Brandon Nuttall, chief data and AI officer at Xceedance, explains why data management is essential if insurance innovators want to supercharge their businesses.
The world is facing an unprecedented range of challenges such as flooding and wildfires - many of them driven by climate change. Pooling risk via the insurance industry is the most obvious way to protect against the impact of these catastrophic events.
The current insurance model, which relies heavily on human intervention, has failed to meet the challenge. Despite the industry’s efforts to broaden the range of businesses, communities and individuals able to procure insurance protection, only six percent of insurable assets worldwide are covered. It is clear that we have a fundamental problem when the people who need insurance most cannot access it.
One of the keys to addressing the ‘insurance gap’ is to build accessible products using innovative technology fuelled by the highest standards of data management. Companies which embed data management excellence at the heart of their businesses are primed for exponential growth.
Generative artificial intelligence (GenAI) – as part of a data strategy - has the potential to supercharge the insurance industry and enable it to create commoditised products, akin to those in the banking industry.
Insurance is more complicated than most other financial products, and, to date, the technology to support commoditisation has not existed. The industry has been hamstrung by a lack of standardised data, as well as by keeping data in various silos. GenAI, however, is the technology that can enable us to rapidly process and standardise data to create more commoditised products that will improve accessibility and drive industry growth.
Data ingestion: begin with the end in mind
When looking to introduce new technology platforms, before beginning the data ingestion process – collecting / moving information to the right location for processing – it is important that insurance companies think carefully about what they are trying to get out of the exercise.
For instance, an organisation that wants to use data analytics to support risk selection will want to have an accurate figure of the actual sums insured – as opposed to the declared value – in an insurance policy. Meanwhile, a general insurer that hopes to improve customer engagement will want to analyse data for the best ways to cross sell, upsell and grow their share of wallet with their end customers.
Focusing on the outcome – and your unique market proposition – is key for optimising your spend.
Integrate for growth
While many organisations think deeply about their API strategy, in which services talk to one another, they don’t devote as much time and attention to their data integration platform. The key to getting the most out of an API ecosystem is to understand the data and how it interacts with all a company’s systems.
One of the key constraints slowing down data integration projects is the semantic understanding of data in context – most organizations have few individuals who have the necessary business and technology skills. Leveraging Generative AI can support these individuals, and unlock scaled data integration capabilities.
Get the fundamentals right: data storage and security
It is critical to bake information security into the front end of the process. It’s not enough to ask a couple of courtesy questions on a Request for Proposal. Instead, an organisation, especially a multinational operating in different jurisdictions, should ensure it has a thorough understanding of the fundamental assumptions made and the controls included in packages they wish to onboard.
Think carefully about how you create products
Organisations can create true differentiation through data analytics. However, some organizations place too much importance on the data products they create and not enough on how they create them. The temptation is to develop one-off products which solve the immediate problem, but that don’t align with the overall data strategy or governance approach.
For instance, an actuary could take an extract from a database to create a model, but if they leave the company or lose focus on it, the model can become “shadow IT” with no clear development or maintenance plan. For that reason, it is essential that any data analytics exercise is carried out in a well-planned, structured, maintainable way to minimise the risk of creating outmoded and unusable bits of technology.
Listen to experts: the role of the Chief Data Officer
It is critical to have a Chief Data Officer who appreciates the need for data governance, and who understands where the information is sourced from, as well as the importance of bringing them together for short- and long-term success. Data leadership must be empowered to break siloes and create data platforms which serve all.
Embrace tech for exponential growth
The time has long passed to view the cloud as a separate aspect of our IT and data infrastructure. Secure cloud operating capabilities allow organizations to keep pace with the changes which Generative AI is bringing.
The outputs of GenAI tools are already drastically better than they were two years ago. Although the insurance industry is beginning to take strides in implementing AI-driven tools, it is still nowhere near enough. We should aspire to an insurance industry which delivers financial security for all – embracing the possibilities brought to us through machine intelligence.
By Brandon Nuttall, chief data and AI officer at Xceedance
No comments yet