Commercial Lines Insurance- the End of the Line for All Data

Commercial Lines Insurance- the End of the Line for All Data

Commercial Lines has more data available than ever before to understand and tune its underwriting practices

I’ve had the pleasure to participate in a few Commercial Lines insurance industry events recently and as a prior Commercial Lines insurer myself, I am thrilled with the progress the industry is making using data and analytics.

However, I do not think Commercial Lines insurance gets the credit it deserves for the industry-leading role it has played in analytics. Commercial Lines truly is an “uber industry” with respect to data. The past and present use of data in commercial lines has set the foundation for modern analytics by compiling and analyzing data across the industries supported with commercial policies.

A Long, Long Time Ago

Since the beginning of Commercial insurance as we know it today, insurers have been using data generated by other industries to assess and rate risks. In the days of Lloyd’s Coffee House, insurers gathered data about cargo, voyages, seasonal weather and the performance history of vessels and mariners to underwrite risks. The data was not created by the underwriters; it all was sourced from the insured enterprises, industry-specific sources or associated third parties. Another historic example is crop and livestock insurance in Germany in the 1700s. Such insurance focused on livestock mortality and named perils such as hail. The German underwriters analyzed historical data such as weather, location, breed, type of crop, and a farmer’s experience to assess risk, underwrite and set price exposures. In this way, the Commercial Lines segment of insurance has really been a user of big data since its inception.

Are Commercial Lines Staying Ahead or Lagging?

While Commercial Lines is an uber industry, it is also sometimes perceived as a laggard in the use of data and analytics. I often hear this at industry events and in conversations with insurers. In reality, we are way ahead in the use of data (possibly hundreds of years ahead!), but behind in our use of tools and technology to manage the data optimally to get the most value out of it. Commercial Lines insurers generally do not analyze the full range of their data, nor do they always have the right processes and technology in place to enable the collection, storage,  and analysis  to operationalize the data.  Commercial Lines could also improve how they enable all types of users to access the data in support of new business use cases.   Here’s where Commercial Lines must catch up in order to move from laggards to leaders again.

As an uber industry, Commercial Lines has more data available than ever before to understand and tune its underwriting practices.  Think of it as applying specialty insurance approaches to more business lines.  There is a wealth of data now available to make this possible.  For example, the types of data sourced from other industries that we can use in the underwriting process include:

  • Manufacturing – sensors (for quality, safety and maintenance-related)
  • Health Care – physician notes, photos, pharma records, MRI scans, telemedicine records
  • Retail – location (and associated risk), type of equipment used, inventory sensors, supply chain data, hours of operation
  • Transportation – Weather, location, aerial drone imagery, telematics

In the last few years, Commercial Insurers have been making great strides in expanding the use of their data.  The approach is very evolutionary; the initial focus tends to be aimed at cost savings and starts with structured data. This results in enhancements in finance reporting or compliance. Then there is a recognition that there is so much more that can be done with the data. The result is creative use cases that improve operational efficiency such as those around natural catastrophe management, telematics/fleet management, underwriting efficiency/document analysis, supply chain and construction insurance.

As an uber industry, Commercial Lines has more data available than ever before to understand and tune its underwriting practices. 

Recently, I learned about some really cool use cases around cargo/marine insurance that specifically underwrite parts of voyages, specific cargo, specific time periods, or ships/cargo only in a specific location. Supported by location data, weather information, sensor identified cargo and historical loss experience, commercial insurers are advancing data driven underwriting. The amount, diversity, accuracy and timeliness of data is enormously better than in the early days of Lloyds.

The tools and technology to analyze this data have advanced also of course. The capabilities exist to collect real-time data and act on it in real-time to be relevant and affect business decisions. For example, this supports underwriting and risk management in agricultural insurance, where all types of weather, location, soil, nutrition and all kinds of real time plant and livestock data improve insuring that business. Another example is fleet management. Location, driver behavior, routing, cargo and rest time data enable fine tuning and individualized underwriting.  The tools exist to interpret semi-structured and unstructured data such as images to help assess damage or compare to other similar images. And the infrastructure exists to accommodate the quantity of data and the processing speed necessary to make this possible. Cloud infrastructures – private, public or hybrid data clouds make this compute possible with flexibility and dynamic scalability.

This data and the technology combined help underwriters assess and rate risks smarter, faster and more accurately. But they are still underwriting, as they have done for hundreds of years. They were data scientists before anybody had ever heard of the term. Edward Lloyd would be proud, though the only clouds he ever knew were those in the sky.  

At Cloudera, we are all about advancing what can be done with data. Watch my short video to hear more about how we enable advancements in underwriting with data.

Monique Hesseling
More by this author

Leave a comment

Your email address will not be published. Links are not permitted in comments.