Connected insurance: best practices from DriveQuant

This second interview is brought to you in partnership with DriveQuant, a company operating its telematics and connected insurance across Europe.

I'm thrilled to have the opportunity to speak again with its CEO, Philippe Moulin, who will enlighten us today on DriveQuant's offering, its technological differentiation, and the ROI achieved by its insurer clients in Europe.

In the previous episode (read more) we delved into what telematics is and why we should specifically talk about connected insurance. We also discussed the latest developments in the field and how this technology can benefit auto insurers who adopt it.

You can check DriveQuant's LinkedIn profile to learn more about their solution.

_

For French ???? speakers, you can listen to that discussion:

• On Spotify or Apple Podcast
• On Youtube

And, listen again to the first episode on podcast platforms.

_

Introduction

Florian Graillot - Hi Philippe! Thank you very much for being with us today, and before we start, could you briefly introduce yourself for those who may not already know you? The idea is to get an overview of your background and your current role at DriveQuant to shed light on what you'll be sharing with us later on the topic of the day, namely telematics.

Philippe Moulin - Hello Florian, my background is very technical. I studied engineering at the école des Mines and have a PhD in applied mathematics. I worked for quite some time in the automotive sector. I started out a bit with car manufacturers, a bit with technology-oriented suppliers, and then spent quite a long time at a research center called IFP Energies Nouvelles, where I was constantly involved in the design and development of algorithms for the automotive world.

While at IFP, in 2017, I founded the company DriveQuant based on technologies we had developed while looking into topics related to connected vehicles. We believed there were opportunities around leveraging vehicle data.

So we created DriveQuant around a driving data processing platform in 2017. And since then, I've been leading the company.

DriveQuant's value proposition

Thank you, Philippe, for that overview that sets the stage nicely. And actually, I'm linking it back to our first episode of our conversation during which we discussed several technical solutions for generating and retrieving data, knowing that this data is obviously the basis of any telematics or connected insurance offering. On this point precisely, can you tell us more about the technical solution or solutions used by DriveQuant? Moreover, it may have evolved over time?

We started by setting up the backend. We initially built ourselves around a driving data processing platform that was agnostic regarding the type of collection. Even today, we have clients who just use our API to process data coming from their own telematics devices.

It's really on the data processing algorithms that we have intellectual property. It's our expertise.

We also worked on a mobile application. Very quickly, we realized that the smartphone was an excellent means of collecting driving data because today, we have a smartphone in every vehicle. It's a way to easily deploy a solution and to make new developments, thus having an evolving solution. The operating systems are very powerful, the devices are very powerful. They contain sensors of very good quality, very rich.

And it's obviously a device that is close to the person. This is very important for prevention, for interacting with users. This is one of the added values of connected insurance programs.

So, we quickly developed tools for our clients to collect smartphone data. We provide our clients with a white-label mobile application. For all clients who do not have mobile applications, this allows them to launch projects without the need to heavily invest in the development of new mobile applications. Otherwise, we provide an SDK for clients who already have a mobile application. And there it makes more sense for ambitious projects that aim for large volumes from the start. These projects justify the development of a dedicated application, which may not necessarily be ours, but we provide the SDK that allows them to access all DriveQuant services related to telematics.

DriveQuant's customer base

It's very clear for this data part. And very interesting, by the way, the backend aspect you mentioned, which allows agnostic processing of data sources. This obviously resonates with the first episode of our exchange, during which we discussed the emergence of connected vehicles, for example, and more broadly, the issue of data in connected insurance. Also interesting is the double possibility of either white-label or integration into an existing app. This naturally leads us to the question of added value. Technology being obviously only a means. To illustrate this, could you tell us a little about your clients? Who are they? Insurers, car manufacturers, startups, others?

We currently have around thirty active clients. We deploy our solution in a software as a service mode. It's a generic solution, and all our clients use the same solution. However, there is a possibility of customization, especially with the SDKs that can be integrated into an existing app.

We have some clients in the fleet management domain, fleet management solution providers who wanted to add telematics functions to their solution.

But our clients are mainly insurance players.

In France, we work with all types of actors. We have startups that are starting out, others that are a bit larger. We have some more traditional insurers. For example, we work with Covea. With them, we have a project called DriveMeUp, which aims to prevent distracted driving and is pushed by MAAF and GMF.

We have been working with the MAIF group for quite some time. With their subsidiary Altima, we developed a pay-per-minute insurance offer, and we extended it with a pay-per-kilometer insurance offer.

We also work with APRIL. We have a behavior-based insurance offer for young drivers.

We work with a small broker called Yeet, which offers behavior-based insurance for VTC drivers.

In short, it's quite rich and diversified.

Another client we really like is Roole. Their offer resembles the concept of a car club in Anglo-Saxon countries. So they provide services around vehicle usage and notably services related to insurance (for example, deductible buyback, 0-kilometer assistance, warranty extension). They use driving data for prevention, to animate their community of drivers, and to better understand them for more marketing purposes.

We now have strong development abroad.

In Italy, we work with Quixa. We do accident detection, and there is an app that encourages eco-driving.

We work in Denmark with an insurer called LB Forsikring, which is one of the market leaders, with whom we do prevention. They have an offer where the deductible depends on the collected data, therefore on the driver's behavior. It's not the insurance rates that are linked to behavior, but it's the deductible, which is interesting.

We also work with an insurer in Tunisia.

And we are going to launch a fairly significant project in South Africa with one of the market leaders.

Overall, there is no pattern among our customers, we really have projects on all types of connected insurance. Often there is a strong prevention aspect. And increasingly, there is a pricing aspect and the assistance part that is growing.

Thank you for these details, it gives a good overview of your client portfolio at DriveQuant. It's interesting to see different use cases as well. And the term connected insurance really makes sense when listening to you! Beyond the direct clients you just mentioned, I believe you also work with reinsurers or at least one reinsurer. Can you explain to us exactly what the model is with this type of actor?

We have established a partnership with the reinsurer SCOR with the aim of supporting rather small or medium-sized insurers worldwide. There are interests expressed and opportunities to launch connected insurance and telematics programs. These actors, small or medium-sized, need support both from a financial and technological point of view. So together, we offer a solution to these clients.

DriveQuant's RoI

Okay, I understand the benefit for end customers well! A benefit that is twofold in this case. Moreover, it brings us back to the question of added value itself. Beyond pricing, pure telematics, could you specify how DriveQuant's solutions impact the entire value chain of the insurer by relying precisely on the experience you have with your numerous clients?

Traditionally, in a connected insurance program, we think about the pricing aspect, and often, it's behavior-based insurance. However, connected insurance is broader than that.

Clearly, there is a prevention aspect that is often very present in programs. Once you collect data, it's important to exchange it with the user. Value must be added for the user. Otherwise, we fall back into patterns where the insured will suspect their insurer of tracking them, and we revert to rather negative thought patterns. Whereas prevention can be very positive, on the contrary.

The role of the insurer is also to help the insured reduce their risk. So, prevention is very important.

Obviously, there can be a pricing component. It can be fairly simple things, like pay-per-mile insurance. Clearly, the risk depends on the duration, it can also be the distance traveled.

We also have collision detection functions. This can be used either to provide assistance in the event of severe collisions, so there can be some automation of the call to the assistance service and then dispatching of emergency services. And in the case of less severe collisions, it helps facilitate or speed up claims management.

Truly, all the verticals of insurance can be involved. And what we see is that the programs with very convincing results are those where the insurer has truly transformed around these technologies and uses data in all its business areas.

We are really in the digitization of insurance, covering marketing, prevention, pricing, assistance, and claims management.

This is where the shift occurs between the term "telematics" that I used at the very beginning and the term you use, "connected insurance." It makes perfect sense when we indeed see the impact it can have on all professions and the entire value chain. In terms of impact precisely, you started mentioning the term "results." Do you have any figures to share with us, or perhaps more broadly: what are the objectives your clients pursue when they decide to work with DriveQuant?

There are plenty of ways to measure the impact of a connected project, notably because the projects vary. Often, we approach it from the prevention angle. In this case, the KPI we are able to measure in our projects is the reduction in alerts or dangerous maneuvers when we conduct prevention actions.

Typically, in this regard, regarding the number of sudden brakes or smartphone usage while driving, we see gains of around 30% when the program is well-executed and the insured are engaged in the program.

The results are quite significant; the impact is really important. There is then a challenge to maintain these impacts over time. When the programs are well-executed, there are significant impacts.

We have clients who have conducted studies on insureds' feelings about the programs and have measured the NPS. Every time it has been done, the results are very good. So, there is really a challenge for the insurer to have very frequent interactions with the insured, which is not usually the case.

Connected insurance offers the opportunity to have very frequent exchanges with the customer.

What we can measure in these cases are the ratings of the apps on the app stores. And generally, we are above four, which is very good for insurer apps.

So, these are programs that are appreciated by the insured. And for our clients, that's important.

If we move on to KPIs related to insurance, I think of the claims ratio to premium ratio. We recently had a client who shared their results with us: they have a ratio below 50%. These results are not obtained solely thanks to telematics technology, but on well-executed programs, we see that we have very satisfactory results!

It's interesting that you mention the sentiment of the end customer. Because I obviously think of a famous insurance joke that says "80% of customers think they drive better than average." In my opinion, and beyond the joke, it illustrates a barrier that is often mentioned when thinking about telematics or connected insurance. It's the gap between, on one hand, the perception of the customer who thinks, as we understood, they drive better than they actually do. Expecting therefore an optimization of their insurance pricing, and on the other hand, the reality which is not always so attractive. Do you get this feedback from your insurer clients or not at all? Maybe you have some best practices to share with us specifically to combat this perception bias?

We're really in the heart of what we develop. We have a data collection technology. What's important is how this data is shared and how it's used. The data must bring transparency.

We try to encourage all our clients to share data with the insured and to build programs that are transparent. What's important then is the insured's commitment to an improvement program.

To get back to the joke, I can also give an example of a manufacturer who conducted a survey on app users and realized that in fact, there was an inverse correlation: the more insureds thought they drove well, the lower their score! It was really the opposite of what it should have been, which is interesting.

Often in the prevention programs we launch, there's the question of assuming that the people who will join and participate in the program will be the good drivers. Yet the challenge and what we're trying to do is to improve the bad drivers. Ultimately, the good ones aren't really the key point of the program.

We also saw in a program - the one we were talking about earlier in Denmark where the deductible depends on behavior data - that it's rather the bad drivers who signed up. And we think it's those who have already paid a deductible who are interested in paying less the next time they have to do it.

There are always lessons to be learned, programs we launch. What's interesting, I think, is the subject of transparency, information sharing. In telematics, the basis of the solution is to measure. Because we can't improve what we don't measure! So the first step is to measure. And then, it's about trying to improve, develop a solution that engages drivers, encourages them to improve to reduce risk.

And on that note, in my opinion, we've developed solutions - including DriveQuant, but also other providers - that can be significantly improved. There's really a lot of untapped potential for improvement on these issues.

Thank you very much, Philippe, for all these insights. It's great to be able to understand concretely how at DriveQuant, you deploy telematics and connected insurance solutions. As we sometimes have a biased or limited view of these initiatives. That's what I understand from our exchanges on this opportunity, anyway. By the way, for those interested, here is the link to the first episode of our discussion. We focused our discussions then on the concept of telematics itself with a focus on data and technology. Today's episode was oriented towards DriveQuant's added value and concrete results.

Thanks again, Philippe, for your time!

Thank you very much, Florian.

_

Learn more about DriveQuant.