Panel Information
Speaker: Yannick Even – Global Business Analytics Partner at Swiss Re
Three main categories of change
- Customer preferences – Customer preferences are changing rapidly as customers look for more value and dynamic digital engagements. Modern customers have emerging protection needs and expect personalization, which poses a challenge for insurers who lack the right data.
- Technology – The prevalent use of intelligent chip sensors has enabled real-time data collection to generate personalized insights, which is set to become more common. APIs, for example, enable ecosystem and platform interaction, whereas cloud computing power allows insurers to process and transform data into meaningful insights.
- Shifting landscape – The current insurance environment is competitive, and the regulatory landscape is evolving rapidly around the responsible use of data and AI models. In some risk pool areas, there is limited growth. Thus, shareholders are asking insurers for more efficiency in the way business is delivered so that better protection can be provided to the masses.
Main tech clusters and tech trends impacting insurance today
When it comes to innovation and evolving technology, some tech clusters are emerging and starting to support insurers in their transformation into data-driven companies. But investing in one technology alone will not do the trick; only understanding how technologies interact with each other and using a strategic approach in bundling technology to solve data and customer challenges will make a difference.
Evolution of data & AI-driven insurance
The insurance target operating model has evolved from what it was years ago. Back then, insurers managed their processes and data in different systems that didn’t interact with each other. Activities such as creating new products, product innovation, assessing and pricing the risks, as well as serving the customers are done in silos. But now, most insurers have moved to a more digitalized target operating model. The customer has a single view across all the processes, and the silos can interact with external parties and have direct connections with data or service partners.
As this transformation continues, we will soon be in a world where data and AI will drive a lot of the decision-making process. To prepare for this, several insurers are adopting end-to-end digital platforms to serve customers across the entire value chain. Similarly, the use of APIs and innovative technologies are enabling insurers to interact with the distribution data and customer engagement partners.
As a result, it is believed that insurance will evolve from being product-centric to service-centric, particularly in embedded insurance.
The decisive role of data & AI in business performance
Data and AI models are starting to play a decisive role in the performance of businesses and innovation of products, and as we’ve seen in other industries, it’s beginning to expand in insurance as well. Historically, insurance companies in Asia did not have access to high-quality customer data. However, as insurers digitize their touch points, they gain access to more data, which allows them to better understand customer needs. This allows insurance companies to tailor their products and make provisions to these needs, improving customer service and augmenting employee decision-making.
In the next few years, 1/3 of all the data created will be real-time, and data collected from AI embedded in smartphones and sensors will be able to react proactively with a greater capability to prevent, understand, and manage risks. COVID-19 has accelerated digitalization, but insurance companies are not yet ready to work with real-time data. Insurers are increasing their efforts to keep up, and it is crucial for insurance companies to focus on data transformation. Getting it right will help insurance companies generate meaningful and actionable insights to improve their business performance.
Harnessing the full power of data
Insurers have traditionally had a large amount of declarative data, and with numerous intermediaries between the points of data collection and the final data, the data collected is not always of good quality or accuracy. However, with the digitalization of operations and customer touchpoints, this is beginning to change.
Insurers will gain access to more transactional data by collaborating closely with distribution partners, banks, and others. Similarly, their ability to create a powerful customer and risk model grows as they partner with big tech companies and integrate into larger ecosystems. This has the potential to alter how insurers manage and price risks in the future.
Translating data into actionable insights NOW is critical for business success
Insurers who adopt a data analytics strategy can harness the full power of the data at their disposal and capitalize on the four key insurance drivers:
- Enabling growth
- Engaging customers
- Improving efficiency
- Optimizing portfolios
However, it is imperative to do all this in compliance with data privacy regulations, and companies must adhere to these principles because regulators are looking into the responsible use of data.
Within an organization, it is critical for insurers to place importance on fairness, ethics, and transparency for AI to scale. Underwriters and actuaries must fully understand how the model makes a decision. So, having a transparent mechanism in place is critical for gaining the trust of talent and explaining to end users how their data is used.
To create actionable insights and value for insurance
It has long been the practice of insurers to use data modeling primarily in underwriting, pricing, and distribution. Yet, it is now prevalent throughout the value chain, with insurers using a data-driven approach (an AI model) to personalize offerings and mitigate risks. In addition to that, they are also leveraging AI to help customers understand the risk they’re exposed to and the value insurance can bring, paving the way for customer retention and cross-selling opportunities.
Still, challenges persist. Many insurers spend a lot of time collecting data, and the quality is not always at the right level. There is also a lack of standardized taxonomy around the data used by the industry, and too much time is spent searching and curating data rather than building a model.
But here’s what insurers can do to break the status quo:
- Have a clear data strategy and define the data quality you want to achieve.
- Understand the data points that have predictive power.
- Identify the data you need to manage your business.
Once you have these sets of data,
- Check if the data is of good quality.
- Cleanse, harmonize, and define it across silos and distribution channels. For example, if there are often different definitions for the same data point, you will need to harmonize it to use it across the organization in the same way.
- Identify and determine the data gaps. Look for third-party, open data, or government data sets to augment and reach the data level you’re hoping for.