Insurance Talks #3 – Future-proofing processes in the insurance industry

Future-proofing processes in the insurance industry

Michael Carl: The fact that insurance companies are under considerable pressure is something that I think we can take for granted in this podcast. Whether that’s digitisation, or the consequences of regulation, lower interest rates, the ever-changing expectations of customers – we could go on. We don’t want to complain here, but rather seek a discussion with those whose job it is, by virtue of their office, to shape the future and ensure that insurance and insurance companies develop in a meaningful and economically sensible way. In other words, successfully.

Today we will talk about processes. A term that, in my estimation, is under a certain tension. On the one hand, we are able to manage complexities of unimaginable proportions, i.e., the largest volumes of data, automated processes, individualised products, and so on. Good IT makes that possible. On the other hand, we also see a certain involuntary complexity, i.e., a wide variety of systems that generate a high system load, making processes slower. In short, where do we actually see IT at this point? As a driver of a transformation toward the future, or rather as a brake, in the interest of secure operations? That’s what we want to talk about.

Dr. Gerrit Böhm: So for us, IT, technology is still a means to an end. Yes, even if you sometimes have the impression that it becomes an end in itself. Let me say that no customer or sales partner congratulates me for having new software running, or for no longer using a mainframe computer. I get something for good, understandable, fast processes – and good products and services. So I get something like customer satisfaction for something that I then build with technology. So for us, it is still essentially a means to an end. At the same time, of course, I can solve problems that I had in the past in a different way today. So I have other solutions, and that’s why I can’t somehow ignore progress. But I think we would be well advised to focus on something that really benefits us, be it service that is nicer, faster or simpler – or a completely new product.

Michael Carl: How do you see it, Mr. Wittmann?

Dr. Thorsten Wittmann: IT is a means to an end. But my claim would be different. When you build a product or a process, nobody thinks about IT. Classic insurance product development works in a way that someone in sales might have an idea – a need -, and then something is built. And my experience is that people actually ask: “Tell me, how expensive is it to build this and couldn’t we maybe change something in the product and it would be simpler?” Does anyone ask how much effort it takes to operate and maintain the software afterwards? And does anyone actually ask, “What about the processes once the product is sold?” Well, we may have come from a world where insurance products or insurance sales say you sell it, and then the customer will get back to you if they wants something. But In today’s retail world, the exact opposite is the case. Everyone is constantly trying to communicate with the customer. Where is that in design? I do expect my development to be a guide on how to build a product or a process as quickly and simply as possible, and how to operate it easily, and how to enable an operation’s department to work through it efficiently.

Better to build together

Michael Carl: Do you also perceive this gap, Mr. Böhm?

Dr. Gerrit Böhm: Yes, it is extremely important that no barriers are created between the business department and IT. For example, we employ – almost exclusively -, salaried developers. That’s not the case for everyone in the industry. They sit right next to the business department. There’s no IT VOLKSWOHL BUND Systems GmbH or anything. There are virtually no hurdles. My only concern is that I am not misunderstood. I’m not a friend of this old world, where someone comes up with something and IT just builds it exactly as it was specified, but rather, as they say today, “challenge” it. I always say to my IT colleagues: “Make suggestions. The business departments don’t always know that the solution space has increased. You have to provide that input, and then they’ll be happy to take it.” That’s why it’s critical that it happens at eye level. We’re not celebrated for having anything new and modern in the background. Customers and channel partners don’t care. But the people who work with technology are of course much more in demand and challenged than in the past.

Michael Carl: I find this point very exciting. Is it enough to communicate at eye level? I mean, precisely because IT has knowledge here that the other departments and related areas don’t have. Especially because we know that the future of insurance will be very digital, shouldn’t IT be more proactive and make its own suggestions before someone comes along and asks?

Dr. Thorsten Wittmann: Well, I think it’s a question of togetherness. If you play it too offensively, then you produce more resistance. I think if you do it well, you can convince others that this consultation adds value. Quite often, a department has a requirement. And now it can’t get it implemented fast enough. What did they do? They looked for their own way, and they have a solution which works for them, but is not integrated. Everyone always thinks they can do the job of the other person. And, of course, I used to observe that the business department knew exactly what software they wanted, and their colleagues in IT knew exactly how to calculate everything. And then the question arises, doesn’t everyone want to retreat to their own area of expertise? If I describe what I want, then an experienced architect could, for example, make a good suggestion about how to integrate it into the system landscape. I believe that if both sides were to row back, and perhaps also admit to each other what competence is available in the various areas, then it would be possible to proceed in a more goal-oriented manner, and then the question you are asking would solve itself.

Dr. Gerrit Böhm: I think what Mr. Wittmann is talking about is a very exciting topic. It’s like negotiating. It’s not about position, but about interest. Position is: “I want to go left” and interest is: “Yes, actually I want to go to the bakery”.

Company culture is key

Dr. Thorsten Wittmann: What I would like to see is more of a look at corporate culture. My impression is that the insurance industry has the phenomenon that for many years it relied on technical expertise in the lines of business. And I think it’s quite difficult at management levels to develop a feeling for feasibility if you have no connection to software. And it’s funny that we always read how the insurance industry has slept through digitisation, which is true. But on the other hand, if any industry can claim to be digital, it’s us. The insurers have the least problems bringing people into the home office, because you just need a keyboard and screen for most insurance work. And yet, we haven’t managed to bring market innovations into our companies, which suggests to me the issue is a question of culture. There was no need to look at how high the IT costs were. It didn’t matter. The customer was satisfied if he got an interest rate of 4% and everything was fine. Today, that is no longer enough. The customer has other requirements. Something completely different is needed for them to experience satisfaction. And that’s where my point would be: I wouldn’t want IT to play such a leading role. I would much rather that more people and managers in the company have a connection to IT and software.

Dr. Gerrit Böhm: I think for both of the topics that Mr. Wittmann addresses, the close connection is so crucial. One is that we don’t want shadow IT everywhere. The specialist department builds something like this, and does something like that, and then falls flat on its face. On the other hand, this close connection is also necessary, which in the beginning may not be so familiar with the latest technologies. And I think that’s the key. You have to break down these barriers, if they exist, between the disciplines. One specialist department alone makes no sense. Nor does IT alone make sense. They have to be able to have healthy discussions together. A specialist department must not be afraid to approach IT and say: “We have the problem here. We’re thinking of the solution. Is there another way to do this?” Those are the individual conversations that are going to determine whether it’s going to be good or not, and that’s where I think you have to try to create a culture. And in this sparring, something cool can come out of it, where a specialist department sees your problem, and a team helps to deliver a solution.

Michael Carl: I find it very exciting and interesting that both of you are focusing so strongly on this aspect of culture. IT as a driver of corporate culture is already a strong statement. So, not long ago, people would have been rather surprised, I think, if you had held on to that.

Dr. Gerrit Böhm: My statement was that, if you have a culture in which the departments meet at eye level, then that is promising. Who is driving this and who is bringing it in, is a different question. And whether anyone can ensure whether it can be governed in this way, is yet another exciting question.

Processes need to be agile

Michael Carl: Well, you can’t govern the corporate culture, of course, but you can certainly initiate corresponding development processes, and at this point I had simply assumed that the person who needs certain development processes for the corporate culture would also be in a position to initiate them. I would like to add a second, culturally charged topic to the discussion. We had a passionate discussion about agility in this podcast previously. It was between Mr. Sommer from Generali and Mr. Burdack from Signal Iduna, who were in the process of restructuring their entire company. We are talking about processes today. Is this a topic that helps us to set up processes in such a way that we can realise the necessary flexibility and agility at all levels?

Dr. Thorsten Wittmann: I listened to that podcast, and I don’t quite see it the way your guests described it. So, when we talk about agility, the question is always, what do we actually mean? I think everyone has their own idea, so it’s always hard for me to discuss it. But what I do like to do is talk to IT specialists who are not from my industry, because I find that you get new perspectives. If everyone comes from the same point of reference, I miss a bit of input. What they describe to me, especially the retail sector, is that they are extremely agile. For me, I have very few systems. It’s a real benefit. But elsewhere it’s difficult for me, if I want to change something on one system, I always have the interaction with another. I have dependencies on a partner system, on a large one. This forces me to test very carefully. What is important to me is that agility requires extreme knowledge of structures and processes.

Michael Carl: Mr. Böhm, how do you see it?

Dr. Gerrit Böhm: Neither a customer nor a sales partner celebrates us on principle for whether we develop in an agile way or whether we develop via the waterfall methodology. No one outside is interested in that. As a rule, it’s the applications that are relevant for sales, for sales partners or customers, that have something to do with web front-ends, and so on. We actually develop in an agile manner, and then we are with Mr. Wittmann 100%. And, for certain problem-solving tasks, we really use this very specifically. But when I build a loop on our accounting system, I don’t do it in an agile way, I write down the business concept in a very classic way. Then I implement it and I’m done. But no one outside is interested in the method we use. They are interested in the result. I think Mr. Wittmann and I are quite close on this.

Processing the future

Michael Carl: I would like to look one step further into the future. We have now dealt with today’s challenges, and I think we have also discussed them in a really differentiated way with questions of culture and collaboration, the selection of methods, and so on. We have every reason to believe that the pressure to change in the insurance industry will not diminish in the coming years, but rather increase. And that the pace at which technology and technological possibilities are developing is also speeding up. Do you have any idea what this means technologically for your businesses? What do you expect? Do you have a relatively clear picture of the future, say, ten years from now?

Dr. Thorsten Wittmann: Well, quite honestly, I’m at a loss when someone asks me to consider the next ten years, or even the next five years. Though I have a clear picture of what I think awaits us, and I know how I have to set up my company to prepare. I would say that we need to be flexible and highly reactive. But of course I don’t know the specifics. So, Mr. Böhm, do you know what will happen in ten years?

Dr. Gerrit Böhm: I also have trouble with these statements. Everything is getting faster. And I say, what is really revolutionary and what is simply incremental? I think the last real revolution was the first iPhone, 15 years ago? That was a long time ago. Of course, computing power increases exponentially. But we also know, and you probably know much better than Mr. Wittmann and myself, that certain technological advances are massively overestimated. Ten years ago, I think a lot of people were saying: In ten years, we’ll be driving autonomous cars. And now we’re doing that in very limited setups. At the moment, I know what we’re working on, and I know where we’re heading with our technologies, based on what’s currently available. But of course, in ten years, something else will be contemporary. Today we have problems every half year, we have to upgrade the Java version and so on. So we’re trying to develop in such a way that we can deal with these cycles, which are partly forced on us, i.e. that systems in third-party software no longer support Java 8. We didn’t think that up, Mr. Wittmann and I, but these are simply framework conditions. So, we are trying, and I think we are all trying – this is not VOLKSWOHL BUND ingenuity – to position ourselves in such a way that we can continue to develop successively, because we don’t know what and which technologies will be available in ten years from now.

Dr. Thorsten Wittmann: We don’t have to look so much ahead into the future, but we have to look at what issues we have at the moment. I think the insurance industry in general, but also our company, if you look at the age structure, you see that the insurance industry is a very old industry. Not only how long it has been around, but also in terms of the average age of employees. But, that means that when I look at my departures, when people retire, there are large numbers of employees leaving the company. This means that know-how is gone and power is also gone. At the same time, I see that the insurance market is not necessarily booming. In other words, we are not seeing more insurers, but rather fewer insurers. Thirdly, if you look at the capital markets, they are not necessarily making it easy for insurers either. So we have to make sure that we can still exist as a reasonable company and offer our services when margins are falling. And then we have the customer, who simply behaves differently. And I believe that a lot of this will be digital. In the case of insurance, the customer wants personal advice at some points. They may not always want it at every turn, but they also don’t want, in my opinion, to handle the insurance business 100% digitally. And if I take that and just look at it as ingredients, then I actually already know what I have to do for the next few years. So I have to be very efficient. It’s always helpful to take a look at other industries. And I would say that the industry that is closest to us is the banking industry. And they are also eight years ahead of us in terms of regulation and everything they have to do. And if you take a look at what’s been happening in the print sector, i.e. newspapers or retail, then you can guess where things are headed. I think you have a good guideline to position yourself well for the next three-four years. And that’s why, I think, the question about where we are in ten years is not necessary.

Dr. Thorsten Wittmann: Well, when I look at it, the insurance industry is still a long way away from the widespread use of AI, just as far away as the car manufacturers are from autonomous driving. But that doesn’t mean it won’t happen. And if you now have a clear plan of what you have to do to implement it, then you will be absolutely well occupied for the next three to four years.

Play to your strengths

Dr. Gerrit Böhm: Well, I would like to add one more point. I think it’s important to be clear about your role. There are many trends and technologies. But, I think you can orient yourself if you are aware of your role. It makes an incredibly big difference whether I’m a life insurer, a brokerage in Germany, and so forth. Or, I have a general store and have all the distribution channels. Then I’m a stock corporation. Then I am an association. You have to be clear about your role. Am I a product supplier? Am I a direct insurer? Then I have completely different issues. I also have completely different technology needs. Whether I have to design something for the end customer and get them to go to the site, even though they don’t want to, because they want to buy a cool iPhone, not unsexy household insurance. Or, am I someone who doesn’t have to worry about the fuss, but who has to have a super good application to his business partners, i.e. brokers, multiple general agents, distributors? If I’m clear on my orientation, then it’s not that complicated to make a roadmap for the next five years. It’s only complicated if I don’t know where I want to go.

Michael Carl: Well, I think it’s a positive thing when you’re in a situation where you can think for yourself and shape things yourself. But I’d like to reiterate – question mark is a decent answer to such a question about the future. If we don’t know, and the requirement is that we must be appropriately agile and flexible to be able to deal with different scenarios, then it requires very active action. And I have now understood the answer quite clearly: find your role! Find your place and make clear where you actually stand right now! That brings us full circle back to culture. It’s a demanding task that you have there.

Dr. Gerrit Böhm: Yes.

Data drives an autonomous tomorrow

Michael Carl: Artificial intelligence is no more distinct than anything else we’ve already had here in this round. But let’s take a brief look at it. When you say, Mr. Wittmann, that we are not really there yet as an insurance industry. What will we be able to do when we get there?

Dr. Thorsten Wittmann: If an insurance company decides to use techniques, i.e. statistical methods, and if you want to apply them, you need data. I would advise anyone who wants to get to grips with the subject not to get carried away, but simply start by looking at what data you have. You will be disappointed by what data you don’t have. But, it would be good to start collecting and simply trying things out. And then you will realise relatively quickly that all these things are not as simple as you think. You can do a lot with statistical methods. We are currently trying this out and are moving forward very slowly, because if you are not careful, you will of course get a lot of difficulties. We bring in people from the outside who have experience. Once you’ve been doing it for one or two years, then at some point you’ll be able to walk on your own.

Michael Carl: What is your perspective on this, Mr. Böhm?

Dr. Gerrit Böhm: I think it’s a good example because it comes from a real problem. So, building and using an artificial neural network like that can be done. This is not rocket science. I’d say simply calculating a tariff in a different way, a logistic regression, or something to separate the two, is pretty damn good. Let’s take Mr. Wittmann’s example a bit further… If you leave out data protection, then you could imagine a scenario where no more questions have to be answered by customers at all, because I already have all the information, because I know their Facebook profiles and so on. And the customer doesn’t have to answer anything anymore, because I already know everything, and I rate it accordingly, and that’s it. This may have less to do with artificial intelligence and more to do with normal statistics. So it’s an important topic. No question about it.

Michael Carl: I’m learning a lot of new things about the charm of looking at a problem. That was a bit of a thread running through our conversation, and I think that’s a nice thought. I would like to conclude our discussion with that. I would like to thank all of you for your time, your commitment and your openness. I have enjoyed it very much. I hope you have, too.

Dr. Thorsten Wittmann: We’ll be in touch.

Dr. Gerrit Böhm: Yes, thank you very much. It was great fun.