Blog article
See all stories »

A Commercial Underwriter’s Guide to Automation, Part 2

a member-uploaded image

Read Previous Blog, Part 1 - Insurance underwriting is a critical differentiator between the leaders and the laggards when one looks at it from an operational performance lens. We’ve drafted this 2nd part for Underwriters who are ready to disrupt, evolve, and drive significant tangible impact across their organization.

Building superior Customer eXperiences

For most Insurers, the key question they ask while making decisions is, ‘what’s the value in it for the Customers?

There are three ways in which brokers suffer during the application process:

  • Delays in acknowledgments for their applications and back and forth communication that’s done to fix that.
  • Rejection of legitimate and fair applications while manually clearing applications.
  • Incorrect pricing of applications that leads to unfairly higher premiums for Customers who should receive credits due to exposure presented.

Fixing the underwriting efficiency problem

The underwriting decision depends on several variables that an AI model can ingest to drive the underwriter’s efficiency.

Finally, we discuss how an underwriter’s efficiency go down due to various reasons. Any manual effort that can be automated affects these KPIs. Underwriters are not well-versed with disparate document types, which demands them to spend more time spotting the sifting through data inefficiently. Also, it’s not easy for them to track applications coming from multiple sources and systems. There’s no smart way to prioritise applications – case in point, an underwriter must prioritise a application from a preferred broker whose policy is expiring in a month over another application that scores low on all these parameters.

a member-uploaded image

Efficient data management for optimize costs

An easy and impactful way to optimize cost is by reconsidering how one runs data management, especially data consolidation and data hygiene.

There are three ways to execute this strategy:

  1. Automating application intake.
  2. Consolidating data for the underwriter to get a comprehensive view.
  3. Accelerate underwriting cycle, and flag critical data (e.g., prior loss information, previous carrier information) that’s absent from the broker application.

If the insurance firm can fix these data management issues upfront, these will yield significant gains in terms of saving underwriting costs.

Machine learning modeling to improve profitability

If the underwriter can effectively assess the risk present in a application, the likelihood of loss goes down significantly. While this is easier said than done, the least insurers can do is to intelligently flag the underwriters of any attributes that accompany a application. The underwriters can go beyond this and leverage intelligent automation through AI, ML, and process automation to improve their decisions.

By developing a scoring mechanism for all applications, the underwriter can bring in fundamental changes to their operations that will result in higher clearances, lower costs, and improved margins.

a member-uploaded image

Automated workflows to simplify Customer eXperience

For driving superior Customer eXperiences for the brokers, insurers need to work on two aspects: encouraging regular and proactive communication with the brokers and incentivising customers with fair premiums as much as possible.

Successful insurers and underwriters automate their email workflows for communications related to pre-defined events like an acknowledgment, requests for additional or missing information, and updates on approval or rejection of applications.

An insurer can benefit from building AI and machine learning-based models to make decisions and reduce the probability of unfair decisions. With every decision that the model churns out, it incrementally improves the accuracy and precision of decision making. E.g., compare two cases of P&C insurance – a application covering a property built in 1980 versus another property that has no history of losses in the last five years. The AI model will ensure the latter is rewarded with a premium credit and thereby delivering better Customer eXperience. Underwriters will miss out on picking these nuances, especially when there’s a massive volume of applications.

Systems integration to improve efficiency and productivity

Underwriters have to deal with a humungous amount of data fed from numerous sources in different shapes and forms. For instance, it might be stored in different file types, can be structured or unstructured, etc. It takes a village to converge all of this data into meaningful information that supports the underwriting decision. In such cases, a smart way to improve the process efficiency and underwriter’s productivity is to automate the data extraction process.

An automated system will free up underwriter’s bandwidth from high bandwidth – low impact activities like manual data extraction, collation, validation, and decision making.

An intelligent automation based system can also save underwriter’s bandwidth by pooling in data from multiple sources and exchange information with numerous systems at-scale. Let’s consider a few examples

a member-uploaded image

Integration with rating and policy administration systems ensures data integrity and consistency across the organization and provides one Customer view.

  • Augmenting existing data with a third party source data.
  • Procuring loss history from prior carriers can bring in significant refinements into the Underwriter’s decisions.
  • Further, by setting up business workflow rules, the applications are routed to the right teams with the right set of flags for appropriate handling. An intelligent rulesbased scoring and continuous learning by the AI model direct help the underwriters prioritize applications that are in the sweet spot for the carrier.

All of these enhancements result in drastic improvements in process efficiency and Underwriter’s productivity.

The Next and Final Blog on this Topic asks - "Why you must rethink and act now?


Comments: (0)

Member since




More from member

This post is from a series of posts in the group:

Artificial Intelligence and Financial Services

Artificial Intelligence and Financial Services

See all

Now hiring