Webinar

Behavioural economics for the insurance customer

Do you find yourself surprised when your insurance customers don’t act how you might expect, especially when it doesn’t seem to be following their own best interests? Would you like to hear about some alternative, proven approaches for you to help guide your customers through the complexities of insurance products: including claims, policy administration and filling out application forms? Behavioural Economics (BE) is an emerging field combining psychology and economics to offer a more realistic picture of how our customers actually behave. In this webinar, Nick Mingo, a Behavioural Scientist from Swiss Re, shares examples of some of the behavioural economics insights that he has been applying with life and non-life insurers globally for the past five years. 

Presenter:

  • Nick Mingo, Accelerator Lead Asia – Behavioural Economics, Swiss Re (Hong Kong)

Ben Telfer: 

Hello everyone. I hope you all are well. Welcome to today’s ICMIF webinar, “Behavioural economics for the insurance customer”.

I’m delighted to be joined by Nick Mingo, Nick is an accelerator lead in behavioural economics at Swiss Re, one of ICMIF’s supporting members. Nick was one of the speakers at last year’s ICMIF Conference in Auckland, New Zealand, and we’re delighted to welcome him back to present on today’s webinar and share some of his insights and answer some further questions on the topic of behavioural economics. Nick, delighted you could join us, and over to you. 

Nick Mingo: 

Thank you, Ben. Very glad to be here with you all. So you see here on the title screen, on the right here, that this is my work from home arrangements in Hong Kong. So you can imagine me sitting here delivering this now, speaking to you. A real leafy outlook in a concrete jungle of Hong Kong, but this is where I’m sitting now speaking to you all around the world. 

So before I get into behavioural economics for the insurance customer, I just wanted to start with a story. This is my son, he’s three years old and he lives with me here in Hong Kong, and I wanted to share about how watching him grow up has really helped me to understand more about aspects of human behaviour. So each day he encounters lots of things for the first time in his life. So I see him trying to make sense of the world. So he gets guidance from us as parents about some specific things, rules about what he has to eat, he has to brush his teeth, how he interacts with friends and when it’s time to play. 

But I also find that he doesn’t respond well to explanations about what to do that are from my perspective, like you have to be quiet now because this is a library, or that you’ll wake your brother. So I also see that he applies what he finds works in one situation and what turned out well for him once, and he applies that to new situations he’s faced with to learn. 

I noticed that he has rules of thumb that he applies, and how he generalizes what worked in one situation to another. Like he’s always saying green means go and that’s a little rule of thumb he uses to generalize across all situations. And he also looks to see how others are behaving and then copies that, which is good and bad. 

Of course as a three year old he also learns by testing those limits, and he sees which rule or sometimes which parent breaks first. So I have to admit that I actually use some insights that I’ve learnt from my study of behavioural science myself to influence his decisions for the better, and we’ll come back to a few of those personal examples a bit later. 

So to kick off into a short introduction about behavioural economics, I like to contrast between traditional economics and behavioural economics as a good way to explain this. So much of the economic theory that people would have studied in university would be around traditional economics, to say well people have this… humans are rational, they have these set preferences, and they always act in their best interest, and they will follow through on what they intend to do and that lots of information helps them to make the best decision. 

The levers we have to pull under this traditional behavioural model is we’re providing lots of information to people, so the more information the better the decision someone will make. We also use financial incentives, so we change the prices of different behaviours, so that the costs and the benefits are easy to weigh up. 

If we contrast that with the behavioural science approach, that understands that people aren’t always rational, and this concept developed probably came into popular lexicon about 10-15 years ago, but we were noticing that this traditional approach doesn’t fit the observed behaviour. We don’t see that humans behave in this traditional computer like way in their decision making. 

Then we found that people are really affected by the context in which they make a decision, so the way they behave in one scenario doesn’t hold true so that people aren’t always making these fully rational decisions. So some of you may have read books like Thinking, Fast and Slow or Nudge or Predictably Irrational or seen TED Talks on this topic. 

That’s really gone into popular culture at the moment, because there’s good science behind it, two of the economists who have won a Nobel Prize for economy in the last few years were actually practitioners of behavioural economics. So it’s really a combination of psychology and economics, so my own background is trained in psychology and a master in behavioural science, and I’ve been working in insurance for most of my career, so it’s bringing those two worlds together. 

Another way to explain it, drawing on Daniel Kahneman’s work from Thinking, Fast and Slow, is that the mind has these two different modes that it uses to make decisions. So system two, which aligns with the traditional economic approach, is really slow, it’s conscious, it takes in all the possible information about a decision, and then analytically works out the best course of action. This takes a lot of energy, it takes a lot of time. So we find that the mind also has this system one, which is fast and unconscious and associative. It makes quick links and it quickly helps us to make a decision that we can act on. This aligns with the behaviours and the theory we see from behavioural science. 

This dual system isn’t a weaknesses in human development, probably the opposite, it’s actually helped us to evolve and survive because we couldn’t use system two, the slow system two, to make every single decision in our day. We make thousands of decisions in one waking day so we’d never even get out in the morning if we used this system two to rationally weigh up the pros and cons of every decision we made. 

In behavioural economic theory we see some patterns in these behaviours that help us to study and to predict how people may behave in the future. So some of these patterns I’m going to bring up on the screen now you might recognize from your own behaviour or from the behaviour of your customers, and in this way we can predict certain behaviours that humans are more likely to take. 

If you look up cognitive biases, which is what we call these, so biases or mental shortcuts are the words I might use today, Wikipedia says there’s 191 different cognitive biases. So they’re social or memory biases. So that’s a whole lot for practitioners of behavioural economics to try to work into their work. Some of them even tend to be opposite, so in that way we recognize that the context of the decision is the most important and to observe and measure how we see the actual behaviour play out. 

But I can recommend two benefits for the customer, when the insurer actually recognizes and understands these patterns. Because firstly you can recognize when these types of biases, they derail a customer from acting on their good intentions. So you’ve already got a customer wanting to purchase your insurance or engage with your insurance, but something stops them taking that last step. So an example may be choice overload where you just give them too many choices of policies or options to choose from, and they find that the more information makes it even more difficult to make a decision. 

But then also help the customer to use these rules of thumb, these shortcuts, to make sense of a lot of information. So when we’re presenting information, we shouldn’t assume that the customer’s going to use that at rational system two to make their decision, we should assume that the customer’s going to need all the help they can get and look for the context in the decision to take a shortcut or make sense of a new situation. So in that way, social norms that we’ll look at a bit later, is a way that they look around and see what are others doing in this situation. And that’s a really powerful cue to how I should behave. 

Some of these shortcuts, they might prompt some questions or maybe even some ideas about how you can use these insights to human behaviour in your work with customers. So if there’s anything here that springs to mind, then from now start sending through those questions that we can address towards the end. So one that I think is really relevant to insurance is that we care about immediate benefits and costs more than those in the future. 

Sometimes millennials get accused of this really heightened present day bias, the gratification effect. For us in insurance this can be a real challenge because we’re asking for money for premium right now for the promise of a payment or coverage in the future. In best case scenario for the customer, they should pay today and hope never to claim. So we’re asking for paying today for a future promise. So compared to other purchases the customer makes, we’ve got a real challenge to overcome this behavioural bias, so we should recognize that and not assume that a customer will see the future safety net compared to the cost of the premium today. 

To apply these to customer behaviour in insurance, so firstly to just general customer behaviour, we’ve got a couple of case studies outside of insurance to help illustrate some of these points. The first one comes from London and the problem statement that the transit authority in London had was how to increase the satisfaction of users for their buses. So they looked at all the feedback coming in and they were very tempted to say well we need to put more buses on, we need to make buses run on time, we need to have more buses so that people wait less times, people are complaining about that. And that’s all very expensive to implement those type of changes. 

They worked with some behavioural scientists and they came out with one change to this setting here at the bus stop that showed the biggest jump in customer satisfaction. In fact, it was this. This is the timetable of when the bus is coming next. So simply by telling the customer when to expect the next bus, the customer satisfaction went up. So they didn’t need to make the buses come sooner, they didn’t need to put more buses on or even shorten this waiting time. Simply by telling the customer when the bus was due was enough to make people feel satisfied about this. 

In behavioural science we give this type of approach a fancy term and we call this operational transparency, in that humans are a lot happier to wait if they can see their progress. We see this in society in some other industries, some of you may, in your countries, have the Domino’s Pizza tracker, or the Uber app where you can visually see your driver coming to pick you up and the route they’re taking. So compared to before these apps were available, then really you made the order and you received a thanks for your order for a pizza or a taxi and then you had this big void between thanks for your order and the doorbell ringing. So you just had to wait during that time. But now there’s things to occupy you in that time and you can see a progress towards what you’re going to do. 

You’ll see that I’ve used this insight in one of our live insurance customer trials with a client in Japan a bit later that I’ll explain. When we conduct actual research to use this insight, we see that customers are more patient and more understanding of delays. They even feel that the waiting time was shorter. So just having these updates plays with our perception of time to say well, it was shorter, I actually waited shorter, even though they may have waited even longer than someone with no information, and they also rate their trust of the process as higher. So lots of good wins for customers there. 

The second example is around the behaviour challenge of smoking. So this is a big and curly public health behaviour challenge to say well, lots of things would be improve if our population quit smoking, stopped smoking, public health and personal health. So lots of strategies are put in place to try to get people to stop smoking. 

I took this photo in Toronto, I was traveling, and it says… Never off as a behavioural scientist, I love to see the different behaviours around. So I couldn’t help but taking this as an example. So this was outside a big office building where the smokers would congregate. So the designer of this sign has taken the opportunity to provide some really rational information here at this time. 

I applaud them that they recognize that this is a time to influence smokers, but maybe not in the most powerful way. They’ve provided really rational examples and information about the risks of smoking. We find that in smoking it’s probably not an information problem that leads smokers not to quit. I think everyone on to planet’s aware that smoking is bad for you, they know the health implications and the disease implications. So providing this rational information is probably not that powerful when the person’s already purchased the cigarettes, they’re there with their colleague, they’re about to light up, they see this sign. It’s unlikely they’re going to put the cigarettes in the bin and say, “Oh, really? That’s new. Cigarettes are bad?” And back to office. 

Here’s another way that I saw, so another example of how the designated smoker area’s been labelled. So here more emotional, makes you feel uncomfortable to be labelled as a smoker, and so it’s a tongue in cheek way of saying you’ll be in the social out group if you are a smoker. You can see the man there, even hesitant to stand in the designated smoker’s area. 

So two different approaches. I would say the one on the left appeals more to the system two, the rational, slow, analytical way of thinking; and on the right, the system one that makes that association straight obvious association that smoking to the coffin and being in the out crowd. 

Swiss Re have quite a history of applying behavioural insights to insurance. That came about because we found the traditional ways of customer research, customer focus groups, asking people what they would do, what they would buy, didn’t predict the eventual buying behaviour. So we’d launch a product and people, even though they said they would buy it, it didn’t sell that well. So we found that assuming customers had insight to their own behaviour wasn’t that accurate, and the people were influenced by the context in which they made the purchase. So we invested in put a lot of time to applying behavioural insight with our insurance customers and clients around the world. 

For a lot of a people buying insurance is a new and unfamiliar experience. So a bit like Chester, my son, what kind of mental shortcuts are they going to use to look around in the context of the information? So Lemonade is an example of an insurance company from the US who embeds behavioural science to the very core of their product and experience design. 

Dan Ariely, one of the most famous behavioural scientists, is their chief behavioural officer. So they have that position. Maybe one day I’ll get to be a chief behavioural officer for someone, that sounds like a pretty cool title. So here he’s saying that if you tried to design a system to bring out the worst in humans, it would like insurance. So not a ringing endorsement for how our industry treats our customers. 

He took on the challenge to use behavioural insights to make the journey better for customers through insurance. For those of you who don’t know Lemonade, they offer gadget insurance, small home insurance, content insurance, like you see in the graphic here. And they use insights to build the trust. So one of the things that they’re quite proud of is they developed this trust with their customers, and they cite the fact that even customers who make a claim for one of these items, they say my laptop was stolen or I’ve lost my phone, they later find that item, realize it wasn’t actually stolen, and they ring up Lemonade and say, “I want to give the money back, I found the item.” So that’s what they see as the trust that they build from their customer centric behavioural principles they use. So they see it as a tech company doing insurance rather than an insurer doing an app to engage with their customers. 

Another insurance example about a particular human bias. An example here of how an insurer has recognized these human biases and used these to guide their customers better through the complex world of insurance. This is for some health insurance or medical insurance in some markets. So they could have presented me with a big table of all the different options, all the different covers, all the different rules for claiming, and said here you go Nick, more information, that will help you make the right decision. 

They’ve recognized that I need to classify the different policies, I need to just know enough to make the next decision as to which policy I’m going to go with and then maybe I can tailor it later. So they’ve labelled them, they made the price really clear so I can compare easily. But then they’ve… So that’s what we call choice architecture and that they see themselves as architects in designing the way that information is presented. So this is a really important role that the behavioural scientists play in organizations, is to be choice architects to design the way that information is presented to people. 

This is an example of good choice architecture, the choice makes it easy for me to make, not too much information. The second insight they’ve used here is one that we call social norms. So you can see here in the green they’ve said, “Well, of these four, you still may be struggling to decide which one, so we’ll give you a little clue. We’ll say, ‘Nick, this is our most popular for people like you.'” So that draws on the bias or the human pattern of social norms, social proof, that says humans are really comfortable to do what other people are doing. And in the opposite they are quite uncomfortable when they feel like they’re in the out crowd or they’re doing things that are inconsistent with how other people like that. And even more powerfully, they like to act similarly to people like us. 

Here the insurer said this is our most popular hospital cover for people like you, so couples in New South Wales, which is a region in Australia. People like you pick this cover. So I feel more comfortable in picking this one and if I had no idea between the four, I know that this is not a disastrous choice. So I can pick this one and feel comfortable that I’m not making a bad decision. 

Again, in making complex choices, quite often humans are happy to just be satisfied or to settle for not the worst option. So they don’t necessarily always try to find that incremental maximum benefit, the very best choice out of a range. Often you’re comfortable and you’re satisfied just to not make a worse decision. And that makes sense evolutionary. We just want to avoid a really bad outcome, we want to avoid eating the poison berry or avoid getting eaten ourselves. So we make these rules of thumb that avoid a really bad decision. And we can live with it if it’s not exactly the best. 

For decisions we need to make the absolute best, for making very complex financial decision, then we can employ system two and really analyze the pros and cons and all the information. And when we need to make hundreds of snap decisions a day, we’re happy just not to make a disastrous decision. And this kind of choice architecture recognizes that and helps me make that decision, so I appreciate this. 

Another example, walking the streets I’ve seen this sign to say most students have dental insurance. So if I’m a student, I’m seeing that, I can’t help but feel a bit anxious or a bit uncomfortable if I don’t have dental insurance. So we find this is a very powerful behavioural economic insight. And marketers have known this for a while, so they apply these insights, so I’d say the difference between marketers applying these understandings of human behaviour and behavioural science, is that behavioural science is more rigorous in the measuring and the naming and the defining of these terms so that we can say yes, we see this pattern, and that it leads to a 14% increase because we’ve run an experiment where one group received one message and the other received the behavioural message, and we kept everything else constant and we can see that this works. So that’s the science in behavioural science. And you’ll see that as I show the examples that I’ve been involved with in insurance, that we’ve always used that scientific approach that we can actually say in this scenario, this is what worked and we learned, but also yeah how much. 

Another human bias that I’ve seen used in insurance, helping to guide the customer through a complex area, this one’s called active choice. And the bias goes that humans can often avoid decisions if we’re not required to make them. So again, we’re nervous about making the wrong decision. So sometimes we’d actually prefer to make no decision rather than make a bad one. So if we’re not that familiar or not that comfortable with a decision, we may just avoid making it altogether. So usually in purchase journeys, that means dropping out and not actually making the purchase, or not following through and changing your policy. 

In customer service and having the customer’s best interest in mind, we want to avoid that because it’s often not in the customer’ best interest to avoid that decision. So here we see that rather than just saying click here to get some insurance, the designer of this webpage or the choice architecture has forced the customer to make a decision. So they haven’t said, “Oh, you must take insurance,” or defaulted in when it can be against the customer’s best interests, they’ve said at least make it an active decision on whether you want the insurance or not, rather than just passively avoiding thinking. 

In this one we actually call this enhanced active choice, because not only do you have to choose yes I want insurance or no I don’t want insurance, for the no option they’ve laid on the implications a bit, so they’ve made it a bit more real about what your answer means. So they’ve said if you answer yes you’ll get all these benefits but if you say no then I choose not to protect my purchase. So in that way we call that enhanced choice because we’re actually guiding the customer as to some likely implications of that choice. And quite clearly in this scenario, steering them towards one particular scenario, but still maintaining their free choice. So we’re not restricting their ability to say no to this insurance. That’s important about behavioural economics, that we still preserve the free choice for the customer. 

Here’s one last example from outside… I’m sorry, from actually the members of ICMIF. So this example comes from New Zealand and here you see a starling nest that’s inside a tractor engine. So you might imagine that can be quite a problem for the farmer. Starlings can build their nests in less than 20 minutes inside the warm engine… inside the engine bay of a tractor. So that can be definitely overnight but even as short as someone going for their lunch break. So they come back and start up the tractor and the nest heats up and suddenly the whole tractor’s on fire. 

That follows through to property motor claims for the agricultural insurer, and so it’s in both the customer’s interest and the insurer’s interest to encourage the policyholders, encourage the farmers to check their engines after they come back from a break so they can clear out the nests. So this is an interesting behavioural challenge, so how do we get the customers to… or get the farmers to check that? 

One approach would be information, so tell them that this is a problem. But that may only be partially effective in actually changing the behaviour on the day when the farmer gets into the tractor. He might not remember the radio ad or the newspaper ad that he read about the information about starling nests. 

This insurer quite innovatively has deployed some stickers, actually, so that they can provide the stickers that go on to the tractor. Just trying to advance through to the next slide. There we go. And you can see here, the example in place, that the sticker goes into the cab of the tractor and so when they get to go, jump in, they see it and are reminded in the context. 

This is recognizing that the context matters about when the decision is made. Simply providing more information that starling nests are a risk, they recognize that the most powerful way to change the behaviour is to have a timely intervention that’s right at the right time and is obvious and salient to the person needed to make the change in behaviour. And that’s what we see here in this example. So well done to our colleagues, our people at FMG from New Zealand. 

I’m going to show you a couple of examples from Swiss Re around the world that we’ve used behavioural science with various insurance problems. So the first one is a bit of a fun for us all on the webinar today. In social science, in psychology, quite often I get the challenge that says this is all just common sense. I don’t know why you studied for five years to learn all this, I could have just told you what would happen. That’s obvious, that’s common sense. 

Often that’s after the fact when they see the answer and say yeah, I could have told you that would happen, but I don’t have a good time machine to take them back and say prove it. But this is a way that we can do that exercise. So in this way I’m going to see well, is it all just common sense, could you predict which behavioural intervention would be most powerful into changing a behaviour? 

So the problem statement here, this is a real one that we worked on with an insurer client. We wanted to get customers to change their monthly premium payments to bank transfer. So at the moment, some customers are still writing checks, going into branches to make payment, or setting up a really inflexible regular payment order from their bank that they need to change it every time that the premium might change. 

It’s really in the customer’s interest to change to automatic bank transfer, but getting people to actually get around to doing that is tough. So here are five different behavioural techniques that we proposed, and then we actually tried all of these. So this is where, we’re coming back to this culture of experimentation, where the only way to find out what works is to try different approaches and measure which one works. 

The first thing we used was a message to the customers to say bank transfer is easy and it’s protected by a direct debit guarantee from the government. So it’s easy and it’s safe. The second one says well your current way you’re paying is quite a lot of hassle for you every time, so inherently you should make a change to this easier way. The third one was if you do what we ask and switch to transfer payments, we’ll give you a $10 voucher, so a bribe. Everyone who changes gets the voucher. 

The fourth one was similar but perhaps a bit more behavioural and it said well, we’ll give everyone a voucher, and then ask politely for you to switch your payment method. So perhaps this is reciprocity, and we’ve done this thing for you and you’re going to want to do something for us in return, so give everyone $5. 

The last one is to make the request seem a bit more like an admin task, so to say well it’s time to update your payment details, here’s the two options. So a bit like an admin message but also bringing in that active choice that they still have to, even if they want to keep it the same way they need to click oh keep mine the same way, or change to automatic bank transfer. 

Ayou’re sitting there listening to the webinar, just have a think about which one you would predict is the most. So we’ve really done this experiment, we know the results, I’ll show you in a second. So in your head you’ll know whether you’re right or wrong straight away. So think if you had to predict or you had to make it better, even, which would you say was the most effective strategy? 

I’m going to show the answers now, so remember your number. So these are the actual results that we found when we did the experiment with different groups of customers. So number two, talking about the payment method being a big hassle. That was the most powerful in actually changing behaviour. For those of you who were at the ICMIF conference last year in Auckland, I did this live on stage, and we had a voting app, and a lot of people predicted that the monetary one, so the $10 voucher, the bribe. 

I’m always surprised when people pick that one because we’re half an hour into a talk about behavioural economics, and I’ve talked about how the traditional economic model doesn’t always explain behaviour, but still they can’t help but assume that customers will be really powerfully motivated by money and pick that one, they’ll pick three. 

The learnings that we got from this example is that people aren’t solely motivated by money, and in fact if you’re trying to change a behaviour and you offer money for them to do that behaviour, then people can feel wary. They feel that well if you’ve got to pay me for me to take this action, maybe this action isn’t really in my best interest. 

So in social psychology we’ll say this extrinsic motivation, the payment, crowds out the intrinsic motivation they already had. So they already may have thought it’s a good idea to change but just couldn’t get around to it, but now we’re saying that we need to pay you to do this, and that makes people think again whether it’s really in their best interest. So that can have a backfire, feel like a bribe. 

And yes, the reciprocity, the give everybody a $5 voucher outperformed the $10 one, but we had to pay out a lot of $5. So when we’re looking at cost per switch, this one was even higher than the payment one. So although it had some behavioural insights, it’s still probably not the best one to apply in this context. But the best performer is the one that passively communicates how the customer can benefits and is at no cost to the insurer. 

A lot of the time, the positives that we see from applying behavioural insights is that we can get meaningful behavioural change quickly and cheaply. So we don’t need to offer a discount straight away, and a few of our projects with clients. At the moment, the way that they’re trying to influence customer behaviour, whether it’s for sales or whether it’s to stop people cancelling, their first go-to is a discount. So they say well, someone rings up to cancel, oh wait, don’t cancel, here’s a 5% discount. We’re finding that if we can apply these behavioural insights instead of going straight to a discount, then we can still encourage the right behaviours but we can even save the insurer some money in the way that they approach the behaviour change challenge. 

The next example I’m going to give is from Japan. So this one is using the customer, and you’ll see all the examples focusing on customer service and customer benefit. So this is where customers call up to make some changes on their policy. For efficiency, they get many thousands of calls a day, so they screen the calls through the automated telephone machine. Lots of you are probably frustrated and hate that and really surprised that it’s still a thing in 2020, but we wanted to make that a better experience. So we added some behavioural insights in, so the challenge from our client partner was well let’s keep more people in the telephone system so they complete their policy change automatically, they don’t need to talk to a human, they can do it at any time, and ultimately they’re more satisfied. 

We added these extra lines in and that in itself was a challenge because quite often when designing these types of choice architectures, we’re looking to minimize and think the less words said the better. We’re actually recommending to increase the words, so we had to convince the client of that. 

But they’re saying that most people like you find this method the easiest, press one. So remembering the concept of social norms that people feel more comfortable taking this action if they’re aware that most people like them have done that. So we added that to the wordings, and then we also, like the London bus example, added this extra line in towards the end, to say, “Well, great news, you’re almost finished.” That was the encouragement to hold on. 

We’re finding by measuring the drop offs at different stages, we saw that this was a key point when people would just hang up and think it’s all too much. By giving them and encouragement, say we’re almost there, giving that operational transparency. You’re giving them a signpost that you’re nearly there. And people then, if they know they’re almost there, they find it quite… they’re a bit uncomfortable to give up their progress. They know they’ve come this far and so they don’t want to give up their progress and start again. But they’re also not thinking this is going to take forever. 

Aligning with the concepts of scientific experimentation, we had the previous version, which didn’t have these extra words in it, then we ran for two weeks with the new version and all other things being similar we found that there was an increase, a significant increase of people being able to successfully complete their bank account change or their policy change through the automated system. So here we can quantify the impact of this behavioural change. This is a good message for CFOs and for the business in general, that we’re making a real change for customers. 

The next example I have is from Australia and around… this time we’re applying the nudge to doctors. So it doesn’t just need to be the customer that you can apply these nudges to, it can be anyone, any stakeholder in your process or your system. So here’s a form that the insurer asks the doctor to fill out for an insurance claim and they said, well the doctors aren’t very good at filling this out. We get lots of incomplete forms out, we get doctors writing, “I can’t answer this,” and it holds up our assessment of valid claims. 

To help them I designed a new table to go into their form, a new way of asking this information. So I applied a few different behavioural insights to this choice architecture. So I made it more relevant to the doctor by adding those examples, I tried to break up the automatic thinking. So by recognize that humans take these shortcuts and they may just think 5 kilos or 10 kilos is the only answer I should give, I’ve broken up the answers so they’re not just multiples of five. I’m actually trying to get the doctor to use their system two here so recognizing they probably use system one, the quick associative one and just tick 5 or 10 kilos, or just ask the patient what they can lift. 

Here I’m trying to nudge them back into using that more deliberative… because this should be something they use their slower system to make this decision, so I’m making in some ways more difficult so that they actually have to use that more deliberate side of their decision making. 

When we measured this compared to the insurer’s own version we saw a really great uplift in the percentage of doctors completing this table. And that has benefits for the customer, be able to get access to their benefits and to get some more support to get back to work. 

Now moving on from these examples from the real world, I thought I’d just spent a few minutes writing advice about how you may apply these types of insights with your customers. So if you have what we call a customer pain point, a behaviour that you’d like to change, how much you use these patterns or an understanding of behavioural economics to benefit your customer? 

I’m going to firstly talk through one framework, and this is not a Swiss Re framework, this is from the Behavioural Insights Team from the UK, and so they were a government department, and a lot of behavioural economics work has come out of government units, so a lot of behavioural economics is used in public policy and how to tackle big behaviour change or small behaviour change problems in government. So they propose this framework that there’s four aspects that you can look at in a behaviour challenge, and that is can you make the desired behaviour easier, can you make it easy to do what they want, what you want? Can you make the nudge timely? So a bit like the sticker on the tractor, the information or the prompt is a lot more powerful if it’s at the relevant time that person is making the decision. 

How can you make the desired behaviour more social? So what social aspects can you bring? So that people feel like they’re doing the right thing that other people are doing, or how can you make their behaviour more socially obvious that they feel more pressure and more desire to do the right thing. Then looking at attractiveness, so when we’re looking at information how can we present it so that our eye and our attention is drawn to the right parts of the information. So we call that salience, how can we make the information salient and also incentives come into play here, how can we incentivize the right behaviour? 

This could be some further reading about that, an easy framework to make some quick changes to your scenarios. And another, I guess, example from the real corporate world is Uber. Uber put this in the public domain about how they use behavioural science to enhance their customer experience and their customer processes. I think Uber are quite a good example of being hyper customer focused, and they’re in the digital world so they have a good platform to test and learn and use different versions to customers and measure in real-time and very quickly the effect of the different versions on customer behaviour. 

You can see here they publicized the process they go through from problem statement, to using science, the behavioural science ideas, so maybe one of those 140 different human biases from Wikipedia, and then looking at the literature from behavioural science about what different approaches are likely to work. Then experimentation. So I mentioned a few times that a key aspect of behavioural science or applying behavioural economics is this experimentation that we test and learn, that we want to actually measure the effect of our intervention. So we learn what works in that context, because just applying what worked in North America with students to a behaviour challenge in Japan, we shouldn’t expect that that will directly work. So we can have the ideas from science but then we need to apply that and learn in the new context. 

The last thing I’m going to cover is around ethical considerations, so we talked about the power of behavioural economics, and with power comes responsibility to use this power well. So a question I get asked a lot is well, how do we know that we’re being ethical in our use of behavioural economics or what reputational risk is there for me to use these techniques? Is this in the same bucket as subliminal advertising, am I brainwashing my customers into doing what I want? 

Behavioural science has a discipline of being quite clear about what is ethical in using behavioural economics, and Swiss Re also has a very strong stance on how we apply behavioural science to only be in the interest of what is the best interest of the customer. So we don’t promote behaviour change to what’s just profitable for the insurance company. 

A true nudge maintains the free choice of the customer. So we’re not trying to restrict or hide the choices, and it should be easy to avoid. So if a customer doesn’t want to follow the nudge they should just be able to carry on with their intended action. 

It’s also important to understand that whenever you present information or a decision to a customer then you are acting as a choice architect. So the order and the format that you present those options, they will affect their behaviour. So there’s no neutral position to presenting information, you can’t just absolve responsibility and say, “Oh, I have no influence on the customer, they have free will.” There’s no neutral position. So the most ethical position seems that we should fully understand and monitor the effect that our choice architecture has on the customer. 

To illustrate what I describe as a sludge, so if we say that nudges are positive for the customer’s interest, then a sludge can be something that’s not in the customer’s best interest. Here’s an example that I came across myself. So it’s a rugby sevens, a huge event every year in Hong Kong. Tickets are tough and they’re hard to get. 

I went on to a ticket reseller in desperation trying to get my tickets to the rugby sevens. This is the experience I had that led me to believe they were using behavioural economics perhaps in not my best interest. So I searched and I found these tickets were available. Then these popups started coming up. They said quick, 10 people have viewed this. Now there’s a countdown that comes up and if I don’t get it I’m abandoning them, it’s really emotive language, and if I want to go back I’m abandoning my tickets. And then they’d say well, all these sections don’t have any tickets, you’ve missed out. 

Then the timer pops up again and I have to actually click it to get rid of it, so my attention is drawn to this timer, and then finally the one that I’m looking at, about to click, suddenly changes and says sold, that someone has just bought it and someone’s going in my place. So this panic, and I found my heart racing, and even reliving it now I’m feeling that effect. So I would say that is not nudging in the best interests. 

Lastly here, this is an example of using the automatic processing. This is not actually a renewals certificate, this is an ad for a new policy. So the insurer here is sending this out and saying, “Well, hopefully they just think this is their renewal notice and just sign it and pay,” but they don’t actually have their policy with that insurer, they’re trying to get the customer to switch, take advantage of that automatic processing. 

I’ve got a bit more time for questions. I’ll just come back to little Chester and how I find myself using the B insights. My point here is to show that these behavioural patterns are relatable to real life situations, so I’m not trying to push my views on parenting. But if I want to encourage him to eat his dinner or brush his teeth, I could take a traditional economic approach and describe all the benefits of, or the risks of brushing his teeth. But they’re really far in the future. 

I find he responds best if I say well, look, all your friends brush their teeth, and even your favourite preschool teacher brushes her teeth. So that’s social norms. Or I make it really fun now, I have a little gamified phone app or I use some strawberry tasting children’s toothpaste. So that’s salience, that’s providing some benefits now rather than just describing the benefits in the future. 

I make it predictable, so I reassure him, I tell him what’s going to happen; pajamas first, story second, brush teeth third, bed fourth. You can probably tell I’ve said a few times, but this helps create his meaning and makes sense of new situations and gives him comfort about what’s going to come next, so a bit like that operational transparency. 

Then I make it easy for him, I bring the toothbrush to him, it’s preloaded with toothpaste, so I don’t expect him to volunteer and run off and grab the toothpaste, I’ll make it easy for him to follow through with his good intentions. So perhaps some of the parents on the webinar have used these strategies and they recognize these patterns. 

Intuitively we know this kind of stuff works but then it seems when we turn our mind to insurance processes, we are more comfortable reverting to our traditional economic assumptions. When we’re always surprised when the customer doesn’t behave as we expect. 

If I was to sum up my summary would be the insights from behavioural science offer this alternative approach. So there’s still room for traditional behavioural change, but they help us design insurance processes for how customers actually behave rather than how we think they should behave. So now I’ll open it up for questions. 

Ben Telfer: 

Thank you, Nick. Thank you for sharing some great insights and some really interesting and practical examples and case studies about how behavioural economics is currently being applied. We do have time for a few questions, everyone in the audience please send any questions you have in. If we don’t get a chance to ask them to Nick on this webinar, we can definitely follow up with you and with him afterwards. 

Nick, there is one question that we received at the ICMIF Conference that I wanted to start with and ask you, and it’s about what courses or research would you recommend to upskill in the area of behavioural economics? 

Nick Mingo: 

Yeah, sure. So you definitely don’t need to get a master’s degree in behavioural science to apply these principles. So I’ve probably prepared three areas that other people can look to increase their knowledge or application of behavioural science. So the first is reading some various books. Some of you who are interested in the topic may have already read these. Even just the authors, noting down the authors, and each of these have written two or three books themselves, so these are big names in behavioural science. 

The second one would be to join an online community, behaviouraleconomics.com have lots of great case studies. So every year they release a huge PDF of all the different case studies of applying behavioural economics in the real world. So that’s a great one to join their newsletter. Then lastly to do. This course has actually just opened up, it’s free, and through the edx.org platform, it’s written by the University of Toronto. So six weeks you can self-paced learning, really focuses on applying all those insights. So the books give great stories about experiments and this is how to apply it as a practitioner to your own behavioural pain points. 

Ben Telfer: 

Thanks for those suggestions, Nick. Another question, where would you suggest we could find the quickest wins to apply behavioural economics for our customer processes? 

Nick Mingo: 

I think two aspects jump out. So I mentioned customer friction. So I think finding in your process, your insurance processes, areas of customer friction where maybe you’re already making things more difficult for the customer than it needs to be. So are there opportunities to prepopulate forms or tell them that they need to have something ready before they start a process. So think of the stamp self-addressed envelope. If you want someone to send you back a form, then if you include something stamped, self-addressed envelope, you’ve already taken a lot of those hurdles and actions away from them. So any unnecessary distraction or hassle that you can remove to stop your customer following through their good intentions, is an easy way. And I know insurance processes are full of these hurdles and hassles. 

Digital processes are always really great for test and learn, because you can get lots of information, lots of data, and you can deploy multiple versions. So I find quickly you can improve a quote flow or an application online very quickly with your behavioural insights. 

Ben Telfer: 

Thank you, Nick. I think we’ve got time for one more question, and it’s quite relevant in the global pandemic that’s going on at the moment, and it’s whether you think that COVID-19 will change purchasing behaviour and how that will impact how customers purchase their insurance? 

Nick Mingo: 

Yeah, absolutely. I think COVID-19 is a fascinating challenge of behaviour change in itself, and probably a whole webinar about the behaviour change of social distancing and we see there are some countries or administrations taking a traditional approach of laws and fines, and then also supplementing that with more of the social norms and you see people approving of these new social norms and reinforcing them and calling out bad behaviour. 

I think for insurers, there’s going to be a heightened awareness of mortality and sickness. So in behavioural economics, we call that availability bias or the recency effect. People are more likely to think about life and health cover. And also insurance in general, so we have to think that it may not always be positive, people may have negative experiences of insurance at this time, and that will shape how they see insurance in the future. We’ve just done some quick research with customers last month, and 30% of particularly the younger people said they’re more likely to think about buying life insurance now because of COVID-19. 

But then only 55% said that they think insurers will pay out if they died because of COVID-19. So they still have that suspicion that insurers won’t be there for them when they need it. So that’s one area that we need to get ahead of that message. And then lastly, distribution. I think people may get used to new ways of buying, particularly online, and away from relying on face-to-face brokers. Again, social norms say well, if everyone else is doing this new way of purchasing, and I’m using it in other ways, that it’s not that powerful anymore, it’s just because I always used to do it this way that I should. There’ll be a new norm and that will very quickly shift towards online or alternative distribution. 

Ben Telfer: 

Thank you, Nick. Very uncertain times but it’s great that you were able to join us today and share your insights about what you’ve been applying to life and on life insurers over the previous five years. As I said before, if you have any further questions please do get in touch, we’ll be happy to send them on to Nick. A couple of people already asked about the slides and the recording. These will be made available to everybody that’s registered for this webinar and you’ll be sent that after we finish today. So finally, thank you everybody for joining, thank you Nick for your presentation, and hope everyone stays safe.  

 

The above text has been produced by machine transcription from the webinar recording. ICMIF has made every effort to ensure that transcriptions are as accurate as possible, however, in some cases some text may be incomplete or inaccurate due to inaudible passages or transcription errors. Listening to or watching the webinar recording will allow you to hear the full text as delivered during the webinar but this is available in English only. Our transcriptions are provided to enable members to select the language of their choosing using the dropdown menu above.

More information

If you would like more information on the topic or case studies presented above, please contact us. We are here to make tailored introductions to your fellow ICMIF members and we can also share other member-only resources with you based on your specific challenges and interests.

Scroll to Top