Inferential Designs

Managed - Gross-To-Net Our Predictive Modeling Solution was developed to address the complex nature of forecasting Gross-To-Net for Contract Sales, Discounts, and Net Retained values. The model features extensive reporting and analytics, guiding the user to the best answer(s), and yielding real-time intelligence to Finance and other stakeholder organizations including Legal, Audit, National Accounts, and Marketing. Each forecast contains 3 years of analysis: Prior Year, Current Year, and Next Year (which is the basis for next year’s plan), with comparisons to Prior Year, Prior Estimate, and Plan, and Next Year vs Current Year. The application also contains a New Product Forecasting Module that converts Scripts to Billing Units to Catalog Sales (given NDC Mix, Daily Average Consumption (DACON), NRX’s, and estimated Patient Retention Rates). The model is designed specifically for small to mid-size marketers of branded pharmaceuticals that offer discounts to certain channels of trade where prices are determined by government mandate or mutual agreement. The application is web-based and built on the .NET Framework using ASP.NET Core MVC. Some of the model's features are as follows:

• Standardization of Forecast Methods - With the exception of Billing Units all forecast ingredients (Wholesale Aquisition Costs (WAC for Sales and WAC for Rebates), discount rates, and government pricing components, etc.) have been updated to the database prior to generating a final forecast. Therefore, the selection of a Forecast Method for Billing Units is all that remains to generate a full 3 year projection with comparisons).
  • Trend – For use when statistical ‘r’ indicates a strong relationship between Actual and Trend Billing Units.
  • Average – For use in various scenarios.
  • Weighted Change – For use when the more recent Actual Billing Unit levels are more likely to have greater impact on future utilization than earlier periods.
  • User Input – Total customization, or a modification of any of the above methods.
• Inclusion of Forensic and Forward-Looking Analytics
  • Reconciliation of Accrual 'True Up' variances allocated to Volume and Price, by Customer/Channel and Brand. This provides the analyst the opportunity to better understand prior Accrual Variances and adjust accordingly going forward.
  • Numerous other analytics at various levels of reporting, designed and included to provide users with rapid and meaningful support and answers to what can be difficult questions. In addition, all Analytics Tables are sortable by any column and can be exported to Excel.
• The model is capable of maintaining multiple Pricing and/or Period Scenarios for any given Reporting Period.
  • Because it's possible to have segregated like forecasts within the same reporting period, the analyst has the ability to perform 'Sensitivity' or 'What If' type comparative analysis. For example, the user may wish to compare next year's impact on Medicaid Rebates of an unplanned 6% catalog price increase versus the planned 4% increase.
• The model requires that Customers be designated as Key or Non-Key. This designation significantly speeds the process of creating a full quarterly forecast.
  • Key Customers require an Annual Plan and Quarterly Forecasts.
  • Non-Key Customers are consolidated by Channel eliminating the need to create a forecast for every Customer.
• The model assumes a one month timing difference between when product is sold into the trade and when the actual 'Pull-Through' occurs.
  • This more accurately depicts weighted average WAC for the calculation of Catalog Sales (i.e. adjusted WAC x Billable Units). Note, WAC for Rebates requires no such adjustment.