DOCOsoft at the London Market Forum
How do you currently use data to drive decision-making? That was the first question asked at the London Market Forums Technology & Innovation Summit 2020. The Forum looked at tech impacting underwriting, claims, data management, business efficiency and on the final day there was a look to the future, with a discussion around innovation and artificial intelligence including the impact on the London market’s greatest asset, its people.
Delegates heard that Zurich is using new technologies and platforms to shape decision-making and augment data from other commercial sources to improve quality of decision-making and efficiency. Zurich is heavily invested in its strategic data asset. Like many other insurance companies, Zurich operates in a very complex landscape data with lots of repositories so to provide a single source of truth for customer and policy data is incredibly important. The strategic data asset allows it to plug into core systems and other platforms as well.
According to one panellist: “We’ve seen an increase in investment within our business of buying data sets from other suppliers to improve the quality of decision making, for example, in the underwriting space. It can help to understand the retail or SME space, the risk profile of that company and its activities. It is now about bringing that data into core systems, to enable underwriters to make quick and efficient decisions both internally and externally turning information into intelligence so we can apply that to our business processes. The ability to underwrite the right business, to look at the risk process and understand it is fundamental to profitability and revenue growth.”
Driving decision-making in Miller was the story of the insurance broker invited to the Forum to speak on behalf of intermediaries. Miller had one of the early online services platforms that shared delegated authority information, both with its cover holders and with their underwriters. What’s interesting about data is that in its raw form, it’s not inherently very useful. It’s the information that one can glean from it, the knowledge that we can add to the data, decision making – that is the focus. Miller has worked with its data team to start building self-serve platforms, so that it can democratize data access throughout its business.
One panellist said: “Culture is important. People are important, the ways that we bring technical and technology literacy to our user communities becomes ever more important, because we’re asking them to make more informed decisions about their own businesses, using the information that we’ve given them.”
The question was asked do you feel that you have enough data at the moment, and is it the right data?
The response was illuminating: “I don’t think you can ever have enough data but the strength of machine learning relies on the depth, width and richness of your data sets. This gets back to the user engagement and culture piece. We need to have a responsibility to make the capture more straightforward easier in the broking or the binding process.
According one insurance carrier participant: “It’s about finding that balance between the velocity of data that comes into the organization then deriving value and presenting it on the right platforms with the right people to make the right decisions. That’s the tricky bit. We really see a huge push in the retail area using data to make better decisions about underwriting and claims, especially for fraud detection. Platforms that can mine the data and derive meaning from apparently disparate events, so there is some really clever stuff out there.”
With fraudulent claims there is software out there that listens to voice calls and can detect stress patterns in people’s voices to help detect whether a claim is valid or not. The important thing is to get people in the business to understand how the technology can help. They are very much focused on solving problems they’ve had in the past and problems they face now, not necessarily problems they face in the future. It’s the same analogy with the military: they equip themselves to fight the war they just fought, not the war they might fight in the future.
It is up to technologists to present these technology options to the business. What meaning can you derive from this data, and operational efficiencies?