How geospatial data is changing insurance

Natural catastrophes caused global insured losses of $154 billion in 2024, according to Gallagher Re. It’s a sobering statistic, but what it points to isn’t new. For four years, insurers have absorbed more than $100 billion in losses related to catastrophes of this kind, and a key reason for the increase is climate change.

The climate crisis is reshaping the world. Already, it affects the lives even of those in the more prosperous countries of the Global North in numerous ways, with the tragic scenes in Los Angeles being only the most recent reminder. For five years running, the AXA Future Risk Report has named climate change humanity’s top risk. And as it intensifies, weather-related disasters increase in frequency and severity – as the tragic events caused by Storm Boris and Hurricanes Helene and Milton have shown.

Traditionally, we focused on ‘peak perils’, such as earthquakes – perils which cause instant devastation. Historically that has made sense. But the threat today comes not from those peak perils, but from the secondary perils downstream of them: wildfires, floods, severe storms. To call these ‘secondary’ isn’t to diminish their impact. Though smaller in scope, they strike more often. Taken together, they have an impact that rivals that of their ‘peak’ counterparts.

So the industry is at a crossroads. Insurers must recalibrate their strategies to make sure that coverage remains both accessible and affordable for the millions of businesses and homeowners at risk from these secondary perils. Some in the insurance sector will doubtless fear that this new world is intrinsically uninsurable. I say otherwise. It is insurable, but embracing new technologies, geospatial technology in particular, will be necessary.

Nassim Nicholas Taleb, who writes about randomness, probability, complexity and uncertainty, imagines a group of people who were able to buy shares in a turkey at the start of the year. Throughout the year, the turkey gets fatter and fatter. The shareholders are delighted. But when Thanksgiving comes, that turkey gets the chop. The moral? The past can be a good guide to the future, but it isn’t always the case. And that’s where we find ourselves now, with the climate crisis. Climate change has aggravated floods, wildfires, storms and other extreme weather events, making them both erratic and localised.

In lieu of the forecasting methods of the past, insurers today need precise insights tailored to specific locations. And that’s the magic of geospatial technology. It offers real-time risk assessments by drawing on the terabytes of data captured by satellites. It’s objective, timely and precise. It’s transformative. High-resolution images and the machine learning technology that interprets them allow for pinpoint risk mapping — whether it’s a house near a floodplain, a factory in a fire-prone area, or a business park at risk of wind damage.

Thanks to geospatial data, we can predict disaster and help to prevent it. In wildfire-prone regions, for example, geospatial data is used to model the likelihood and spread of wildfires, using wind, vegetation, and other factors to accurately evaluate the risk. Businesses can receive early warnings, enabling them to clear brush, fortify buildings, or evacuate if necessary.

Geospatial data also refines coverage by aligning premiums with actual, rather than likely, risk. This keeps protection fair and affordable. If the risk is low, or if the insurance holder acts to lower their risk, then the insurer can adjust that person’s premium dynamically. It’s a win-win. It’s no surprise, then, that geospatial data is being used more and more worldwide.

The climate crisis is a major problem, perhaps an existential one. Around the world, people are working hard on solutions. While they do, we need to consider resilience – how we can fortify societies so that if disaster strikes, the damage done is minimal and the work of rebuilding is straightforward. This is no easy task, but thanks to geospatial data, we’re getting better and better at it. In the future, perhaps the near future, we’ll see personalised disaster alerts, guidance on protective measures, dynamic modelling, and increasing cross-sector, as well as public-private collaboration. That amounts to more robust societies for which disaster damage, financial and human, is as low as it can be. Insurers have a key part to play in this evolution.

by Pierre du Rostu, CEO of AXA Digital Commercial Platform