This is Archived content. Explore our latest Insights for recent updates and perspectives.
With great enthusiasm, insurers in Denmark, Norway, Sweden, the UK, Germany and many other countries are moving towards data-driven business models. But the majority still prefer to ignore the opportunity. It seems too challenging to achieve tangible results. Together with an insurer operating on TIA, we can demonstrate that it is not that difficult.
For a long time now, the challenge for financial institutions has not been getting enough data. The amount of data we produce every day is truly mind-boggling. At our current pace, 2.5 quintillion bytes of data are created each day, and that pace is only accelerating with the growth of the Internet of Things (IoT). Over the last two years alone, 90 percent of the data in the world was generated.[1] The Harvard Business Review found that:
The challenge is rather how to act on all that data in order to gain actionable insights and achieve meaningful business results. Working with such quantities requires non-traditional approaches, hence the emergence of machine learning and deep learning (artificial intelligence). Microsoft’s Principal Solutions Architect, Scott Seely, defines eight business use cases for the insurance industry:
The extent to which each of these use cases is already used by insurers varies significantly.
Machine learning, for all its clear advantages, is employed in insurance much less often than it could be. Most companies consider it a highly complex technology and therefore opt against activities that would employ such methods. In fact, this reason is cited by 53% of insurance decision-makers as a main reason for not using ML or AI.[4] This is not surprising: effective use requires technological know-how, mathematical expertise and industry knowledge to guide the entire process towards business-oriented goals. But the fact is that effective use of machine learning is much easier than many decision-makers believe. There are many data science platforms on the market, some of them operating on an open-source basis, such as h2o.ai.
Sollers is currently exploring and developing a machine learning solution for the insurance business. The large amount of consumer data stored in core systems offers numerous possibilities to support the business with valuable insights. Conclusions based on machine learning analysis may, for example, increase the effectiveness of insurers’ cross-sell campaigns. This can be achieved by answering questions such as:
We chose an approach using h2o.ai, one of the leading machine learning platforms. It is widely adopted and operates on an open-source basis. It also requires a relatively simple technological setup.
Of course, industry knowledge is required to make sense of the data, select the most important data points and guide a machine learning solution towards analysis that has a chance of providing useful and actionable insights. Creating appropriate algorithms can involve very complicated calculations, but even relatively simple mathematical approaches can help achieve significant results.
The approach was to integrate the TIA insurance core system with the h2o.ai engine and train the engine on customer data using machine learning. The input data needs to contain structured and relevant information, including, for example, personal data, policy data, claims data and transaction data. The scheme below presents an architecture that can be applied to put h2o.ai into action.
A machine learning model organised as shown in the graphic and trained on a file containing 1.5 million data sets with 100 columns was able to select the best model with 180 decision trees. As a result, the model was able to identify 1% of consumers with a 10-times higher likelihood of purchasing a given product at a given time. For the top 20% of customers, this likelihood was nearly three times higher.
Preparing such a model requires only a few weeks and a small team. This demonstrates that large investments and months-long projects are not required to reap early benefits and evaluate the usefulness of machine learning for a company. Confirming that usefulness is of the utmost importance. Wolfgang Wahlster, founder and former CEO of the German Research Center for Artificial Intelligence, admitted in an interview: “I have already seen three hype cycles for AI – it’s very important, as I tell our researchers, not to over-promise and under-perform.”[5]
It is easy to get involved in trendy, flashy technologies and devote significant resources to projects that might be interesting but are unlikely to deliver tangible results that justify the investment. According to BCG research, navigating an unpredictable field of technology such as machine learning and artificial intelligence requires an adaptive strategy.[6] It is therefore advisable to initiate a series of small, low-cost proofs of concept aimed at validating different business use cases. Those that deliver the most promising results can then be scaled up.
Piotr Pastuszka, Senior Manager at Sollers Consulting
Footnotes: