How to Successfully Implement a Telematics Insurance Product

In this episode of the Insurance Telematics Podcast, Yasmin speaks with Harald Trautsch, CEO of Dolphin Technologies, about the keys to successfully implementing a telematics product in the insurance industry.

They explore why insurers adopt telematics, how to set strategic goals, the importance of prioritization, and common mistakes in RFI/RFP processes. You’ll learn why a lean and consultative approach beats a complex rollout and why focusing on customer value should be the top priority.

Why Do Insurers Implement Telematics?

Harald explains that insurers implement telematics to improve specific performance indicators, such as:

Reducing claims frequency
Increasing customer lifetime value
Engaging new demographic groups (e.g., young drivers)

The core goal should always be strategic improvement—not just technology adoption.

Setting a Strategic Foundation

Implementing a telematics product should begin with defining clear goals and priorities. Harald compares this to building a house:
“You don’t start by picking plumbing materials. You first decide how many bedrooms you want and who the house is for.”
Key Takeaway: Insurers should define what they want to achieve and let the telematics experts handle the implementation details.

Keep It Lean and Focused

Rather than launching all features at once, Harald advises starting small and focusing on what supports your strategic goal:

Target a specific demographic
Build a clean customer experience
Avoid trying to please everyone

“If you try to please everybody, you end up pleasing nobody.”

Technology Should Follow Strategy

Too often, insurers focus heavily on technology in their RFPs. Harald emphasizes that the value proposition should come first:

80% of questions in RFPs are tech-related—this should be reversed
Features like GPS sample frequency can be vanity metrics
Focus on how telematics can create value for both customers and the business

Use RFIs and RFPs Correctly

Request for Information (RFI): Used to explore what’s available in the market
Request for Proposal (RFP): Should only be issued once strategic goals are clear

Skipping the RFI often leads to unclear RFPs filled with technical questions that miss the strategic mark.

Learn from Real-World Examples

Harald shares insights from over 30 insurers worldwide:

Demographics vary—solutions must be tailored
Use pilot programs like Dolphin’s Move pilot to experiment quickly
Consider behavioral economics, customer incentives, and cultural context

Avoid Vanity Metrics

Examples of misleading metrics:

High GPS sample frequency drains battery and worsens UX
Fast event detection is useless without marketing automation

Final Advice for Insurers

Keep it simple: Ensure the product and message are clear
Focus on the user: Prioritize customer experience over features
Be brave: Offer fewer features but deliver them well
Iterate constantly: Launch, measure, adapt

“It’s never a finished product—it’s a continuous process.”

Episode Recap

Start with strategy: Set clear goals and priorities
Keep the product lean: Launch with minimal features and scale based on feedback
Avoid tech overload: Don’t focus on irrelevant metrics
RFI before RFP: Know what you want before inviting proposals
Work with pros: Bring in experts who know the landscape

Yasmin
Hi everyone and welcome back to the Insurance Telematics Podcast. Today we’re diving into a topic which insurers often struggle with, which is how to actually implement a telematics product successfully. And I’m joined again with Harald, the CEO and founder of Dolphin Technologies. How are you, Harald?

Harald
I’m fine. Thank you very much. How are you?

Yasmin
I’m good, thanks. Can I already ask you some questions?

Harald
Sure, sure.

Yasmin
So, why do insurers even implement telematics?

Harald
So, they normally want to improve something. They want to improve one ratio. So they want to reduce claims frequency, they want to increase customer lifetime, they want to engage a new demographic group. So, this is the underlying, or it should be the underlying desire when you start thinking about implementing a telematics product.

Yasmin
So let’s talk implementation. What does the ideal process look like?

Harald
Well, I see it like when you’re building a house. When you’re building a house, you need to address what you really want. You need to understand what is your family situation. So you have kids. Do you have pets? Do you want a balcony, a terrace, a garden? Do you want a pool? How many bedrooms? How many bathrooms?

You need to describe what your ideal house or apartment looks like. But you’re not going into the details of technology. You’re not talking about material of plumbing. You’re not talking about material of construction. You’re basically understanding who’s the owner of the problem or the desire, and let the professionals do the solution.

This is my approach. You should be able to formulate what you want, what you’re trying to achieve, what your strategic goals are, and not go into the technical details.

Yasmin
So it’s about setting clear strategy with goals and priorities which then the insurer can share with the telematics provider.

Harald
Absolutely.

Yasmin
Okay, let’s move on. I have a question about, I can imagine that insurers when they see all the possible options you get to do with telematics, they get quite excited and they want to implement all the possible features because at the end of the day, there are really amazing features which you can add with telematics. Is this something you would recommend doing? For example, implementing all at once or rather keeping it lean and slowly scaling the product over time?

Harald
I’m a big fan of setting priorities. I’m a big fan of creating a clean customer experience. And what you should do in my opinion is that you focus on your strategic goal and you implement exactly what is needed to achieve that goal.

If you try to please everybody, you’re basically pleasing nobody. This is really, really important. You cannot address everybody with a telematics product. You will have customers who don’t like it. You will have customers who don’t appreciate it. You will have customers who don’t understand it, but that’s okay. You need to build a group of fans and it’s totally fine if this is 10, 15, 20, 30% of your portfolio, because I truly believe if you try to please 100%, you’re pleasing nobody.

Yasmin
I have a follow-up question regarding more the technological aspect of things. As you mentioned earlier, insurers do tend to ask a lot of tech related questions. Should this be a very important part of implementing the telematics solution? Or should this rather be a side of the other important factors to consider?

Harald
When we participate in tenders and we look at the tenders, we mostly discover that about 80% of the questions are technology related and 20% are really value related. We believe it should be the opposite because the technology part is relatively, not simple, but you can outsource it to your telematics provider.

I truly believe that the focus should be on how to create value for the customers and in the long run, how to create value for the company.

I give you an example. I want to address a new demographic. Let’s say, I want to address young drivers. And I’m saying this because many telematics programs aim for that demographic. I want to select good drivers. I don’t want the reckless young drivers. I want the good ones. And there are a lot of them. About 70-80% of young drivers are actually very good drivers, and they’re very careful. Then obviously, you have these outliers that create a higher claims frequency in this demographic.

So, what you could do is a try-before-you-buy product. Show me how good a driver you are. Or you could go for a campaign where it’s actually cool to be a safe driver. Understand that the clear strategic goal is I want to address young drivers. This is my benefit with a low frequency as an insurance company. On the other side, I’m giving a benefit to young drivers by saying, we believe you’re a good driver and you deserve a good premium. So we give you the option to actually prove that you’re a good driver.

Have a clear story that you can tell and then everything works out.

Yasmin
So have a clear strategy, with priorities, with goals, and then we can help you do that since we have the technology.

Harald
Absolutely. Absolutely. Let us apply the tools based on your strategic goals.

Yasmin
And from your experience, is it rather an iterative and consultative process implementing a telematics solution or is it more a straightforward process with insurers?

Harald
It’s highly iterative. There is this strange situation that I always face when selling technology. Think of the following: you are hiring a new person for a new position. It’s a high potential. It’s a great person. They have great enthusiasm, etc. You’re not expecting this person to work perfectly from day zero, right?

Yasmin
Yeah.

Harald
You need to help the person become part of the culture of the company, part of the team, explain the tools, processes, get to know the customers, the products and services, etc. So there is an iterative process.

In technology, it should be the same way. We can apply knowledge from 30+ insurance companies we’re working with. But demographics are different. As an insurance

company, you’re different. You have your specific tone of voice. You’re addressing a specific demographic. You’re in a specific geographic region, in a specific cultural environment. So there are some things we need to fine-tune along the way. There must not be the expectation that everything works perfect from day zero.

Yasmin
Yeah. Of course not.

Harald
So this is important.

Yasmin
So I want to shift to the topic of request for proposals and request for information. Can you explain the difference between both to the audience and then dive into the common mistakes insurers make when creating a request for information?

Harald
A request for information (RFI) is to gather information, understand what’s possible, and see what the market offers. That’s totally fine. Every potential provider can communicate what they have and showcase their portfolio.

It becomes tricky when we skip the RFI and go straight to the request for proposal (RFP). The RFP is sometimes a mix of both. If you haven’t defined your strategic goals or the technology you want to use, then you get proposals with different technologies, and they’re hard to compare. For me, an RFP should compare similar offerings—meaning, I already know what I want. The RFI is for investigating how to solve the problem.

Even before an RFI, you should know what you want to achieve and communicate this to providers. Often in RFPs, most questions are technical, and only a fraction is about customer experience, retention, or added value. I strongly recommend doing this differently.

I want to refer back to episode five on our Move pilot where we give insurers the opportunity to test telematics without a complex internal process. You can build your own telematics pilot in minutes and focus on what really matters.

Yasmin
So insurers have the opportunity with the Move pilot to test out the tech aspect and get all their questions answered, so they can focus on strategic goals and added value.

Harald
Absolutely. In many cases, insurance companies hire consultants for such processes.

Yasmin
Yeah.

Harald
What consultants tend to do is quantify results. For example, we see in nearly every RFP the question: what is the sample frequency of GPS points?

Yasmin
I don’t know. So…

Harald
That means: how often do you ask the smartphone for geographic location? How precise is the path you can build for the customer?

It seems that a higher sample frequency means better data. But it’s not true. It’s a vanity metric. The more location points you get, the more you drain the user’s battery—which is a huge customer experience issue. In a straight road scenario, I don’t need one point per second. I can take one every 10 or 20 seconds because direction and speed didn’t change.

Smartphones aggregate multiple sources for location—not just GPS. So consultants may push technical questions that don’t make much sense in reality.

Yasmin
And the best way to not fall into this trap is to set your priorities, focus on customer experience, and have a clear strategy with goals.

Harald
You’re correct. Bring someone on board with deep technical knowledge. Telematics sounds easy—collecting points and interpreting driving behavior—but it requires expertise in smartphone tech, IT, data science, behavioral economics, insurance, radio signals like Bluetooth or Wi-Fi. It’s holistic.

If you don’t know what to ask, you’re missing the point. For example, another provider told us about a client demanding fast detection of driving behavior. They focused on milliseconds but didn’t have marketing automation. Yet, 90% of support cases are about revoked app permissions.

So if you’re focused on vanity metrics and miss the key issues, you’re off-track. Think customer experience, behavioral economics, and incentives.

Yasmin
Thank you for sharing these examples. Are there any final tips you’d like to share before we recap?

Harald
Keeping it simple is really important. Make sure the customer clearly understands the proposition. Don’t mix messages.

For instance, if you build a product for young driver safety, and the message is “we care about your child’s safety,” then you should aim that message at parents, because they usually pay for insurance. Don’t dilute that message by adding gamified features that contradict the tone.

Prioritize wisely. If you prioritize everything, you’re prioritizing nothing. You can’t please everyone—if you try to please everybody, you please nobody. So be courageous, have an edge, and build a great product for your demographic.

Yasmin
So it’s always better to keep it lean and scalable and see telematics as an iterative and consultative process rather than implementing everything at once without clear strategy.

Harald
Absolutely. Do two or three things right instead of ten things mediocre. Have the courage to not offer everything at once. Be evolutionary—add features, see how customers react. Use A/B testing, use marketing automation, use data. Telematics gives you insight into how users use your app. Learn, adapt, remove what doesn’t work.

The app is only part of the story. There’s communication, value proposition, marketing automation. It’s never a finished product—it’s a continuous process.

Yasmin
Thanks for sharing this last point. Can we recap the main points?

Harald
Absolutely. You need to help me.

Yasmin
Okay, here’s what I’ve learned: It’s important to set a clear strategy, goals, and priorities before launching an RFI or RFP. It’s better to keep the product lean and scalable than to launch all features at once. Want to add more?

Harald
Yes. Get a pro on board. Get someone skilled to help you.

Yasmin
And ask the right questions.

Harald
Yes, and let them ask the right questions. Not everything that’s quantifiable adds value. Sometimes it’s the opposite. Work with professionals and let them help you. That’s the key point.

Yasmin
Great. Thank you so much for watching this episode. Please subscribe and leave feedback.