Valen’s AuditRight^® enables insurers to optimize the premium audit process by predicting the net return for each type of audit thereby maximizing the insurer’s use of audit resources.  

Using Valen’s AuditRight^®, insurers maximize the returns on their premium audit expense.  Valen’s data experts gather data from our customers, cleanse and organize the data, and combine that data with our ValenNetworks^® database of historical information.  Our analysts use that data to build predictive models that estimate the net return for each audit method at the insurer’s disposal.  Our models are deployed in a Software-as-a-Service approach that provides ready access to the results of sophisticated predictive analytics.  Our customers typically see a minimum of a 15% improvement in premium recoveries.

The Challenge:

Commercial lines property and casualty insurers, especially those that write workers’ compensation policies, base the premiums that they charge on payroll volume and job classification.  These two key characteristics are likely to change throughout the year and will influence the amount of premium due.  In order to verify that the correct premium is being charged, insurers rely on premium audits.  These audits take on several forms ranging from the mailing of forms to the insured company to an onsite physical audit.  The different types of audits methods produce different levels of accuracy and carry significantly different costs.  Furthermore, the audit may discover that the insurer must refund premium or charge a lower premium due to changes in the payroll.  In order to maximize the efficiency and effectiveness of the premium audit process, insurers need to predict which method will provide the greatest payoff after factoring in the cost of the audit. 

Valen's Solution:

Valen AuditRight is a web-deployed predictive model that presents the user with an estimated recovery dollar amount for each type of audit.  Using this information the auditor or underwriter selects the most appropriate audit method.  The solution includes the gathering and preparation of data, model building using the client’s historical data combined with our ValenNetworks data and the java client software that resides on the insurer’s users’ desktops.

Implementation:

When it comes to predictive modeling, most vendors focus on the model and give little consideration to how that model will ultimately integrate into the insurer's business processes.  At Valen, we recognize that the implementation of a model is equally important as its creation. Therefore, we have developed a five step process for ensuring that our customers receive full value from their AuditRight model.  Implementation is conducted in five major phases: project definition, data gathering, modeling, model deployment and model monitoring.  The modeling phase is highly dependent on the amount of historical data available.  Valen's XData^® can be used to enhance our client's historical data, filling gaps and improving the modeling process.   The model deployment phase will depend on the amount of customization required by the carrier.  Typical implementations require 7 to 12 months.