The Seven Things Most Insurers Don't Know About Hazard Data

| No Comments | No TrackBacks
iStock_000000743223Small.jpg

Unless you spend everyday thinking about, creating, and updating hazard data there's probably a lot you don't know about it.  We were asked what we thought the seven top things that insurers didn't know about hazard data were.  Here's our list:

 

1)     There's a cost to not using the data.  Most hazard data is sourced on a per transaction basis or on annual license based on anticipated transaction volume.  In other words, data costs money.  But that cost is minimal compared to taking on unnecessary risk or underpricing a policy.  Very few underwriting departments examine the cost of making a bad decision to save a dollar or two on hazard data.  The thought process is almost always, what's this report going to cost?  Not, what's the cost of writing a bad policy?

 

2)     Hazard data changes.  The hazard data that you used last year to quote and bind a policy may not be the same this year.  Hazard databases are evolutionary not static.  Some change annually, some quarterly, and some even monthly.  Events that happen or didn't happen this month, quarter, year, will affect the update of the hazard data and can change the scores of your models.  For example, a wildfire will take a high risk area to a low risk area literally overnight.  It may take three to five years or longer for the vegetation to grow back to make it a high scoring area again.

 

3)     The more hazard data the better.  Typically companies only look at the hazard variables they are interested in.  And that makes perfect sense.  It's also a little short sighted.  By acquiring hazard data beyond the basic needs an insurance company can start building up a history of scores associated with a property and find out if other variables are predictive for them and understanding the likelihood of a claim being filed.    

 

4)     Hazard data should be used to educate the consumer.  You may recall the Farmer's commercials that showed how to prevent claims.  They were humorously teaching consumers how to mitigate risk.  Underwriting and Claims should be leading the charge on mitigation.  Hazard data reports provided to the consumer can go a long way in educating the policyholder and reducing claims.  As odd as this may seem, very few consumers know their true hazard exposure.

 

5)     Exposure analysis and underwriting should work hand in hand.  Some people look at the big picture, some people look at the small picture.  Macro versus micro.  But the underlying hazard data to look at both pictures should be the same.  When you have differences in data, or versions of data, different departments can be sending different messages throughout the company.

 

6)     Using hazard data does not have to be hard.  Through the use of LandVision Insurance and APIs, as well as geospatial files, data delivery can be tailored to meet your specific needs.  Companies can also vary their hazard data delivery by department, with some departments receiving geospatial files, some reports, and some the API.  The easier it is to us hazard data the more likely it is to get used.

 

7)     The more you use hazard data the better your loss ratios become.  Knowledge is power according to Sir Francis Bacon.  And the more you know, the more you know when it comes to accurately writing, quoting and binding a policy.  Greater knowledge at the beginning of the policy process equates to better policies written which correlates to better loss ratios.

 

If you came up with some other answers, please send them our way.  Clearly there are more than seven things that aren't typically known about hazard data, but we stopped at seven.  But without getting into all of the technical details of sourcing, update schedules, modeling techniques, etc., the key point is, if you have access to hazard data, use it.  If you don't have access, get it.  And that the more data you have the better your decisions will be.

 

No TrackBacks

TrackBack URL: http://blog.melissadata.com/mt-tb.cgi/485

Leave a comment

Categories