IRS Key Stats
Department: Business Intelligence
Region: North America
language process developed to reduce thousands of labor hours spent finding and fixing contract description errors
transparency in the description and accuracy of procurement contracts
of concept in days
Improving data transparency in government through the DATA Act
Enabling greater data operations for the IRS through innovation
The Internal Revenue Service (IRS) is a sub-agency of the Treasury, whose primary functions are tax collection and tax code administration. In support of its mission, the IRS contracts with the private sector to buy services and products. In 2019, the IRS launched a program to identify innovative approaches to complying with the DATA Act. The program, known as Pilot IRS – DATA Act Improvements, has the following goals:
- Improving federal procurement data which resides in the Federal Procurement Data System – Next Generation (FPDS-NG)
- Limiting the amount of manual work required by government personnel
- Achieving incremental improvement in IRS data in the near term
The IRS faces increasing pressure to achieve more with less. Over the past five years, the U.S. government has seen a 23% compounded annual growth rate in contract volume. As the amount of work increases, the IRS has also seen a 40% decline in its workforce. In addition to maintaining its current workload, the IRS is faced with correcting contract errors to ensure compliance with the DATA Act and other policies. Finally, many of these procurement processes are performed manually across various IRS systems. Although this story is about the IRS, the Federal Government faces similar challenges across its more than four hundred agencies.
Like most large enterprises facing the challenges of legacy systems and processes, the IRS wanted to explore new and innovative ways to improve its data and operations. From extensive work in both public and private organizations dealing with similar procurement compliance issues, we know that the average organization loses approximately 7 hours per data worker in productivity due to manual legacy data processes.
A daunting process for federal analysts
Federal procurement analysts are tasked with drafting contracts, which can be a grueling process. They must gather inputs from multiple systems and stakeholders while maintaining compliance with various federal acquisition regulations. As they dig through emails and folders for information to complete multiple pages and hundreds of data fields, they often find themselves trying to interpret vague, incorrect, and incomplete data. This can include whether an address is up to date, if an award description is meaningful or not, or if contract dates are accurate.
In some cases, analysts can perform reference checks with third-party websites, while in other cases, they send emails, make phone calls, and schedule meetings to confirm the data they need. In all cases, they try to make the best decision possible with the limited information they have.
As analysts validate the new information they have gathered, they can only hope that all the information they’ve received from upstream processes have been adequately validated. As their long day comes to an end, they notice the large backlog of new contracts that just arrived in their queue, ready for tomorrow. They can think of dozens of ways to improve this process, but they believe it will require software development and data engineering resources. It just feels like it will be too complicated and take too long before they can see any meaningful improvement.
An all-encompassing solution to accomplish multiple tasks
ResonantLogic, an Alteryx partner, developed a proof of concept (POC) to address this problem under the Pilot IRS program. ResonantLogic used Alteryx Designer and Alteryx Server to develop a working proof of concept for the IRS in less than 20 days. The solution used the following methods to:
- Alteryx Designer was used for ETL processes for federal procurement data and Alteryx Server was used to schedule this activity to ensure the latest information was always available for users.
Verify and Validate
- The system translates validation rules into an easy-to-use workflow. This lets users build and test algorithms without the need for complex code. And because Alteryx makes managing data workflows easy, users can simply update the workflows as conditions and validation rules change over time.
- The system also uses natural language processing to interpret the meaningfulness of user-generated Award Descriptions.
- The system identifies errors and exceptions and intelligently suggests corrected data.
- The system automatically performs reference checks and provides corrected values using third-party sources such as Dun & Bradstreet and U.S. Postal Services.
- The system uses a machine learning model that can be trained by users to determine whether a user-generated award description is meaningful. The system allows users to update a training library and view the results of analyzed award descriptions.
- The system creates a report that lists the contract awards, the data fields with errant values, and the recommended values.
- The system can provide custom reports to identify defect trends that can be used to support user training and other continuous improvement activities.
3 Reasons the IRS Chose Alteryx:
More Great Content
Clearer Skies Ahead: Delta Air Lines Enables Technicians with Alteryx Automation
Delta Air Lines future proofs aircraft maintenance to avoid long-term grounding time with analytics automation
Armor Express Uses Predictive Supply Chain Analytics to Help Protect More Lives
Armor Express took predictive modeling to the next level with real-time data feeds that inform supply chains—saving $500k through inventory optimization.
McLaren Racing fast tracks data analytics in the race to accelerate
McLaren Formula 1 team consolidates 11.8 billion data-points to maximize race performance.