IRS explores HVH technology to improve appeals outcomes
During Fiscal Year 2018, the United States Internal Revenue Service (IRS) collected nearly $3.5 trillion, processed more than 250 million tax returns and other forms, and issued almost $464 billion in tax refunds. That same year, the IRS had 92,430 appeals on the books, which translates to a significant amount of outstanding revenue.
In a review of processes and procedures and considering how it might be more efficient and accurate regarding appeals, the IRS turned to HVH Precision Analytics and partner Perspecta for help.
Apply predictive analytics to IRS tax returns to:
- Determine which cases are most likely to go to appeals.
- Improve efficiency by predicting the number of hours IRS staff will spend on an appeal, in relation to the revenue the IRS will make on the return. The IRS may decide it is more efficient to follow through with a course of action other than an appeal.
- Determine which appeals are most likely to be successfully audited, resulting in revenue to the IRS.
HVH used IRS historical data on appeals and outcomes, including extensive tax return data, the payer type of entity, and whether or not a return was appealed. Data also included information about appeals, the appeal process, number of hours IRS staff spent on each appeal, and the outcome of each appeal. With that data, HVH applied our advanced analytics platform and machine learning expertise to build models to predict which future appeals are most likely to be successful.
HVH will provide the IRS with actionable insight it can use to:
- Improve efficiency of the appeals processes and outcomes
- Determine which appeals cases are most likely to result in revenue to the IRS
- Make decisions regarding whether to send a case to appeals, or to consider another course of action, which may require fewer staff hours and result in a better outcome