There’s no doubt that to achieve profitability, commercial insurance companies must be accurate with their pricing. Forecasting future loss per exposure is critical for insurance carriers and reinsurers for pricing, but also producers and program managers to prioritize their efforts.
In the practice of underwriting, selection and pricing require Olympic-level balance. If an insurance provider never writes any risks, they go out of business, however, if they write too many high-risk accounts without adequate premium, they become overexposed. The state of the art is to accurately forecast the frequency and severity of future losses, understand the uncertainty around these forecasts, and then use these to price insureds appropriately. Doing so brings in the right amount of revenue to accommodate the costs of claims and expenses.
But there are other factors that come into play when pricing adequately, including:
\n
- Setting premiums high enough to adequately exceed total loss, and simultaneously low enough to be competitive and retain customers
\n- Revising premiums often enough to reflect changing exposures and economic realities
\n- Identifying exposures and risky conditions that can be mitigated by risk management initiatives
\n
This is where predictive modeling comes in. Data science can be used to assess the risk profile of an insured to help set the right premium for their risk attributes. While many insurance providers have in-house actuaries or data scientists to build predictive models—Convr offers these modelers additional information, rich artificial intelligence (AI) features and risk scores to improve model accuracy. And the tools inside the Convr Underwriting Command Center bring an added layer of information and real-time insights to underwriters to directly improve the decision-making process.
\n
Sometimes using traditional methods that lean on historical loss data and policy level information—is not enough. Convr’s platform can help customers gain a competitive advantage in pricing through several available features.
To aid insurers in this level of analysis, we’ve built:
\n
Risk 360 AI™ where users can streamline their research and enhance applicant data with the power of AI. Risk 360 AI leverages Convr’s data lake—comprised of the digital footprint of millions of businesses—built with an underlying knowledge graph that unleashes detailed insights from the intersection of tens of thousands of data variables.
\n
Answers AI™ streamlines and centralizes all available information about an applicant’s business and answers underwriting questions directly through insurance trained AI models. There are two key components of d3 Answers: standardized questions to confirm or predict risk characteristics, and business classification. These answers can be included in an insurer’s pricing models and underwriting rules.
Intake AI™ eliminates manual submission processing by digitally ingesting, preparing and analyzing submissions. The product quickly extracts structured and unstructured data and applies AI to digitize data and assess accuracy. For every submission that flows through your business we extract key data points and enrich that information with third-party data to broaden and deepen the risk profile.
By automating and digitizing the insurance application process, underwriters use enhanced application data to quote faster, with more confidence. And they achieve more nuanced insights. There is no room for debate—greater accuracy and speed leads to better decisioning. And what provider does not want that?
All three of these products give the underwriter better information to modify the formulaic suggested premium. And they are better equipped to make more accurate pricing decisions—charging more or less based on the information provided through Convr.
And these capabilities apply to more than just single submissions. They can be used to evaluate entire books of business, rollover books of business and for bordereau ingestion.
So what are the benefits of better up-front risk awareness?
\n
- Greater efficiency with a focus on better risks
\n- More information to go into loss forecasting models
\n- Better understanding of the appropriate premium for each risk
\n- Providers are better able to compete for best risks
\n
When models fail portfolio management
Another way Convr’s Risk 360 AI, Answers AI and Intake AI can be applied is when providers need to update their pricing models. Actuaries and data scientists at leading insurers need to experiment and test new modeling methodologies, model frameworks, and data sources regularly. Not updating models on routine cadence can result in models getting less and less accurate over time.
This is also true for the data that feeds models that price risks. Outdated data can lead to lost profits and cause better customers (those who are less risky) to leave their provider and poorer customers (those who present greater risk) to pay insufficient premiums. Outdated and/or insufficient data linked to insureds can also result in errors in risk management decisions, jeopardizing profitability or even solvency. Accuracy and current information is key.
\n
Here is a case when you don’t know what you don’t know!
You can explore your out-of-the-box and customer modeling options at convr.com and reach out to Suzanne Vranicar at suzanne.vranicar@convr.com if you have any questions about the products and services listed.
\n
