ReportGem ReportGem 中文

GLOBAL RESEARCH ARCHIVE

MCO: MCO Exceedance Conference Takeaways: AI is Leading to Greater Data Consumption and Product Innovations

Published: 2026-06-04Institution: Wells Fargo Securities, LLCCompany / ticker: MCO.NPages: 8Original language: 英语Evidence page: 3

Research evidence excerpt

MCO: MCO Exceedance Conference Takeaways: AI is Leading to Greater Data Consumption and Product Innovations

Moody's Corporation Equity Research

Exhibit 1 - Insurance represented 19% of total FY25 MA ARR Exhibit 2 - Insurance ARR growth stepped up significantly in

14%

16.0%

26%

14.0%

19%

12.0%

10.0%

12% 8.0% 29%

6.0%

Banking ARR Insurance ARR KYC ARR

Research and Insights ARR Data & Information ARR

4.0%

2Q23 3Q23 4Q23 1Q24 2Q24 3Q24 4Q24 1Q25 2Q25 3Q25 4Q25 1Q26

Source: Company reports; Wells Fargo Securities, LLC

MCO's Insurance segment has been largely built by acquisitions.

MCO purchased RMS, a leading provider of catastrophe (CAT) models, for ~$2bn in 2021. These models, developed by teams of PhD's and

other experts, are designed to predict the occurrence, severity, and ultimately financial losses for the insurance industry from catastrophic

events like hurricanes, tornadoes, earthquakes, floods, hail, wildfires, and others. These models utilize troves of data from a variety of

sources, including contributory insurance claims data, imagery captured from satellites and low-flying planes, purchased real estate data,

and extensive historical loss cost data and projections produced from the models. MCO also has agreements with other vendors, such as

colleges and universities, to capture wind speeds in different regions of the US. Some of this data, including prior insurance claims data, has

been collected over the last ~30yrs, which we believe is difficult to obtain elsewhere. We see CAT models as largely defensible against AI

given the expertise and data required to create them.

We see MCO and VRSK's CAT models as the industry standard. They primarily compete on the accuracy of the models, as well as their

compute power and level of definition.

The English excerpt is extracted automatically from the cited source page and may contain layout or recognition errors. It is never batch translated.

Open report viewer