Mathematics is an exact science that has been used since the inception of insurance to calculate premiums and risks. Traditionally, this was done by actuaries. However, with the advent of machine learning (ML) and artificial intelligence (AI), it is now possible to analyze large amounts of data using advanced mathematical formulas (algorithms) and uncover patterns and trends. This has already led to more accurate premium calculations (for example, the VPI-box). The combination of leveraging AI and the vast amounts of data inherently held by insurers can also be used to make acceptance and claims handling processes much more efficient and offers opportunities for personalized customer service at scale. All this means that tomorrow's winning insurers will dramatically lower their cost loading while significantly improving customer service.

The harsh truth is that this will further pressure the intermediary's earning model. The intermediary becomes a relatively expensive distribution channel if informing, advising, and managing can largely be done by algorithms for a fraction of the commissions currently paid out.

Consumers, in turn, can increasingly find the information they seek on the internet with ease. For instance, you can simply ask ChatGPT what the twenty key points are to consider when insuring a camper. There are already various apps on the market that use AI to provide tailored financial advice, such as Parthean or Mint, although they are not yet suitable for linking to your Dutch bank accounts.

In this rapidly changing environment where AI has a major impact on the insurance sector, it is of crucial importance for intermediaries to make the right strategic decisions. On one hand, by focusing even more on the use of human qualities and, on the other, by learning to harness the power of AI to support their work.

The Advisor's Strength

Human traits like friendliness, empathy, understanding, trust, and respect are hard for AI to replicate. These qualities have been mentioned in this magazine before and remain essential for building strong relationships with clients. The word 'client' comes from the French word ‘chalant’, which means ‘attention’. We all know the word ‘nonchalant’. With personal attention, advisors can create a loyalty factor that extends beyond the purely transactional relationship between a client and an AI system. Especially if you also provide proactive support, ensuring customers feel valued and well cared for.

Advisors can also specialize and build deep knowledge within specific niches or product areas. By staying up to date with the latest developments, advisors can use their intuition to provide valuable insights that go far beyond what AI can deliver.

So far so good, but how can you as an intermediary with possibly tens of thousands of customers give everyone that personal attention? Deploying more advisors is not scalable and far too expensive. So how do I ensure that the scarce time of the advisor is used as effectively as possible and invested in the right customer at the right time? Smart AI tools can help with this.

The Power of AI

AI is a systemic technology, like electricity and the combustion engine. A systemic technology always has a major impact on society that cannot be foreseen in advance. For intermediaries, this means they must embrace this new technology, whether they like it or not, to make their work more effective and efficient. We distinguish two main types of AI tools: 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 on the website openai.com: ChatGPT-3 . GPT stands for Generative Pre-trained Transformer. It is a type of neural network architecture used to generate naturally sounding language. Advisors can now effectively use GPT to generate powerful answers to frequently asked questions, create relevant content for websites, or quickly request a summary of lengthy legal texts. Insurance software companies such as Wegroup are experimenting with this to also use GPT in customer service. In the US, Sixfold.ai has developed a GPT model that automatically assesses the risks of insurable objects and provides suitable coverage advice within provided acceptance guidelines.

For now, generative AI has several important drawbacks. Manual corrections and quality control are always necessary, especially when high-quality and accurate data are required. The content also needs validation due to it being a black box. Understanding the exact reasoning behind the generated content is very difficult. Finally, generative AI requires vast amounts of training data to learn effectively and generate new data. These huge amounts of data are available on the internet but are almost never available within a company for its own exclusive application. There are also legal risks, such as those noted at www.arag.nl/nieuws/chatgpt-juridisch-risico.

Specific AI

Besides generative AI, there is specific AI, also known as narrow AI. It focuses on efficiently performing a specific task with a particularly high level of expertise and accuracy. This is something very important in the insurance world.

For the following use cases, AI technology can already be successfully used by intermediaries and brokers: increasing policy density, preventing cancellations, setting up active customer management at scale, deploying the advisor more effectively, streamlining acceptance or claims processes, improving the combined ratio, identifying bleeders and feeders, monitoring car portfolios, and improving (digital) customer service. Narrow AI is also used in applications like image and speech recognition, recommendation systems, and autonomous vehicles.

This AI requires (only) training data that is specific to the task for which it is designed. Suppose you want to use AI in this way to predict the acceptance (chance) of an STP motor vehicle, then you only need historical data of accepted and not accepted vehicle applications. The relevance and representativeness of the training data are crucial for the performance of these AI models. Here are some applications that are already being successfully used by Dutch intermediaries and brokerage companies.

'Advisors can use GPT effectively to generate powerful answers to frequently asked questions.'

Churn Prediction: Based on historical customer and policy data, machine learning models can be trained to predict the likelihood of customers canceling their policies within a certain period (churn). Algorithms such as logistic regression, decision trees, or neural networks are used for this. Advisors can work very targeted with this information. Customers with a high churn probability need direct attention and should be proactively contacted for a maintenance call and/or with an incentive (defend strategy). Customers with a low churn probability are of course the loyal customers, and there the relationship can be strengthened in a structured way (nurture strategy). Direct action is not needed. In addition, the model also reveals at a diagnostic level why the risk of cancellation is high or low, enabling strategic response.

CLV Prediction: Customer Lifetime Value (CLV) is an important metric for intermediaries who want to set up customer service systematically and future-proof—sustainably profitable. Technically, CLV can be calculated using a machine learning model that analyzes the same historical customer policy data. By combining this with advanced algorithms such as regression analysis or survival analysis, it predicts the future value for each customer in euros while considering customer lifetime and cross- and upsell potential. With this, the intermediary can make the right strategic decisions—logically, by investing targetedly in (the acquisition of) customers with a high predicted CLV.

Customers with a high predicted CLV and a high predicted churn probability should naturally receive the highest priority from the advisor. Thus, market and customer-focused activities can be deployed efficiently, and personalized services can be offered 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 be used to predict which additional policies or coverages are most relevant and attractive for each individual customer. This utilizes techniques such as collaborative filtering, where patterns and similarities between customers are discovered to make recommendations. This allows the intermediary to automatically and continuously inform all customers about relevant coverages. If a customer shows interest, the advisor can directly engage in advising the client (enlarge strategy). Thus, working towards a higher policy density. An added benefit is that customers with more policies are generally more loyal.

Robotic Process Automation (RPA): RPA is a technique used to automate repetitive and time-consuming tasks. This can be done using only smart business rules, but it becomes even more effective when combined with machine learning (ML) models. The application is mainly in STP lanes for acceptance among brokers or large intermediaries. Acceptors must verify submitted data and check it against acceptance guidelines. RPA can automate these verification processes, eliminating time-consuming manual work for 'bulk products'. The RPA is also configured to report any discrepancies to the acceptors. This keeps the human in the loop and makes their work much 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 questions 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 brokers:

• Automatic claim processing: NLP algorithms can analyze the content of claim forms to extract relevant information, such as the nature of the claim, the parties involved, and the details of the incident. The claim can also be checked against policy conditions. This enables claim handlers to process claims faster and more accurately, resulting in improved customer satisfaction.

• Sentiment Analysis: With these NLP algorithms, the intermediary automatically and continuously gains insight into how customers feel about services or products. Real-time patterns in customer feedback and potential causes of declining customer satisfaction are identified. Sentiment analysis can already be enabled as a standard in the Microsoft Platform for both written text (e.g., emails via Outlook) and spoken text (e.g., calls via Teams).

• Chatbots: With techniques such as intent recognition or named entity recognition, chatbots can understand the customer's intent and generate relevant responses. This reduces workload and improves customer service. Everyone knows, of course, the 'bad' chatbots. Much depends on the setup. InShared's well-trained chatbot is capable of automatically answering more than 95% of questions.

AI Roadmap

The good news is that AI already offers intermediaries diverse possibilities to improve customer service. However, each office has its own course and clientele, so unfortunately there is no one-size-fits-all. Therefore, it is wise to work on an AI roadmap, where AI is effectively employed to benefit the company's goals (KPIs). It is a good idea to first explore possible use cases within the company's goals and strategy together with the MT. The use cases are then jointly prioritized using a business value versus effort quadrant. Use cases with high expected business value that, in the judgment of AI insurance experts, can be relatively easily realized with AI for the office are the quick wins. These come first in the roadmap. More challenging use cases, the major projects, are addressed later. This way, one, at most two use cases are selected, for which a pilot is set up in an operational environment with a limited scope. It is important that the business case can be calculated from the pilot. In other words, does the investment in AI weigh against the expected business value in line with the organizational objective? Goals must be formulated SMART.

'AI already offers advisors various possibilities to improve customer service.'

It is very important to also involve employees and give substance to change management. After that, the implementation must be regularly evaluated and adjusted to meet changing needs and technological developments. Also, consider ethical considerations, data security, and compliance with laws and regulations when implementing AI solutions.

The ball is in the intermediary's court to quickly explore the possibilities of AI. There is no need to reinvent the wheel. AI solutions are available that can be deployed directly. Consult software suppliers or seek advice from AI experts. By harnessing AI tooling, the intermediary can further leverage the power of the advisor, 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 been an AI strategist for over ten years and helped organizations like Johnson&Johnson, CZ, BasicFit, Corendon, Sligro, Samsung with deploying big data and AI applications.

View the article on LinkedIn or download it as a PDF