
Christof Mascher: Absolutely. The problem is that insurance itself is actually an information-based product. It’s not as if we have specific physical material products like watches or cars. We are an information-processing industry, in the broadest sense of the word. It is to a certain extent already anachronistic that we have not actually pursued an ecosystem approach earlier in the area of the basis of risks. The other side of the coin is the disappointment of the customer who doesn’t feel understood, both in the underwriting process and in the claims process, when it comes to rejection or partial rejection, for example. And that is what we now have the opportunity to solve with modern technology, with a really good understanding of the information and the data. In the past, it was expensive to organise the data and to get a comprehensive picture of the customer’s risk situation or ecosystem. Today, this is much cheaper and we are therefore obliged to really create a new quality in value creation.
Data is everywhere, they key is understanding it
Simon Dufour: Mr. Mascher, on the subject of the information ecosystem. Where is this information?
Christof Mascher: Go to the archives of insurance companies, there is tons of information describing the situations. But there is also information that is lying on the street. Take a company. It doesn’t always have to be the private customer segment. We know exactly about the information, for example, the value chain, how a certain system works. This information has to be organised, and structured, and then used in processing, in calculating the right coverage of the right price. But, of course, also in processing and in carrying out claims processing. So the information is there. So if someone claims that the information, the data is not there, then that is simply a complete misjudgement of the past, the present as well as the future, because the bottom line is that in the information-based society that we are clearly in, the information is available.
Benedikt Kalteier: I would totally underline that. Of course, there’s always exciting information that’s outside, and you can also use that, and I think it’s also a lot about asking the customer less directly for information and getting it from intelligent sources. That makes it a little bit easier, of course. I think we have the vast majority of the information we need. We just use it far too little. But there is still so much potential, not only for insurance, but above all for what you said, Mr. Mascher, about advising and supporting the customer properly and saying: “Look, you have a risk here and you may even be over-insured.” I think there’s huge potential there.
“We are an industry that is based on being able to deal well with data, and I think that’s where the challenge lies, to change the way we look at data.”
Simon Dufour: But why aren’t we exploiting that potential more now? Is it because of the structure of the data? Is it because of the system that’s processing the data?
Benedikt Kalteier: I think it’s a mixture. On the one hand it’s what Mr. Mascher said. It was just, I think, more expensive to get to in the past. The technology wasn’t there yet. It is there now. What is also paradoxical is that, in addition to the fact that we are an information-based industry, we are also an industry that has priced quantitatively, so to speak, from day one. We are an industry that is based on being able to deal well with data, and I think that’s where the challenge lies, to change the way we look at data. I don’t have to teach someone to look at data in a new way, but I have to convince them that it makes sense to look at data in a different way, and I think that makes it a challenge. And the third aspect is, I don’t think I’ve been in the industry that long, but I could also imagine that the pressure wasn’t there before. For many years, it was relatively easy to earn very good money in our industry. And now there is simply more competitive pressure, more interest rate pressure, etc., and now we have to do more optimisation and more readjustment.
“This separation between product or insurance knowledge on the one hand, and IT and information processing on the other, must be overcome.”
Christof Mascher: Perhaps I can add to that. I would find two major lines of argumentation to answer your question, which partly coincide with what Mr. Kalteier said. I think there is an economic argument and an intellectual argument. The development of information technology over the last decades, always with this famous halving of data processing costs within 18 months, has already led to the fact that a completely different use was possible. I can remember, many years ago, my first IT project, cost wise, was to organize 50 GB of disk storage and that was an activity at that time where I had to go to the board several times to get appropriate approvals. That was really only due to the financial impact. So you can see how big the difference is there. So that has already made a lot possible and enabled a lot. That then leads to the second point. This is an intellectual challenge. I believe that the insurance industry has long made a distinction between the people who make the business, products, sales, depending on the segment in which the respective company was or is active, and the people in the basement, so to speak, who provided the manufacturing. And that doesn’t just apply to IT, by the way. Insurance has always been quite early to embrace modern technology. But it’s not just modern to use a modern tool, it’s modern to translate or transform your own business model, your own insurance expertise, what that particular product is, into software, into information processing. And that has to be done. This separation between product or insurance knowledge on the one hand, and IT and information processing on the other, must be overcome. I think that the current crisis in particular is once again acting as a real accelerator.
Better ecosystems require new skills and perspectives
Michael Carl: Now we have perhaps mapped out and described this field. Now you, Mr. Kalteier, used the term “decreasing production depth” earlier. I translate questions of cooperation also across company boundaries. Mr. Mascher, you talked about platforms. How does that fit together now? So is this something that accelerates and facilitates development, or is it more like the central hurdle that companies that want to be successful have to overcome?
Benedikt Kalteier: Well, I think it will accelerate the whole thing. We will need new skills to integrate into ecosystems, or to collaborate with partners. Those are skills that we’re rebuilding as an industry, and they’re not trivial. But that will lead to faster development, and everyone focusing on what they’re particularly good at, and that will lead to a better experience for the customer and a better solution. I believe that this will accelerate everything, and will also lead to stronger competition, and not everyone will survive this competition. I don’t think that in 10 years we will still have all the insurtechs that we see right now. Not fewer, but others. But also the long-established insurtechs will not be able to keep up at this speed, and will not have the investment resources to continue.
“The important thing is that we have to overcome the consumerism of the insurance companies, that we always want to develop and provide everything ourselves.”
Michael Carl: How do you assess that, Mr. Mascher?
Christof Mascher: I think that the depth of value creation, or the definition of this depth, is a crucial issue. I believe that it must be reduced and can be reduced. But it must be expanded. I would distinguish between vertical and horizontal value creation. If you look at the normal development within a sector, such as the construction industry, or any production, they used to produce electricity themselves. Nobody thinks about that. So we as insurers have to develop appropriate software tools etc. No one would write a database themselves nowadays. But it is very much necessary to go beyond the boundaries of conventional insurance. Ecosystems have already been mentioned. Partnerships have been mentioned. How can they deliver claims and performance efficiently for a customer, both in private and for an industrial customer segment, or an operational customer segment, if you don’t work together through integrates. And that means that they have to understand the partner and the business processes/expertise of the counterpart or partner. And that means more IT. But this is necessary nowadays, to find the right balance in this area of tension. That’s the real art. In insurance, we have a great opportunity, because we are an information-processing industry, to eliminate at least the physical standardisation. If we succeed in standardising the information blocks, i.e., in converting them into data, where we then exchange information with each other, communicate with each other, and then essentially work together in a decisive manner, then this has great starting advantages over other industries. The important thing is that we have to overcome the consumerism of the insurance companies, that we always want to develop and provide everything ourselves.
The importance of partnerships
Benedikt Kalteier: I would tie into that. What you said, understanding the partner vis-à-vis, I would underline that very much. I think that’s a big challenge between the very traditional large insurance group, and on the other side the hip Berlin start-up, with unfixed processes, unfixed standards that are always adhered to. But, in my opinion, that’s exactly where the great opportunity exists. When you bring these two very different systems together, a great deal of innovation, a great deal of creativity can emerge and then development can take place. But that is, I think, still a challenge that we face today and that we should not underestimate. So we have many advantages, I would also confirm, as you say, Mr. Mascher, that we can also connect a lot “simply” via data. But on this cultural collaboration level, I think there’s still a challenge. That’s also one of the core learnings from my work in the Insurtech Hub. When you look at which startups cooperate with which corporations. Why does that work and why doesn’t that work? It’s usually not the desire of both sides to work together. So how do I work together with the start-up, how do I work together with the corporation? If we can, then there are often huge opportunities for good collaboration.
“The real challenge of the problem statement is how can I translate insurance expertise into software?”
Michael Carl: Can I ask a more pointed question? We are now talking about different competencies. We’re talking about cooperation, instead of just wanting to do everything ourselves. Where do we actually get the assumption that, for example, such a platform for insurance must necessarily be offered by an insurance company? Shouldn’t such roles be easier, simpler and perhaps even more compatible for other companies?
Christof Mascher: I think the question is directed at me. I don’t think so. We have to distinguish between information technology and information processing. If it were the case that it’s just a matter of getting better at mastering a technology, then I would agree with you. But that is not the challenge. The real challenge of the problem statement is how can I translate insurance expertise into software? That’s the real problem statement. That means that I personally believe that all companies in the long run that want to outsource this thing, or want to put it outside the company boundaries, have no future, because in the end they no longer do information processing. I believe that there is a misunderstanding behind this, namely that people come from a very restrictive understanding of information technology. They say, “okay I’ll make a new tool, I’ll make a new database management system”. But we don’t need to do that. IT-Vendors can really do that better, and they should provide us, if you stay with the example of building a house, the bricks as a component to build the house. But how the house looks in the end is something we have to design ourselves. That’s why I believe that this cannot be outsourced, that this must not be outsourced to the boundaries of the company, because otherwise the insurance company would lose the opportunity to really design the processes, and even further, the products correctly. That’s why I think such an insurance platform, not the IT tools that lie underneath, but the insurance platform itself is inseparable from the insurance expertise, how the products, how the processes work, how they are described. After all, if you follow the digitisation discussions, that’s exactly the problem, that in this apparent antagonism you end up not understanding each other’s objectives. You have to force them together and do information management. That is the real challenge. In that respect, if you provide an insurance platform, the only way to go there is through insurance. I don’t think large IT companies can do that because they don’t have the knowledge, they don’t have the information, they don’t have the expertise.
“I think we still make too many tools ourselves today, and should buy in more, but that doesn’t mean that I give out what I do with these tools.”
Benedikt Kalteier: I would absolutely support that. I see it the same way. I think what you described, Mr. Mascher, is the core of the insurance. And anyone who gives up their core, so to speak, what does that leave them with? So from that point of view, I would absolutely underline that. I would separate that from “Does that then automatically mean that the insurance company is the orchestrator in an ecosystem, so to speak, or do the things that happen with partnerships then also have to happen on that platform?” I don’t think so. I think there a few that, at least in our industry, overestimate the importance of the insurance industry in the ecosystem. I don’t think the world revolves around the insurance industry in a lot of ecosystems. But, we must never give our cores away. I think we still make too many tools ourselves today, and should buy in more, but that doesn’t mean that I give out what I do with these tools.
Simon Dufour: Yes. I might come back to this reduction in the depth of production. People say yes, this has to happen. An ecosystem is also built by that. It needs, of course, and I think you also said this Mr. Kalteier, other skills within companies. Where should these skills actually come from?
No need to reinvent the wheel
Benedikt Kalteier: Well, the simple answer is: from the industries that are already doing it today. I think one big mistake that we often make is to think that we have to develop and discover and invent everything ourselves. I don’t think we have to do that. There are a lot of industries that have been working with a lot of partners for a long time, and that’s where we need to get our expertise. In my opinion, these can be very different industries. The one that has virtually perfected this is the automotive industry. They know how to work with partners in the value chain. But I think there are many other industries that are already good at this too. And I think it’s a skills issue. But it’s also a cultural issue. It’s that letting go. And I think that’s the even bigger hurdle than the skill hurdle.
Christof Mascher: You actually distinguished between communicative skills and between technical skills, if you can even make that questionable distinction. I think that in the communicative skills, I would fully agree with you, the division, the industry has to learn to develop a broader horizon and build the understanding and collaboration of the partnerships and so on. The expertise itself, the content expertise, you’re not going to be able to export that through partnerships. Take health insurance as an example. You need to understand how that works. You have to understand how a hospital works, how the business processes there work, how the whole processes, or stages of an illness in a hospital, are documented, processed and passed on as information. Just sitting and waiting without understanding the relevance and the depth of that information, and interpreting it for your own information system or for your own industry, and interpreting it correctly, that’s not going to work. It’s like car insurance. You just have to know how an auto insurance repair works. Otherwise, you’re just out of business. You see that with some startups that have no understanding about claims management. So that doesn’t mean you can’t work well with someone, if you understand well what the other partner does.
Michael Carl: I would like to look ahead a bit more. At this point in the podcast, we have talked several times about the question of which technologies, and which technological developments we actually believe will have the greatest potential for change and development in insurance in the coming years. Interestingly, in response to those questions, I’ve often heard assessments like, “If we get rid of our mainframes in the basement and bring our technology up to where it is today, we’re making the much bigger leap than if we adopt blockchain.” How would you two describe that? On which technological track do you expect the biggest development steps?
“At the end of the day, it’s nothing more than watching how can I structure quantities of information in order to gain better insights, better information, better support for the products and for the services that are linked to them.”
Data integration supports the value chain
Christof Mascher: Well, I agree with your assessment that if you replace a mainframe with a server of whatever kind, that might be technologically interesting, but it’s almost irrelevant for value creation within insurance. It is much more critical that we understand the information objects that describe our business model, our value creation, and that we can better integrate the information processing tools into our value creation, and thus increase the level of automation. And so I think the tools that you mentioned, like Data Analytics, Artificial Intelligence, Blockchain, etc., these are the real change opportunities because these tools will help us, on the one hand, to collect the information, to process information and to understand it better. What is data analytics, artificial intelligence? At the end of the day, it’s nothing more than watching how can I structure quantities of information in order to gain better insights, better information, better support for the products and for the services that are linked to them. A second aspect is, of course, everything that supports communication with the customer. So whether it’s portals or voice control, platforms of some kind, etc., that’s simply a clear issue and we don’t need to discuss it any more. I think everyone has understood that by now. Although I read in the briefing that someone doubts the role of customer portals. I wouldn’t understand that now. There’s still a lot to be done there, of course. So in particular these poor customer portals that we more or less currently provide to the customer is of course also driven by the complexity, and not understanding the insurance product. So, we are no longer able to present it in a simple, transparent way. But the real innovation, from my point of view, starts with the data and then building on that are the objects, the functions that help us manage the subjects better. I mentioned these tools at the beginning. There we will create a lot of new things and also an extremely high level of automation. That is almost more important than the depth of value creation. Finally, a degree of automation. Because it’s actually shamefully low when you look at our industry.
Benedikt Kalteier: I would also support and confirm that, as boring as it is. I would want to add one more aspect. I think what will also make the difference is to set up the technology in a way that I can react to new tools and trends. I think today I would mention exactly the names and mention exactly the topics that you also said. But if we were to meet here again a year from now, I would be surprised if we didn’t have one or two new topics. I think we have to be on the move in such a way that we can also incorporate and use them again as quickly as possible, because that is honestly the main problem with this large mainframe in the basement. As long as everything works well, it’s great. Having the ability to tackle new topics and work on new ones is, so to speak, is what makes it so difficult right now. I would perhaps add that as another aspect.
Michael Carl: And that brings us back to a very nice point. This already defines the initial question for our discussion in a year’s time. Namely, which topics have actually been added? Simon and I would like to thank you both, Mr. Kalteier, and Mr. Mascher. All the best to you both. Stay healthy.
Christof Mascher: Thank you very much for the invitation.
Benedikt Kalteier: Thank you.
Better data requires better partnerships
Our world is awash in data, and some industries are ahead in parsing that data into genuinely useful information that feeds back into delivering better customer experiences and building smarter products and services. For the insurance industry, resistance to sharing our collective skills with rivals, start-ups and insurtechs, is still a barrier to creating genuinely modern ecosystems, but it is a barrier that is being chipped away day by day by technological progression and changing expectations. At the heart of all this is IT, which is not going anywhere soon, but insurance companies wishing to remain competitive will need to start looking at IT not from the perspective of private in-house ownership, but as a piece of a larger puzzle that can be only be solved by forming new partnerships with other technological players in and outside the insurance industry, which will enable insurers to deliver the ecosystems of tomorrow that will attract the modern customer and keep them there.