At the request of VVP, the platform for financial service providers, our CTO and AI strategist Dennie van den Biggelaar explains how to apply specific AI and machine learning to "advice in practice. In different editions, the following topics will be highlighted: 

• Starting with specific AI and ML

• Operationalizing in business processes

• Integrate into existing IT landscape

• Measuring = Learning: KPIs for ML

• Ethics, regulation 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?

Read the text of the article here:

Integrating AI software

The insurance industry is on the brink 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 value (CLV) and make recommendations for cross- and upsell, among other things, allowing advisors to make better informed decisions. But how do you integrate these algorithms into your existing IT landscape? How do you make sure your employees have these predictions and suggestions available to them at the right time to work more easily and effectively? In this article, we discuss some concrete technical tips for successfully integrating AI decision engines into insurance systems.

Define successful integration

I firmly believe that IT issues should always serve a business purpose. Therefore, a successful integration always starts with asking the question, "when will this integration be successful? Creating a user story can help with this, an example:

"As [digital marketer of proxy company X] , I want to know [weekly which customers are in need of an additional product Y], so that I can [set up a targeted automated marketing campaign for this group] with the goal of [generating weekly (new) leads for my field consultants]."

This is a great starting point to present to the technical experts what is required of them. More in-depth questions usually follow:

• What specific information does the user want to see?

• How often should it be refreshed?

• How will we measure the success of these automated campaigns?

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

Analysis of existing IT landscape

Successful integration begins with a thorough analysis of the existing IT infrastructure. Many integration attempts fail because there is no good understanding of current systems, leading to compatibility issues. With which existing IT systems, databases and interfaces should the AI algorithm "interoperate"? What volumes of data should be passed? When and how fast?

In practice, this means collaborating and coordinating with various IT partners of backend and front-end systems. Therefore, start this inventory early and include all (external) stakeholders in your plans. 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 experienced organizations that had innovative plans, but that their IT landscape was set up too rigidly. Therefore, design a modular and scalable architecture to enable future expansion and change so that you are agile as an organization. This is why it is best practice today to use microservices architectures, where each functionality runs as a separate service. This makes it easier to add new elements, replace or update existing ones without overhauling the entire infrastructure.

Consistency and quality

Data quality is critical to the success of AI. Many AI systems perform poorly due to inconsistent, incomplete or outdated data. Therefore, implement a data cleaning and preprocessing pipeline that ensures that all data sent to the AI decision engine is clean and up-to-date. Automated data integration and validation tools can help with this, ensuring 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 that is essential for successfully training and using AI models.

Testing, validation 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 testing in a simulated environment that mimics the production environment. Validate the output of the AI models with historical data and scenario analysis. Involve end users in the testing phase to ensure that the models meet business and user requirements.

Use APIs

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

Security and privacy

Data security is critical, especially given the sensitive nature of insurance data. Insufficient security can lead to data breaches resulting in loss of customer trust and invasion of consumer privacy. Therefore, use only data that is actually needed and anonymize as much as possible. Do you really need certain sensitive data? Then apply encryption. Ensure that all data transfers between systems and the AI decision engine are over an encrypted connection. Use access controls and audit logging to ensure data security and regulatory compliance.

Conclusion

Thorough integration is a prerequisite to successfully landing AI in your organization. You not only have to pay attention to a scalable business case and the user, but also think about agility, security, privacy and quality. As a result, you therefore need a team with different competencies and will have to coordinate and collaborate with IT partners.

Therefore, first make sure that you or someone in your own organization is sharp on what the business frameworks of a successful integration are. Make these explicit so that you can transfer this. Then appoint someone who feels responsible for the realization and the associated project management. If you don't want to or can't free up resources for this yourself, you can use one of your trusted IT partners. So you can focus on your own core business!