Author name: Author

This event will take place on 12th and 13th June in Amsterdam

DIA is well known for being Europe’s thought-leading insurance and insurtech event and has an exhilarating line-up:

• Best of the best innovative insurtechs
• Thought-leaders from inside & outside the industry
• Exciting side-events (kick-off party, ferry networking ride, iAnnet)

Please visit our stand or check us out in the DIA – app.

For more information and tickets go to ITC DIA Europe – Amsterdam 2024

This week our CTO and co-founder Dennie van den Biggelaar  was the keynote speaker as listed in this linkedin post on AI in insurance for 120 members of the Stichting Assurantie Registratie (SAR). It was a great turnout of advisors under the RMiA, RGA, and RPA recognition schemes.

The Stichting Assurantie Registratie (SAR) is a Dutch foundation focused on the certification and registration of insurance professionals. SAR aims to uphold high standards within the insurance industry by recognizing and accrediting professionals who meet specific educational and professional criteria. These certifications, such as RMiA (Registered Mortgage Advisor), RGA (Registered Insurance Advisor), and RPA (Registered Pension Advisor), ensure that members possess the necessary knowledge and skills to provide quality advice and services to clients. The foundation also organizes events, training, and continuous professional development opportunities for its members to stay updated with industry trends and regulations.

𝐌ö𝐜𝐡𝐭𝐞𝐧 𝐒𝐢𝐞 𝐞𝐫𝐟𝐚𝐡𝐫𝐞𝐧, 𝐰𝐢𝐞 𝐒𝐢𝐞 𝐈𝐡𝐫 𝐊𝐮𝐧𝐝𝐞𝐧𝐩𝐨𝐫𝐭𝐟𝐨𝐥𝐢𝐨 𝐦𝐢𝐭 𝐞𝐢𝐧𝐞𝐦 hashtag#𝐊𝐈-𝐚𝐮𝐭𝐨𝐦𝐚𝐭𝐢𝐬𝐢𝐞𝐫𝐭𝐞𝐧 𝐀𝐧𝐬𝐚𝐭𝐳 𝐨𝐩𝐭𝐢𝐦𝐢𝐞𝐫𝐞𝐧 𝐤ö𝐧𝐧𝐞𝐧?

Gerne sehen wir uns am kommenden Dienstag, den 28. Mai und Mittwoch, den 29. Mai, auf der Messe insureNXT in Köln an Standnummer C011 mit Dialog Group !

At the request of VVP, featured in  the latest issue  VVP, the leading platform for financial advisors in the Netherlands, our CTO and AI strategist, Dennie van den Biggelaar, explains how to apply specific AI and machine learning to ‘advice in practice.’ In various editions, the following topics will be highlighted:

• Starting with specific AI and ML
• Operationalizing in business processes
• Integrating into existing IT landscapes
• Measuring = learning: KPIs for ML
• Ethics, regulations, and society
• AI and ML: a glimpse into the near future

In this third edition, we answer the question: how do you integrate a trained algorithm with your existing IT landscape and tooling?

Integrating AI Software

The insurance sector is on the verge of a technological revolution. With the integration of AI decision engines, insurers can significantly improve customer service and achieve better business results. AI algorithms can predict churn, calculate customer lifetime value (CLV), and make recommendations for cross- and upsell, enabling advisors to make better-informed decisions. But how do you integrate these algorithms into your existing IT landscape? How do you ensure that your employees have these predictions and suggestions at the right time to work more easily and effectively? This article discusses several concrete technical tips for successfully integrating AI decision engines into insurance systems.

Define a Successful Integration

I firmly believe that IT issues must always serve a business goal. A successful integration always begins with the question: ‘When is this integration successful?’ Creating a user story can help, for example:

“As [digital marketer at underwriting agency X], I want to [know weekly which customers need an additional product Y], so that I can [set up a targeted automatic marketing campaign for this group] with the aim of [generating (new) leads weekly for my field advisors].”

This is a good starting point to present to technical experts what is expected of them. Usually, further questions follow:

• What specific information does the user want to see?
• How often should it be refreshed?
• How will we measure the success of these automatic campaigns?

By asking and answering these questions, the team naturally identifies the framework of a successful integration. This is not a one-man job: it is important that both business/users and technical experts are represented in this exercise!

Analyze Existing IT Landscape

A successful integration begins with a thorough analysis of the existing IT infrastructure. Many integration attempts fail due to a lack of understanding of current systems, leading to compatibility issues. With which existing IT systems, databases, and interfaces must the AI algorithm ‘collaborate’? What amounts of data need to be transferred? When and how quickly?

(Successful integration means creating synergy between backend and frontend systems.)

In practice, this means working and aligning with various IT partners of backend and frontend systems. Start this inventory early and include all (external) stakeholders in your plans. If you don’t have the time or resources for this yourself, appoint one of your IT partners to do this project management for you. After all, it is their expertise!

Clear, Scalable, and Agile

Unfortunately, I have often seen organizations with innovative plans whose IT landscape was too rigid. Therefore, design a modular and scalable architecture to allow for future expansions and changes, ensuring your organization remains agile. Nowadays, it is best practice to use microservices architectures, where each functionality runs as a separate service. This makes it easier to add, replace, or update new elements without overhauling the entire infrastructure.

Consistency and Quality

Data quality is crucial for the success of AI. Many AI systems perform poorly due to inconsistent, incomplete, or outdated data. Therefore, implement a data cleaning and preprocessing pipeline to ensure that all data sent to the AI decision engine is clean and up-to-date. Automated tools for data integration and validation can help ensure the reliability of AI outcomes. Use ETL (Extract, Transform, Load) processes to extract data from various sources, transform it into a uniform format, and load it into a central data repository. This ensures a streamlined data flow essential for successfully training and using AI models.

Testing, Validating, and Monitoring

Thorough testing and validation are essential to ensure that AI models function correctly within existing systems. Insufficient testing can lead to errors and unexpected problems after going live. Therefore, conduct extensive tests in a simulated environment that mimics the production environment. Validate the output of AI models with historical data and scenario analyses. Involve end-users in the testing phase to ensure that the models meet business requirements and user needs.

Use APIs

APIs (Application Programming Interfaces) are essential to connect AI decision engines with existing systems. Without standardized interfaces, communication between systems can be inefficient and problematic. By developing and implementing APIs that can receive and send data, integration becomes flexible and scalable. This ensures that the AI decision engine can robustly communicate with both back-office and front-office systems.

Security and Privacy

Data security is crucial, especially given the sensitive nature of insurance data. Insufficient security can lead to data breaches, resulting in loss of customer trust and privacy violations. Therefore, use only the data that is truly necessary and anonymize as much as possible. If you really need certain sensitive data, apply encryption. Ensure that all data transfers between systems and the AI decision engine run over an encrypted connection. Use access controls and audit logging to guarantee data security and compliance with regulations.

Conclusion

A thorough integration is a prerequisite for AI to land successfully in your organization. You must consider not only a scalable business case and the user but also think about agility, security, privacy, and quality. Therefore, you need a team with different competencies and must coordinate and collaborate with IT partners.

First, ensure that you or someone in your organization has a clear understanding of the business framework of a successful integration. Make this explicit so that you can convey it. Then appoint someone responsible for the realization and the associated project management. If you do not want or cannot free up resources for this, you can easily engage one of your trusted IT partners for this. You can then focus on your own core business!

“It’s easy to create a self-learning algorithm. What’s challenging is to create a self-learning organization.” – Satya Nadella, CEO of Microsoft

In short: devising, building, and validating a robust algorithm is only phase one of successfully implementing AI in practice. In the next edition, we will delve into how to integrate it with existing IT systems and workflows.

In this second installment of the series on strategic entrepreneurship published in VVP, the platform for financial advisors in the Netherlands, we offer ways for independent intermediaries to respond to this consolidation pressure.

Due to consolidation, large companies are increasingly encountering the limits of their growth: an abundance of customers, but a scarcity of staff. In our view, this problem can only be solved by smart use of data and AI.

Special thanks to Marcel van Dijk of MarshBerry – Europe for the inspiration for this article.

Intermediaries were once fragmented and nationally oriented, but this is now quickly shifting to a consolidated and international playing field. In this second part of the series on strategic entrepreneurship, we delve into this significant business trend and offer ways for intermediaries to respond to this consolidation pressure.
“Five Arrows, the investment arm of the Rothschild investment bank, acquires a stake in insurance company Voogd & Voogd Groep.” This 2017 press release headline was about one of the first internationally operating investors entering the Dutch intermediary market. The turnover of Voogd & Voogd (now Alpina) grew from 50 million in 2017 to over 200 million in 2022 through mergers and acquisitions. A recent example of international expansion in the Dutch market is the acquisition of VLC & Partners (formerly owned by De Goudse) by Howden Insurance, which operates in 50 (!) countries. Söderberg from Sweden is also a successful international player, now ranking among the top 5 largest intermediaries in the Netherlands.
What makes the intermediary market so attractive for consolidation? And why are investors willing to invest large sums here? Firstly, intermediaries typically have a loyal customer base with annual recurring revenues, making them relatively insensitive to economic cycles or crises, such as the COVID-19 pandemic. Profit margins are still generous compared to other industries. Additionally, aging within the intermediary sector plays a significant role, both among advisors and owners. The average age is over 50 years. Finally, many smaller intermediaries are hesitant about the necessary deep investments in digitalization and increasing regulations regarding transparency and data privacy. Larger, stronger merged companies have an advantage here, at least according to many buying and selling parties.
Currently, there are still more than 900,000 intermediaries active in Europe. This averages to one intermediary for every 600 inhabitants in Europe. There are significant differences: in the Netherlands, there is one intermediary for every 2,953 inhabitants, in the United Kingdom one for every 6,650, in Germany one for every 442, and in Italy one for every 250 inhabitants. These figures from BIPAR – the European association for intermediaries – confirm that the consolidation trend manifested much earlier in the UK than in the Netherlands, while countries like Germany and Italy lag behind. Consolidation literally means merging. The UK now has 4,000 intermediaries and Germany still has 45,000, while both markets are approximately equal in terms of premium volume. Unlike other countries, both the Netherlands and the UK already have strict regulations (such as a ban on life insurance commissions) and lead in digitalization. These are two other trends that will continue and that many consolidation parties want to anticipate to turn inefficiency into profitability.
Many international Private Equity funds (PE) are now active in the European intermediary market, such as Blackstone, Rothschild, HG Capital, and KKR. They all follow a recognizable buy-and-build strategy, taking a financial interest in a prominent regional intermediary and using it as a platform company to add smaller intermediaries. This ‘stringing beads’ almost naturally creates value, as large companies have higher multiples over revenue than small companies.
Stringing Beads: Acrisure illustrates the potential of this ‘stringing beads’ strategy. It was founded in 2005 by Greg Williams and Ricky Norris in Grand Rapids, Michigan. With a thoughtful growth strategy, the company quickly reached a turnover of 30 million dollars. This success triggered PE Genstar to step in, and with the capital, the company acquired hundreds of intermediaries in the US and various European countries. In the Netherlands, Raetsheren was added. Acrisure now generates annual revenue of 4 billion dollars. The company is expected to be valued at 20 billion dollars in an IPO. For comparison, the market value of Nationale Nederlanden is around 10 billion euros.
“In the heart of every great change lies an opportunity waiting to be discovered.” – John C. Maxwell
In this dynamic time of consolidation, Dutch intermediaries face both challenges and opportunities. As a Local Hero, they have the power to choose their own path amidst the global players. However, maintaining independence and entrepreneurship requires smart strategies and flexible business models. Here are some practical options:
1. Work on Strategic Positioning: Identify your office’s unique strengths. If these are still unclear, consider distinguishing yourself in a niche or specialization in the market and build a loyal customer base. Building a strong brand and business identity is crucial. Clear positioning can attract customers even amidst consolidation.
2. Accelerate Digital Transformation: Invest at least 6% to 8% of your revenue in digitalization to operate more efficiently. Look at what software providers offer and focus not only on the back office but especially on a better digital customer journey. This requires continuous attention to improving data quality. A strong online presence (website, social media) increases competitiveness, especially if you are active locally. Stay alert to innovative developments in the sector, such as the rise of AI.
3. Collaborate and Network: Seek strategic collaboration with fellow intermediaries and insurance partners to achieve economies of scale without full consolidation. Select service providers that offer products aligned with your strategy and are at the forefront of digitalization.
4. Invest in Talent Development and Team Spirit: Invest in staff development to build expertise, not only technically but also in emerging areas such as digitalization, risk management, and compliance. Consider how to turn your star players into a winning team. Start by establishing an inspiring ambition (mission, vision) for your office.
5. Strengthen Customer Relationships, Minimize Bleeders, Maximize Feeders: Focus on building sustainable customer relationships by providing excellent service and truly customer-oriented solutions. Satisfied customers are essential for organic growth. First, determine who your feeders and bleeders are. Feeders are profitable customer segments for your office. Bleeders cost you money because handling costs (visits, emails, phone calls, changes, claims handling) are higher than revenue from commission or service subscriptions.
6. Combine Different Business Models: The commission model seems less sustainable in the trend of increasing transparency. Ensure you are agile. In addition to the service subscription, there are alternatives such as:
• Hourly billing: The more specialized and unique, the higher the hourly rate.
• The platform model: Bringing supply and demand together (in line with your specialty) on a digital marketplace.
• The freemium model: A free basic product or service to attract customers, upgrading to paid options based on their needs.
• Embedded insurance: Seamlessly offering insurance as an integration into a product or sales process. Well-known examples: travel insurance at a travel agency, car insurance at a dealer, or warranty insurance with equipment purchases.
• Outsourcing: Outsourcing certain activities to specialized service providers so you can focus on your core activities. Think of outsourcing specialized pension advice, absence management, or your own IT services.
7. Manage Your Own Financial Planning: Financial stability offers more control over strategic decisions. Be prepared for economic fluctuations and strive for a solid financial foundation. Continuously benchmark and monitor the key KPIs of your office. Engage with M&A (merger & acquisition) experts to understand how they view the value development of your company and how you can optimize it to become an attractive merger or acquisition target in the future. A good understanding of the legal aspects is essential to safeguard your company’s interests.
Big isn’t always better, and sometimes a creative dwarf beats an industrial giant. An inspiring example from outside the insurance industry is Pixar Animation Studios, which in the early 1990s under the leadership of visionary Steve Jobs defied the established order of Disney with their groundbreaking computer-animated films like “Toy Story” and “Finding Nemo.” The highlight of the battle between this David and Goliath came in 2006 when Disney acquired Pixar for over $7 billion. This collaboration between seemingly unequal forces was more than a business transaction and led to creative synergy and extraordinary growth in the entertainment industry.
Jack Vos is a member of the Entrepreneurs Panel, former intermediary, and founder of the high-tech data science company Onesurance.

At the request of VVP, featured in  the latest issue  VVP, the leading platform for financial advisors in the Netherlands, our CTO and AI strategist, Dennie van den Biggelaar, explains how to apply specific AI and machine learning to ‘advice in practice.’ In various editions, the following topics will be highlighted:

• Starting with specific AI and ML
• Operationalizing in business processes
• Integrating into existing IT landscapes
• Measuring = learning: KPIs for ML
• Ethics, regulations, and society
• AI and ML: a glimpse into the near future

In this second edition, we answer the question: how do you operationalize a trained algorithm?

Challenges in Operationalizing AI in Business Processes

Imagine you and your data science team have designed a promising AI algorithm to predict cancellations, aiming for advisors to proactively address them. This process was discussed in part 1 of this series, ‘AI in Practice.’ The potential is there, but soon you realize that several complex hurdles must be overcome to implement it practically. What are these hurdles, and how can you overcome them?

Measurable Results Are Lacking

A clearly formulated goal precisely identifies what the AI algorithm should achieve and aligns with business objectives. The scope, on the other hand, directs the project by defining relevant data sources, budget, timelines, and expected outcomes. What are the steps to achieve this?

A data science project is generally an investment where:

1. It is unclear what it can deliver.
2. It is uncertain if your team can realize it.

Therefore, make the project as small and manageable as possible without losing its value and impact if it succeeds. Try to achieve results as quickly as possible to prove you are on the right track.

If you don’t achieve those results, evaluate and adjust with the team. If you do achieve them? First goal reached! Then, create a compelling story and present it to your business stakeholders, discussing how to scale it within your organization.

Questions About Consistent Data Quality

A common stumbling block is the quality and consistent provision of up-to-date data. Inconsistencies and missing values can jeopardize the accuracy of the AI model. The solution? A thorough exploration of which data is always Accurate, Available, and Consistent (the data ABC).

If essential data does not meet these criteria to achieve your goal, apply extensive data cleaning, such as handling missing values, extreme outliers, and incorrectly entered data. Then you will need to structurally ensure these cleaning steps in a data transformation pipeline and associated process so that you can add this data to your foundation for a reliable operational model.

Insufficient Trust in the AI Model

Insufficient understanding and trust in ML models form a barrier to acceptance among non-technical users. If you don’t pay enough attention to this, distrust and resistance will arise. A solution is selecting transparent models with good explainability and smart methods where complexity is translated into an understandable concept. Visualization and clear (process) documentation increase trust, pushing the “black box” objection into the background.

As with any change, it is important to carefully involve your colleagues in this process. Give them enough time to ask questions and get used to this new technology and its possibilities. Realize that their questions and feedback are essential inputs to make the intended application successful in practice.

Concerns About Security, Privacy, and Ethics

It goes without saying that the security and privacy of (customer) data are prerequisites to start with. Fortunately, much new legislation has been implemented in the past five years, and organizations are also applying it practically and structurally.

Trust is not only an issue of legislation and technology. Ethical objections can also be expected from various angles:

• Are we sure the algorithm is fair?
• And what does that mean?
• Are certain groups worse off with an algorithm?
• Do we find that ethically responsible?
• How do I prevent my algorithm from discriminating?

Fortunately, the Dutch Association of Insurers has established several guidelines that you can incorporate into your algorithm and approach. Want to make sure you don’t overlook anything? Assign one person responsible for this.

The Feedback Cycle Is Missing

Listening to user experiences and using this feedback creates a dynamic iterative cycle, allowing the model to evolve according to business requirements. A structured feedback mechanism is crucial for the self-learning ability of the AI model. How you set this up correctly differs per AI application.

In the specific case of ‘preventing cancellations,’ have advisors, for example, record what they did with the prediction: called the customer, visited them, or did nothing. This way, you can measure over time what effect it has on cancellations.

Insufficient Monitoring

The motto should be: “keep the algorithm on the leash.” You don’t want ‘hallucinations’ or unexpected performance degradation, for example, during a trend break. This means that a careful monitoring and warning system must be in place to track model performance. A sustainable application requires detailed documentation of parameters and used data so that the model remains transparent and reproducible.

The Model Is Not Scalable

An algorithm must be ‘by design’ part of a system with scalability in mind. Safe cloud solutions and scalable infrastructure such as MLOps technology (the ML variant of DevOps) are generally necessary. Consider growth projections and ensure a sufficiently flexible system that adapts to evolving business requirements. Making the right choices for integration with the IT landscape is essential (e.g., real-time or batch processing). More on this in the next edition.

Last But Not Least: Insufficient Involvement

According to innovation professor Henk Volberda, the success of innovation is only 25% technical and 75% dependent on human adoption. Successful adoption starts with “CEO sponsorship,” as water flows from top to bottom. Leadership must ensure sufficient training, communication, and support when deploying an AI model. Invest enough time and energy to make this new technology part of your organization, from strategy to operation. Because that’s where the real return on investment lies: the successful collaboration between human experts and AI technology.

“It’s easy to create a self-learning algorithm. What’s challenging is to create a self-learning organization.” – Satya Nadella, CEO of Microsoft

In summary: devising, building, and validating a robust algorithm is just phase one of successfully implementing AI in practice. In the next edition, we will delve into how to integrate it with existing IT systems and workflows.

In this interview with Odette Bakker (CEO) and Indra Frishert (CMO) of DAZURE, which was published in VVP, the platform for financial advisors in the Netherlands,VVP,   Odette and Indra state that “The brilliant minds at Onesurance are building something innovative for our medical process. Soon, many prospective insureds will have to answer significantly fewer medical questions because the developed algorithm can make predictions based on historical data. As a result, policies can be issued much faster”.

“For me, Onesurance epitomizes responsible and effective AI implementation, yielding optimal outcomes. Their boundless enthusiasm adds a layer of enjoyment to collaboration.”

Odette Bakker, CEO DAZURE

“When applying for life insurance, the insurer wants to know exactly how (un)healthy a person is to estimate the risk of death. This is done by medical specialists. The applicant must fill out a ‘health declaration.’ what an unfriendly word, actually… but anyway. If it shows that you have (had) something, you have to answer even more questions or undergo examinations. We know from experience that most people are not looking forward to this.

When applying  for life insurance grilling is no longer needed.

Dazure offers life insurance, and owners Odette Bakker and Indra Frishert have always viewed this medical process critically. They take signals from applicants seriously. People sometimes cancel their application upon hearing they need to undergo a medical examination. During the COVID period, this issue became more pressing due to shortages of medical personnel. Medical specialists were more urgently needed in hospitals.

DECLINE IN LIFE INSURANCE POLICIES

Nowadays, people expect that when they order something online, it will be delivered the next day. The medical process sometimes causes long delays. The decrease in the number of life insurance policies being taken out is partly due to this process. Financial advisors often recommend taking out life insurance, but they do not always actively mediate it. The lengthy process of issuing the policy can detract from their advisory process. ‘In the past,’ life insurance was taken out in 60 percent of cases when taking out a mortgage; now, it is only 16 percent. It is extremely important to have life insurance to be financially secure.

AI IMPLEMENTATION

Dazure has long had the ambition to make the medical acceptance process faster, more efficient, enjoyable, and easier. We have now developed an intelligent system with Onesurance based on AI (Artificial Intelligence). And yes, of course, with a ‘human in the loop’ and an ethical framework. This system allows us to safely and more accurately predict which applicants can go through a shorter process. This group does not have to bother medical specialists or exhaust themselves with lengthy questionnaires.

HOW DOES IT WORK?

We have built a fully transparent algorithm that can make reliable and highly accurate predictions based on the completed health questions using predictive analytics. This algorithm is trained on historical data.

IMPRESSIVE RESULTS

The algorithm immediately correctly predicted half of the medical processes with an error margin of 0.1 percent. More than half of the applicants could be accepted right away. The processing time for the rest was reduced from 29-71 days to 7 days! It is estimated that the number of cancellations will be at least halved.

By making the process better, more efficient, and faster, we expect to meet modern expectations and for more people to find their way back to life insurance.

At the request of VVP, featured in  the latest issue  VVP, the leading platform for financial advisors in the Netherlands, our CTO and AI strategist, Dennie van den Biggelaar, explains how to apply specific AI and machine learning to ‘advice in practice.’ In various editions, the following topics will be highlighted:

• Starting with specific AI and ML
• Operationalizing in business processes
• Integrating into existing IT landscapes
• Measuring = learning: KPIs for ML
• Ethics, regulations, and society
• AI and ML: a glimpse into the near future

Naturally, we start in this first edition with the basics: what is it and how do you get started? If you have any questions following this article, feel free to contact Dennie (dennie@onesurance.nl).

AI vs. Machine Learning (ML) AI is a machine or software that performs tasks traditionally requiring human intelligence. Machine learning (ML) is a specific subset of AI, allowing a machine or software to learn from historical predictions or actions. The most well-known and discussed example of ML software is ChatGPT, designed to generate meaningful text for users. However, there are countless other issues where machine learning can assist us. There is often (still) no ready-made solution like ChatGPT that you can use directly. To build such a usable AI solution, you must bring the right competencies together at the right time. It is the AI strategist’s task to work with a multidisciplinary team of business experts, ML engineers, data engineers, and data scientists to determine what you want to predict, how (accurately) it should be done, which techniques to use, and finally how everything is operationalized and secured to achieve the desired results.

Example: Predicting Cancellations As a firm, you want to ensure that the right customers get the right attention from your advisors at the right time, minimizing cancellations. Ideally, you would know which customers are likely to cancel. But how do you translate this to the team? Often, a customer cancels a single policy, which is usually just a change and something you don’t want to contaminate your ML model with. Suppose a customer cancels all policies within the main liability branch but not the others (yet). Is this a customer about to leave? And what if they cancel everything within the main fire branch but still have legal aid and term life insurance? Have any policies been internally transferred? What is the actual cancellation rate? These are all things you want to determine before putting an ML engineering team to work. Additionally, you must consider your forecast horizon: how far ahead do you want to predict? Do you want to know which customers will cancel in the next 1, 3, 6, or 12 months? This may seem like a detail, but under the hood, it means you will train a completely different ML model.

Finding Patterns Once you have clearly defined what you want to predict, it’s time to see if your data is sufficiently Accurate, Available, and Consistent (the ‘data ABC’). The main reason customers cancel often boils down to receiving too little attention. The question is, of course, from whom, when, and why there was ‘too little attention.’ This information is not in your data warehouse and must be constructed through feature engineering. What features (characteristics) significantly affect the likelihood of cancellation? This is an analytical and creative process where the knowledge and experience of insurance experts and data scientists come together. Once a solid initial table with features is shaped, you can finally start with machine learning. Experience shows that predicting cancellations is best modeled with classification or survival analysis. There are hundreds of different ML techniques theoretically suitable for this. In your choice, it’s important to consider: to what extent does the algorithm need to be explainable, how complex can the patterns be, and how much data is ABC?

Validating Patterns After the ‘machine’ is set to work to find patterns and make predictions, there’s always an exciting moment… how accurate are the different models? The ML engineer has an extensive toolbox for this. First, they keep part of the data separate to test and validate a trained model. This ensures the robustness of the discovered patterns and prevents a model from making inaccurate predictions in the ‘real world.’ Next, they look at the false positives and false negatives and their costs. For example, a false positive prediction that someone will cancel next month isn’t too bad. The advisor calls the customer and concludes that nothing is wrong: it only costs 15 minutes of their time. If the algorithm incorrectly predicts that someone will remain loyal (false negative), this is much more costly: you lose a customer. Based on, among other things, precision, recall, and AUC scores, the best ML model is determined. Additionally, it’s possible to adjust algorithms to be stricter or more lenient, so they better fit the intended business process. This is called parameter tuning, and an experienced ML engineer knows how to do this responsibly.

Making It Usable Next, you integrate the algorithm into operational processes. How can data be transferred back and forth safely and efficiently? How can the advisor easily use the prediction? This is the work of data and software engineers. Finally, you also want the advisor to provide feedback on the algorithm’s quality so that it learns from the user. The algorithm becomes smarter and more effective the more it is used. That is the real ‘AI’ component, but more on this in the next edition!

AI is not always ML. For example, the algorithm Deep Blue (which defeated chess grandmaster Garry Kasparov) in 1997 is AI but not ML. ML is always AI.

Three Sentences About Dennie Himself Dennie is an econometrician with 12 years of experience in designing, building, and implementing machine learning solutions in practice. As co-founder and CTO of Onesurance, he is responsible for developing AI solutions and successfully operationalizing them for clients in the insurance sector.

The NVGA is the most important community in the Netherlands for independent insurance brokers.

The selection committee consisted of Levent Türkmen (general director of SUREbusiness), Marijn Moerman (CEO of Alicia, winner of the 2022 innovation award), Annet van den Berg (editor of AMweb_nl), and Caro Sala (communication advisor at NVGA). They reviewed the sixteen entries received this year.

Our nomination is for the AI Engine. This is a modular AI-driven decision engine capable of supporting advisors by extracting data from back-office systems, continuously performing analyses with complex algorithms, and then delivering targeted predictions in the front-end systems in use.

The NVGA and AM aim to boost innovation in the underwriting agency sector with the NVGA AM Innovation Award.

A  personal story by Heidy Bouwmans-Bierman in this article, which was published in VVP, the platform for financial advisors in the Netherlands, on how the personal team at Rivez and Zuiderhuis Assurantiën collaborates with Onesurance’s  TopDefend application.

“The way data can be used to predict, for example, the churn probability of a relationship fascinates me. It provides us with a great tool to serve our customers. We can also predict the next best policy for a relationship based on data.”

Heidy Bouwmans – Manager personal insurances – Rivez Zuiderhuis

What does her colleague say:

“The AI Assistant delivers on its promises, generating enthusiasm among employees as it seamlessly integrates complex technology onto the shop floor. This translates to reduced churn, increased sales, and heightened satisfaction among both customers and staff”.

Michael Dubelaar – COO Rivez Zuiderhuis

Ensuring that each customer receives timely and tailored attention is crucial.

Text by Willem Vreeswijk

Heidy Bouwmans, Manager Personal Insurances and Advice at Rivez-Zuiderhuis, which has 165 employees, says this. The insurance company, generating over 20 million euros in revenue, was formed in 2022 from a merger between Rivez Assurantiën & Risicobeheer and Zuiderhuis Assurantiën. The company is part of the Söderberg & Partners network and has branches in Helmond, Veghel, Deurne, Schaijk, Tegelen, and Venlo. Rivez-Zuiderhuis is active in insurance, risk management, absence issues, mortgages, and tailored financial advice for both individuals and businesses. Additionally, there is a specialization in real estate/real estate corporations and trade associations.

“Better Customer Service Where Most Attention Goes to the Right Customer”

Heidy has been active in her current role for four years but has a long career in the financial sector. “I have mostly worked in the business segment. About 35 years ago, I started at Aegon. After a training period, I was seconded to one of their advisors. Then I enjoyed working in the Motor Vehicle Companies department in Leeuwarden. After the business part of Aegon moved to The Hague, I worked for a short period in the private department in Leeuwarden. Due to a move, I ended up at Quintes, where I worked for twelve years as a Business Relationship Manager and also co-approved business delegated policies and claims with a colleague. Then I worked as Head of Internal Affairs at a smaller office. Here, too, I dealt with claims and business relationships. After holding this position for about ten years, I ended up at Rivez-Zuiderhuis in my current role. I have always gained a lot of energy from providing relationships with the best possible insurance solution and advice. I noticed that I spent a lot of time answering questions from employees and taking on a more coaching role. I enjoyed this so much that about fifteen years ago, I switched to a more leadership role. Despite the challenges that come with this, I still derive a lot of satisfaction from it.”

BETTER CUSTOMER SERVICE

“For a long time, AI and the use of data were a ‘far-off show’ for me, and I associated it with robot technology. Until I read how AI is used in hospitals, for example, to extract information from patient data in delivering COVID-19 diagnoses and cancer research. Through my children, I came into contact with an app like ChatGPT, and a whole new world opened up for me. The use of AI is so much broader and more integrated into daily life than many people realize.”

“Last year, we came into contact with Onesurance, and my interest in AI was truly piqued. The way data can be used to predict, for example, the churn probability of a relationship fascinates me. It provides us with a great tool to serve our customers. The developments are rapid. For example, they can also predict the next best policy for a relationship based on data.”

“What we aim to achieve with the use of data is better customer service where most attention goes to the right customer. The predictions we use create a sense of relevance among employees. They also feel that they are maintaining the right customer, which is very motivating. Especially because there has been a shift in the approach to private customers in recent years. The days of the advisor coming over for coffee in the evening to review the package are long behind us. Providing good customer approach while meeting the duty of care remains a challenge, especially with the current struggles in the labor market.”

LEARNING POINTS

“We started small with a few employees who provided a lot of feedback, allowing continuous adjustments and the program to begin working more optimally. Gradually, the program is being rolled out across the entire private department. Employees handle a number of leads daily.”

What are the learning points for fellow advisors? “An important point is making time and instructing employees. In the hustle and bustle of the day, you need to ensure there are enough opportunities to contact a number of relationships each day. Facilitating this sufficiently for our employees remains a challenge.”

Insurance Data and Onesurance proudly announce a strategic partnership in this linked post, combining their unique strengths to dramatically transform the use of data in the insurance market. This marks a new phase of progress in the industry.

As a leading expert in streamlining, integrating, and visualizing insurance data, Insurance Data has built an impressive track record in advanced Business Intelligence. Complementing this, the team of experienced AI engineers at Onesurance has developed an innovative AI Engine. This modular engine is specifically designed for insurers, underwriting agencies, and intermediaries, providing advanced churn predictions and highly accurate AI underwriting for business risks.

Jack Vos (CEO Onesurance): “This collaboration increases our impact in the market, as we serve the same select customer group.”

Lex De Bruijn (CEO Insurancedata: “This alliance not only offers greater visibility but also allows us to add direct value for our customers.”

The urgency to leverage data and AI for critical themes such as active customer management, efficient workflows, scalable growth, and compliance is driven by the consolidation of increasingly large portfolios and the growing scarcity of qualified personnel. With over 12 months of intensive collaboration behind the scenes, the data experts from both companies have built the AI Engine, enabling a rapid time-to-value for customers.

In the insurance world, timing is crucial. The AI Engine addresses this issue while maintaining accuracy and customer focus, keeping companies ahead of the competition. An example is the generic M&A module, which predicts portfolio valuations using algorithms.

To quickly meet the growing market demand, Frank Rensen (RGA) has been appointed as an experienced insurance expert. He is well-known in the market.

On December 18th, 2023, we were featured in this article in Het Financieele Dagblad with an extensive interview where we somewhat debunked the ‘AI hype’ and focused particularly on the ethical aspects of AI.

THE ®EVOLUTION OF AI IN THE INSURANCE SECTOR

Introduction: It is remarkable that many still think AI is limited to generative AI, such as Chat-GPT. However, specific AI has been used in the insurance sector for some time, particularly by major insurers for premium and risk optimization. The data science company Onesurance focuses on specific AI to support financial service providers in better serving their customers.

Dennie Van Den Biggelaar, econometrician and data scientist with 12 years of experience as an AI strategist, emphasizes that in the insurance sector, it is more about an ‘evolution of specific AI’ rather than a ‘revolution of generative AI’. This (r)evolution is driven by a growing need to serve customers in a scalable and personal manner. Due to consolidations, portfolios are increasing in size with a growing shortage of advisors. Insurance companies are thus reaching the limits of their growth. “This problem can only be solved by using data and AI more intelligently,” says Van Den Biggelaar, who, along with his partner Jack Vos, founded Onesurance. Vos, who is well-acquainted with the insurance world, explains that their AI modules are not intended to replace advisors and experts but to support them in their work. “In the world of insurance, mortgages, and pensions, it ultimately revolves around trust: it is and remains a people business.”

Onesurance consciously chooses to deploy specific AI to solve very specific problems for insurance companies. Based on historical data, reliable predictions are made for acceptance, risk of termination, customer lifetime value, or effective customer management, ensuring that the right customer receives the right attention at the right time. “Our AI modules are ‘by design’ explainable, transparent, and comply with the Ethical Data Frameworks of the Dutch Association of Insurers,” notes Van den Biggelaar, who is responsible for the technical development of the AI modules. Vos adds, “These are qualities typically required in the insurance industry, but which generative AI cannot yet provide.”

The potential to do more with data and AI is considerable. According to a study by the CapGemini Research Institute among 204 insurers, ‘data masters’ in the insurance sector achieve a revenue per FTE that is 175% higher and are 63% more profitable. This fact obviously appeals to parties that are constantly making acquisitions in their ambition to grow scalably.

At the same time, the dilemmas around privacy, bias, and liability remain significant. If AI is deployed without context and without the so-called ‘human in the loop,’ unintended errors can occur, leading to customer dissatisfaction, ethical issues, and a loss of trust in the insurance sector. “You must not make ethical and moral compromises when working with machine learning and AI,” says Van den Biggelaar. “It is crucial that AI experts feel responsible for steering this in the right direction. And we take our role in this very seriously.” Therefore, Onesurance collaborates intensively with Brush-AI, the first Dutch company dedicated to systematically managing the ethical component in AI.

That this responsibility is also taken seriously by the insurance sector is evident from the KOAT advisory committee, which started under the leadership of SIVI in 2019. SIVI develops and manages standards for digital business in the financial world. Jack Vos, who is himself on the committee, says: “KOAT stands for Quality Unmanned Advice and Transaction Applications. By ‘unmanned applications,’ we mean smart software capable of replacing tasks, not people. The increasing use of such automated applications in the financial sector, combined with new (European) regulations, makes quality control of unmanned applications increasingly important to safeguard customer interests.” The Financial Services Complaints Institute (Kifid) sees more and more complaints about digital services. The SIVI platform offers insurers, underwriting agents, and software companies a checklist and a knowledge base, among other things. It is not unthinkable that the checklist is a good first step towards a quality standard. This is important for the end consumer but also for the financial advisor themselves. After all, the financial service provider remains liable for the advice given, even if the unmanned application is developed by an external software supplier.

“A technology becomes powerful only when it is embedded in the ethical and human context.” Satya Nadella, CEO of Microsoft.

Fact Box: Since September 2022, Onesurance has been part of the Cronos Group, the largest ICT service provider in Belgium, with 10,000 employees, 500 of whom are based in the Netherlands. This has enabled Onesurance to scale quickly, as all complementary ICT experts are directly available within the Cronos units. Founders Jack Vos and Dennie van den Biggelaar have gained years of experience in successfully developing and implementing data science solutions at Building Blocks, which is part of the listed company CM.com. In 2019, they won the NVGA Innovation Award with the Data Driven Underwriting solution. Onesurance has offices in Amsterdam-IJburg, Breda, and Kontich (Belgium).

We concluded by emphasizing the importance of the Advisory Committee KOAT of SIVI, which includes representatives from the Dutch Association of Insurers, Adfiz, the Contact Group Automation Foundation, Leiden University, and Tilburg University.

KOAT stands for Quality Unmanned Advice Applications. This refers to smart software that can replace tasks, but not people. Insurance is and will remain a people business.

The rise of artificial intelligence (AI) seems to have reached the insurance sector through the ChatGPT hype, and the discussion about its effects is anything but black-and-white. The question is how we can find a balance between the positive and negative aspects of the rise of AI, as this balance is crucial in no other sector as much as it is in the insurance world.

AUTHOR – JACK VOS – ONESURANCE.NL

To begin with, the concept of artificial intelligence is certainly not new. The term itself was first used during the “Dartmouth Workshop” in 1956, where scientists gathered to discuss how machines could exhibit intelligent behavior. Since then, we can distinguish three main phases in the development of AI within the insurance landscape:

Rule-based systems (1960-1990) In the early years of AI, these systems only used manually inputted rules to make simple decisions based on specific inputs. For example, acceptance based on simple predefined rules. These systems were still too limited to make complex decisions.

Statistical modeling and data analysis (1990-2010) As computers became faster and analysis software became more intelligent, Machine Learning models were deployed to discover patterns and trends in large amounts of insurance data. This was particularly helpful in risk assessment and fraud detection.

Machine Learning and Predictive Analytics (2010-present) With the advent of more advanced Machine Learning techniques, such as neural networks and deep learning, it has become possible to perform even more complex analyses. This includes predicting customer behavior, determining rates based on individual characteristics, and detecting fraudulent activities with higher precision. AI is also used to improve customer service with chatbots, virtual assistants, and automated interactions.

And now, since November 2022, there is Chat-GPT, in which Microsoft wants to invest €10 billion. This is a Large Language Model (LLM) that predicts the next word in a text, mimicking human language. Among seasoned data experts who have worked for years on refining algorithms and understanding big data, the recent AI hype caused by ChatGPT has elicited a mixture of excitement and concern. Two tweets from prominent AI figures encapsulate the contrast between the positive and negative aspects.

The positive side: “It’s going to be glorious” Tweet 1 is from well-known venture capitalist Marc Andreessen, in January 2023: “We are just entering an AI-powered golden age of writing, art, music, software, and science. It’s going to be glorious. World historical.” The excitement around AI is certainly justified in the insurance sector, as it has introduced new possibilities that once seemed unthinkable. AI promises efficiency and accuracy at a new level. AI can analyze very large amounts of data and create insights into risk assessment, claim handling, or customer service that human assessors could never achieve. The benefits of AI are simply too attractive to ignore, which is why more and more decision-makers are putting AI on the agenda. It enables insurance companies to gain competitive advantages, improve customer experience, and reduce costs simultaneously. In short, a unique opportunity to lead as a ‘Data Master’ in a traditional industry subject to change.

The negative side: “a misleading impression of greatness” The contrast with tweet 2 is stark: “ChatGPT is incredibly limited, but good enough at some things to create a misleading impression of greatness. It’s a mistake to be relying on it for anything important right now. (…)” The author of this tweet is Sam Altman, the founder and CEO of OpenAI’s ChatGPT itself. Have you ever heard a CEO say about their own product that it is still good at creating misplaced greatness? The concern is justified because persuasiveness is indeed the main factor in allowing false information to do its devastating work. The persuasiveness of ChatGPT is overrated in this regard. This means that anyone professionally involved with the truth must be alert, as asking the right questions can produce highly credible and persuasive nonsense. According to Professor Terrence Sejnowski, author of The Deep Learning Revolution, language models also reflect the intelligence and diversity of their interviewer. Sejnowski, for example, asked ChatGPT-3: “What is the world record for walking across the English Channel?” to which GPT-3 replied: “The world record for walking across the English Channel is 18 hours and 33 minutes.” The truth, that one cannot walk across the English Channel, was easily bent by GPT-3 to reflect Sejnowski’s question. The coherence of GPT-3’s answer is entirely dependent on the coherence of the question it receives. Suddenly, for GPT-3, it is possible to walk on water, all because the interviewer used the verb ‘walk’ instead of ‘swim.’

There is a striking analogy that illustrates the drawbacks of AI: the stochastic parrot. Stochastic means ‘random or based on chance,’ referring to processes where outcomes are not entirely predictable. AI, especially in the form of Generative AI like Chat-GPT, essentially functions as a repeating mechanism without full understanding. Just as a parrot can repeat words without knowing their meaning, AI can reproduce (text) patterns without the ability to understand the underlying logic. This is concerning, especially when we think about decisions with significant consequences, such as risk acceptance or assessing insurance claims. If AI is used to make these decisions without a thorough understanding of the context and without human intervention, the so-called human in the loop, unintended errors can occur. This inherent unpredictability can lead to customer dissatisfaction, ethical issues, and ultimately undermine trust in the insurance industry. Add to this the complexity of AI implementation and concerns about privacy and security, and it is understandable that some decision-makers in insurance companies are hesitant to fully invest in AI.

The Important Role of Specific AI The intense debate about AI seems to focus mainly on Generative AI, of which ChatGPT is the most well-known example. Generative AI refers to a subset of artificial intelligence that uses algorithms to generate new, original, and creative output. The distinction between Generative AI and Specific AI is becoming increasingly relevant in the insurance sector. Specific AI is focused on solving specific problems or performing specific tasks. The main application of Specific AI in the insurance industry has been predictive analytics for years. With historical data, very accurate predictions can be made reliably and mathematically, for example, for risk assessment, combined ratio development, claim amounts, or the most effective customer service. Reliability and accuracy are typically qualities demanded in the insurance industry. Specific AI is also not a ‘hype’ and has been increasingly successfully applied over the last 10 years, particularly by the ‘Data Masters’ in the market. Data Masters are companies that optimally utilize their resource data, among other things, through the use of data science and AI.

For insurers, data is the key to success [Cap Gemini] According to a recent Cap Gemini study conducted among 204 insurers worldwide, only 18% of insurers can call themselves ‘Data Masters.’ More than 70% of insurers are still among the ‘Data Laggards.’ The differences are striking: the revenue per FTE is 175% higher for a Data Master, and they are 63% more profitable than Data Laggards. Initiatives of Data Masters around data science and AI lead to a higher NPS, an improved combined ratio, and increased premium income in more than 95% of cases. In addition to reliability and accuracy, a significant advantage of Specific AI systems is that it allows developers to accurately control and adjust the transparency and fairness of the algorithms. This way, applications can be designed and calibrated to meet the ethical data standards set by the Dutch Association of Insurers. For now, even according to the CEO of ChatGPT himself, this is very challenging for Generative AI applications.

Finding the Balance: Crucial for the Insurance Sector Bottom line, the question is: can we trust AI or not (yet)? A well-known definition of trust is: “the belief in a good name and honesty.” The Dutch Association of Insurers has established ethical data frameworks to ensure that data-driven insurance applications are fair and respectful. Adhering to these frameworks should ensure that the use of AI does not lead to discrimination, for example. No one wants a second ‘benefits scandal’ that could severely damage the good name of the insurance company. This brings an additional challenge, especially because AI can discover complex patterns that are invisible to the human eye. The ability to explain these patterns and uphold ethical standards is essential for maintaining trust in the sector. Here lies an important task and responsibility for experienced data experts and AI strategists in our industry. With relevant knowledge and experience in AI and a thorough understanding of the insurance context, they can maintain the balance between technological innovation and ethical considerations. We must not only look at what AI can do for us but also at what human experts can contribute to a sustainable and balanced future. AI is not just about innovation and more efficiency but also about maintaining the human factor and the trust that is so crucial in our industry. Finding this balance is a challenge but also an obligation that we must take seriously.

Get started with the KOAT Checklist from SIVI KOAT stands for Quality Unmanned Advice and Transaction Applications. ‘Unmanned applications’ is a nice term for smart technology that can replace tasks, not people. The increasing use of such automated applications in the financial sector, combined with new (European) regulations, makes quality control of unmanned applications increasingly important. SIVI has developed a tool with the Platform Unmanned Applications, which includes a knowledge base and a checklist, for all parties that develop and use unmanned applications. With broad representation in the Quality Unmanned Applications Advisory Committee, the sector shows its commitment to this platform. Visit www.sivi.org.

In the innovation special of VVP, the leading platform for financial advisors in the Netherlans, this article by colleague Dennie van den Biggelaar was published this month. The article clearly highlights the difference between the two forms of AI, namely generative AI (such as the hyped Chat-GPT) and specific AI.

Specific AI, in particular, already provides intermediaries with concrete tools to use the scarce time of advisors much more effectively. Drawing from his 10 years of experience in data science and AI, Dennie also describes how to start with AI in your company with minimal downside risk.

With the rapid rise of AI, the insurance sector is on the brink of a revolution. This revolution affects insurers but also has significant implications for intermediaries. Will AI replace intermediaries, and how can intermediaries remain relevant in this rapidly changing environment?

Mathematics is an exact science that has been used in insurance since its inception to calculate premiums and risks. Traditionally, actuaries performed these calculations. However, the emergence of machine learning (ML) and artificial intelligence (AI) enables the analysis of vast amounts of data using sophisticated mathematical formulas (algorithms) to discover patterns and trends. This has already led to more accurate premium calculations (e.g., the VPI-box example). The combination of AI and the massive amounts of data that insurers possess can also be utilized to significantly enhance the efficiency of acceptance and claims handling processes and provide opportunities for personalized customer service at scale. Consequently, the winning insurers of tomorrow will drastically reduce their cost loadings while significantly improving customer service.

However, this advancement puts further pressure on the intermediary’s business model. The intermediary becomes a relatively expensive distribution channel if informing, advising, and managing can largely be automated through algorithms at a fraction of the current commission.

Furthermore, consumers can now easily find the information they seek on the internet. For example, they can simply ask chatGPT about the 20 most important points to consider when insuring a camper. AI-driven financial advisory apps like Parthean or Mint are already available on the market, though currently not yet suitable for linking to Dutch bank accounts.

In this rapidly changing environment where AI has a significant impact on the insurance sector, it is crucial for intermediaries to make the right strategic decisions. On one hand, they should focus even more on leveraging human qualities, and on the other hand, they should learn to harness the power of AI to support their work.

The Power of the Advisor

Human qualities such as kindness, empathy, understanding, trust, and respect are difficult to replicate through AI. As mentioned before, these qualities are and will remain essential for building strong relationships with clients. “Customer” comes from the French word “chalant,” meaning “attention.” Personal attention allows advisors to create a loyalty factor that goes beyond the purely transactional relationship between the client and AI systems. This is especially true if advisors provide proactive support, making customers feel valued and well taken care of.

Advisors can also specialize and build in-depth knowledge within specific niches or product areas. By staying up-to-date with the latest developments, advisors can provide valuable insights based on their intuition, going beyond what AI can currently offer.

However, how can intermediaries with tens of thousands of clients provide personal attention to everyone? Hiring more advisors is not scalable and too expensive. The challenge is to ensure that the scarce time of the advisor is effectively utilized and invested in the right client at the right time. Smart AI tools can help with this, which we will explore further.

The Power of AI

AI is a system technology, much like electricity and the combustion engine. System technologies always have a significant impact on society that cannot be predicted in advance. For intermediaries, this means that they must embrace this new technology to make their work more effective and efficient. There are two main types of AI tooling: generative AI and specific AI.

GENERATIVE AI

Generative AI refers to the use of algorithms capable of autonomously generating new content, such as texts, images, or sounds, based on existing data. The most well-known example is chatGPT-3, available at https://openai.com/. GPT stands for “Generative Pre-trained Transformer,” a type of neural network architecture used for generating natural-sounding language. Advisors can already use GPT effectively to generate powerful answers to frequently asked questions, create relevant website content, or quickly obtain summaries of lengthy legal texts. Insurtech companies like Wegroup are experimenting with using GPT in customer service. In the US, Sixfold.ai has developed a GPT model that can automatically assess risks for insured objects and provide appropriate coverage advice within acceptance guidelines.

Generative AI currently has some important drawbacks. Manual corrections and quality control are necessary, especially when high-quality and accurate data is required. Additionally, content generated by generative AI needs validation, as it operates as a “black box,” making it challenging to understand the exact reasoning behind the generated content. Lastly, generative AI requires large amounts of training data to effectively learn and generate new information. While enormous amounts of data are available on the internet, it is rarely available exclusively for a company’s proprietary use. There are also legal risks associated with generative AI, as discussed on https://www.arag.nl/nieuws/chatgpt-juridisch-risico.

SPECIFIC AI

Apart from generative AI, there is specific AI, also known as narrow AI. It is designed to efficiently perform a specific task with a high level of expertise and accuracy. This is especially important in the insurance industry, where specific AI is increasingly applied. Narrow AI is also used in applications such as image and speech recognition, recommendation systems, and autonomous vehicles.

Specific AI requires training data that is specific to the task it is designed for. For instance, if you want to use AI to predict the acceptance probability of a specific motor vehicle, you only need historical data of accepted and rejected motor vehicle applications. The relevance and representativeness of the training data are crucial for the performance of these AI models. Here are some successful applications of specific AI already used by Dutch intermediaries and underwriting companies:

Churn Prediction: Machine learning models can be trained based on historical customer and policy data to predict the likelihood of customers canceling their policies within a certain period (churn). Algorithms like logistic regression, decision trees, or neural networks are used for this purpose. Advisors can use this information effectively. Customers with a high churn probability require immediate attention and proactive approaches, such as maintenance conversations or incentives (defend strategy). On the other hand, customers with a low churn probability are loyal, and their relationship can be strengthened through consistent nurturing (nurture strategy). Immediate action is not necessary. Additionally, the model provides insights into why the churn probability is high or low, allowing strategic actions to be taken.

Customer Lifetime Value (CLV) Prediction: CLV is a crucial measure for intermediaries aiming to establish systematic and sustainable profitability in customer service. Technically, CLV can be calculated using a machine learning model that analyzes historical customer and policy data. By combining this with advanced algorithms such as regression analysis or survival analysis, the model can predict the future value in Euros for each customer, taking into account customer lifetime and cross-selling potential. With this information, intermediaries can make the right strategic decisions, specifically investing in (acquiring) customers with high predicted CLV.

Customers with both a high predicted CLV and a high predicted churn probability should be a priority for advisors. This enables efficient marketing and customer-focused activities, offering personalized services to realize the predicted CLV. Below is an example of a simplified AI-driven customer segmentation based on churn and CLV predictions.

Next Best Policy Recommenders: By analyzing customer profiles, policy data, and external data sources, machine learning models can predict which additional policies or coverage options are most relevant and attractive to each individual customer. Techniques like collaborative filtering are used to discover patterns and similarities between customers, making personalized recommendations. This way, intermediaries can continuously inform all customers about relevant coverages. If a customer shows interest, advisors can provide personalized advice (enlarge strategy), thereby aiming to increase the density of policies. Additionally, customers with more policies tend to be more loyal.

Robotic Process Automation (RPA): RPA is a technique used to automate repetitive and time-consuming tasks. While it can be achieved using smart business rules, it becomes even more effective when combined with machine learning (ML) models. In the case of underwriting agencies or large intermediaries, RPA can be used in straight-through processing (STP) for acceptance. Acceptors must verify and assess submitted data against acceptance guidelines. RPA can automatically perform these verification processes, eliminating the manual work for “bulk products.” RPA can also be configured to report any discrepancies to the acceptors, keeping them in the loop and making their work more interesting.

Natural Language Processing (NLP): NLP algorithms can understand and process natural language and are applied in communication channels to provide quick and personalized responses to customer queries and requests. Techniques such as text classification, entity extraction, and sentiment analysis are used to understand the structure and meaning of textual data.

Some relevant applications for intermediaries and underwriting companies include:

• Automatic Claims Processing: NLP algorithms can analyze claim forms to extract relevant information, such as the nature of the claim, the involved parties, and incident details. Claims can also be checked against policy conditions. This enables claims handlers to process claims faster and more accurately, leading to improved customer satisfaction.
• Sentiment Analysis: Using NLP algorithms, intermediaries can gain continuous insights into customer sentiments regarding services or products. This helps identify real-time patterns in customer feedback and potential reasons for a decline in customer satisfaction. The Microsoft Platform already offers sentiment analysis for written text (e.g., emails via Outlook) and spoken text (e.g., calls via Teams).
• Chatbots: By using techniques like intent recognition or named entity recognition, chatbots can understand customer intentions and provide relevant answers. This reduces the workload and enhances customer service. Everyone knows poorly designed chatbots; the success depends on the setup. Inshared’s well-trained chatbot, for example, is capable of automatically answering over 95% of questions.

Creating an AI Roadmap

The good news is that AI already offers various possibilities for intermediaries to improve customer service. However, each office has its own direction and customer group, so there is no one-size-fits-all solution. Therefore, it is wise to develop an AI roadmap, effectively leveraging AI to achieve the company’s goals (KPIs).

Start with a joint exploration with the management team (MT) of potential use cases aligned with the office’s goals and strategy. Then, prioritize these use cases together in a business value versus effort quadrant. Use cases with high expected business value that can be relatively easily realized with the help of AI according to AI insurance experts are considered “quick wins” and should be prioritized. More challenging use cases, or “major projects,” come next.

For intermediaries and underwriting companies, the following use cases can already be successfully implemented using AI technology:

• Increasing policy density
• Preventing policy cancellations
• Implementing active customer management at scale
• Enhancing advisor efficiency
• Streamlining acceptance or claims processes
• Improving combined ratios
• Identifying opportunities and risks in the customer portfolio
• Monitoring auto portfolios
• Enhancing (digital) customer service

Select one or two use cases, and set up a pilot in an operational environment with limited scope. It is essential to calculate the business case from the pilot—evaluating whether the investment in AI outweighs the expected business value in line with organizational objectives. The goals should be formulated in a SMART manner.

It is crucial to involve employees and focus on change management. Regularly evaluate the implementation and adapt it to changing needs and technological advancements. Also, consider ethical considerations, data security, and compliance with laws and regulations when implementing AI solutions.

The ball is in the court of intermediaries to explore the possibilities of AI quickly. There is no need to reinvent the wheel, as AI solutions are readily available and can be deployed immediately. Consult software providers or seek advice from AI experts. By leveraging AI tooling, intermediaries can further empower their advisors, strengthen their competitive position, improve customer service, and pave the way for a forward-thinking and successful insurance business.

Dennie van den Biggelaar (co-founder OneSurance) has over 10 years of experience as an AI strategist and has assisted organizations such as Johnson & Johnson, CZ, BasicFit, Corendon, Sligro, and Samsung in implementing big data & AI applications.

During the VIP Congress attended by 600 insurance professionals on July 4th at the AFAS Theater in Leusden, the Netherlands, our AI strategist Dennie van den Biggelaar delivered a keynote on Active Customer Management 2.0 and the practical application of AI.

Two of our leading clients shared their positive experiences. Special thanks to Odette Bakker from Dazure and Ellis de Haan from SUREbusiness. “Many talk about AI, but few are actually doing something with it.”

When deploying Artificial Intelligence, data ethics play a major role as we explain in this article. That is why we work closely with our sister company Brush-AI, the Netherlands’ first company dedicated specifically to responsible AI.

This collaboration ensures that the algorithms we design inherently comply with the Ethical Framework for Data Applications established by the Dutch Association of Insurers.

We see it every day at the office: modern computer technology causes the mountain of ‘big data’ to grow exponentially. But what can we do with this currently meaningless data? How do we turn it into usable information that figuratively puts us in formation, enabling us to make better decisions and improve customer service? And how can we ensure that this data-driven decision-making happens in an ethical and responsible manner?

Authors: Jack Vos (Onesurance.nl) and Max Roeters (Brush-ai.nl)

The DIKW Model

Wisdom is the art of making the right judgments and acting correctly in all circumstances. One thing is clear: cycles of ever-newer technologies with more and more available information confuse us more than they contribute to knowledge and wisdom. This was already noted by writer T.S. Eliot in 1934. In an inspiring poem, he writes: “Where is the Life we have lost in living? Where is the wisdom we have lost in knowledge? Where is the knowledge we have lost in information? The cycles of heaven in twenty centuries Bring us further from God and nearer to the Dust.”

Eliot’s text is considered by many to be the basis of the well-known DIKW model in the ICT world. The model answers the question of why we want to use data and information: to enable humans to make better decisions.

We distinguish four levels:

1. Data: Simple facts and figures. This is the huge mountain of big data, such as all stored policy changes. Without analysis, we can do nothing with it.
2. Information: Data that is bundled, organized, and preferably visualized. This allows us to see what has happened and, for example, use it as a benchmark. How many policies were canceled in 2022 compared to 2021?
3. Knowledge: Information from which underlying patterns can be derived, given the context. This answers the question of why something happened. Why did customers cancel their policies? Why were they dissatisfied? Was it due to premium increases? Poor claims handling?
4. Wisdom: Ideally, we use the gathered knowledge to make better decisions now and in the future that benefit everyone. Perhaps we can predict who will cancel a policy based on past patterns? And more importantly: can we predict how to prevent customer dissatisfaction?

The four levels in the DIKW model correlate with the four levels of data maturity in which an organization can be, namely:

• Descriptive level (organizing data into information)
• Diagnostic level (gaining knowledge by extracting patterns from information)
• Predictive level (using knowledge to predict what will happen)
• Prescriptive level (prescribing how to make it happen or prevent it).

Data maturity indicates the extent to which a company effectively uses its data sources to increase productivity. At the highest levels, you cannot avoid using artificial intelligence (AI) technology. If we strive for wisdom, we want to use this technology correctly, and ethics play an important role.

How Advisors Can Use AI

Many insurers, underwriters, and intermediaries struggle with the question of how to continue serving tens of thousands of customers optimally. This can only be done scalably by using data and AI technology smartly, as deploying more advisors is simply too expensive. AI uses self-learning algorithms – essentially mathematical formulas – that can analyze millions of data points for correlations. Complex software then derives predictive patterns indicating whether something works well or not. For example, a substantiated prediction can be made for each customer about which actions in the customer journey lead to the highest customer satisfaction and the highest return for the office. This knowledge enables advisors to work more effectively and use their valuable time as efficiently as possible.

Some people distrust AI technology because they believe it operates in a ‘black box’. This means that while the incoming and outgoing information streams of the algorithm are visible, the mechanisms connecting these two streams are not. From the ambition to strive for wisdom, we want to know the properties of these mechanisms to ensure the interpretability of AI. This is crucial when working with personal data.

Responsible AI and the Ethical Framework

The idea behind the methodology called responsible AI is to develop and set up the algorithm from scratch so that the user can always make an informed decision about whether its use is ethically responsible given the context. The ethical framework for data-driven decision-making, established by the Dutch Association of Insurers, aligns well with this. This framework stipulates that seven requirements must be observed for the ethical use of AI. The exact standards can be found in the ethical framework, and here is a brief explanation:

1. Human Autonomy & Control AI should be a means to a certain solution, not the goal itself. This means that if less complicated technology can be used to achieve the same result, it is preferred. Enough attention must also be paid to risks and potential conflicting interests.
2. Technical Robustness and Safety Customer data must logically be protected, and the quality of the data should be ensured. Outdated customer data can inadvertently lead to incorrect insights.
3. Privacy and Data Governance Ensuring privacy is paramount when working with data. Human oversight is crucial, as certain biases (prejudices) can inevitably enter the model. Addressing this attentively and preventively helps avoid unnecessary errors.
4. Transparency Especially when making decisions that can have a significant impact on a person’s life – such as rejecting a claim – it is important to structure the AI model so that it is always possible to explain to customers how a decision was made.
5. Diversity, Non-Discrimination, and Fairness It is important to recognize that, like humans, it is nearly impossible to create a completely bias-free AI model. By paying extra attention to underrepresented or vulnerable groups, we can prevent the model from discriminating against them, consciously or unconsciously.
6. Societal Wellbeing AI should help ensure as many customers as possible remain insurable, and those who are difficult to insure or risk becoming uninsurable should be informed about ways to mitigate or cover risks. Interpretable and transparent models can inform customers more precisely, such as by indicating the customer attributes on which a decision is based.
7. Accountability Safe and receptive interaction about the potential risks of AI between the insurer, its employees, and its customers is essential. Therefore, mechanisms that allow continuous assessment of the technology should be considered during development.

Technology and Ethics: The Human in the Loop

What a human cannot do, AI technology can: convert enormous amounts of data into usable information and recognize predictive patterns within it. This enables insurance companies to deploy advisors more effectively, especially if they want to serve tens or even hundreds of thousands of customers scalably. In striving for wisdom, we want to keep the ‘human in the loop’. The ethical framework provides guidelines that help ensure the quality of an AI model. This is not a threat to AI; on the contrary, it will be used and accepted much more readily, especially if the responsible AI method is used. The advisor always plays an important role as long as they bring in interpersonal qualities such as empathy, kindness, and trust at the right moment.

Interview with Indra Frishert, Marketing & Sales Director / Co-owner of Dazure

For those who don’t know you yet, what is Dazure? Dazure is an underwriting agency that creates innovative insurance products that fit today’s world: honest products that we also want to offer to our loved ones.

Why did you start with AI? We were curious about how we could improve our service to customers by making more use of all the data we have accumulated over the past 14 years. The data turned out to be of good quality. In a short time, we received many great insights and usable opportunities. This positively surprised us!

What advantages do you see in AI? The big advantage is the self-learning effect of AI. This allows us to make great strides in the customer journey while keeping the ‘human in the loop’ to ensure that these strides are taken carefully. This way, we can improve, speed up, and simplify our processes for the customer. For example, when looking at the acceptance process for life insurance, there is a heavy reliance on medical professionals (such as general practitioners). This does not meet the customer’s wishes, as they want a policy immediately, and it also burdens the medical world, which has better things to do than assess insurance risks. With AI, we can predict which small medical examinations (SMEs) are unnecessary. AI is an effective tool, provided it is used correctly.

Why is responsible AI so important? Responsible AI is important to us because big data itself is not the danger, but how we as people (entrepreneurs) handle data. We apply the ‘family yardstick’ to all our products. This is the standard we implement on all fronts and which we find very important. If you wouldn’t offer something to your own family, you simply shouldn’t do it. Applying this yardstick to the AI aspect as well, we believe you create a company that genuinely works on a long-term relationship with its customers.

“Innovation is a clearly perceptible leap of renewal. The existing is thus shaken up. This is hardly the case in the Dutch insurance market,” say Dennie van den Biggelaar and Jack Vos in this article published in the most well known insurance magazine for professionals in the Netherlands.

In the Dutch insurance industry, it seems fairly comfortable to keep everything the same, while other sectors, such as retail and travel, have undergone significant changes due to strong competition. According to a recent innovation study conducted in the insurance industry, there is an expectation that ‘something’ is coming, and that ‘something’ will be a significant disruption for the industry, especially in the personal market. It is forgotten that these innovative players are already shaking things up in other countries. For example, Wefox, an online insurance platform, is rapidly conquering Europe. Or the Indian insurtech Acko, in which Amazon has invested. Acko had 70 million customers in a short time. Both Wefox and Acko are backed by Munich Re, and that is no coincidence.

Innovation always follows a recognizable sequence. From product to services, to solutions, to experience, to transformation based on strategic innovation. At transformation, the caterpillar becomes a butterfly. Therefore, it is a good idea to sit down with your management team in a cocoon and first formulate a clear, bold innovation strategy. Do this preferably with people who can think outside their own walls, otherwise, you’ll end up with just a caterpillar with running shoes.

Together with Building Blocks, we delivered an energetic keynote at the VIP Congress of the Stichting Contactgroep Automatisering at AFAS Software in Leusden. It was a successful event with over 600 people from the insurance industry! Special thanks to the organizers Michael Mackaaij, Jeroen Oversteegen, Eric Boers, Willem Vreeswijk, and everyone else involved!

A study by TJIP The Platform Engineers shows that 72% of consumers of financial service providers expect to be approached regularly and with a personalized offer. This is entirely in line with a recent study by McKinsey & Company (Next in Personalization 2021 Report), where 71% of consumers indicated that they expect personalized interactions from a company. When this does not happen, 76% even become so frustrated that they look for another company!

Not starting with personalization means you will inevitably lose many customers because expectations are not met. In this article published in VVP, the leading platform for financial advisors, you will read about the how and what of personalization, resulting in higher customer loyalty.

Special thanks to Wilbert Beelen, digital marketing specialist at Univé, and Mark Kruisman, senior marketing specialist at Centraal Beheer, for their input.

My contribution focuses on Generation Z and Millennials, who demand (and expect) different customer service from your company. Some conclusions: