In his October Cabinet meeting President Donald Trump directed agency secretaries to develop plans for an across-the-board 5 percent cut of their respective budgets.
Even with the uncertainty around when Trump’s call for cuts will actually be executed, the order forces federal planners to immediately look for opportunities where unnecessary low-value activities can be eliminated or digitized to trim agency budgets. Regardless of what the final appropriated budgets may be, these budget decisions, rooted in the President’s Management Agenda, will become part of the annual congressional appropriations battle next year.
The challenge to cut costs without reducing services always creates uncertainty during the planning process. Agency secretaries can address this daunting task by looking at a few of their forward-thinking colleagues who are already reducing costs while enhancing service by utilizing robotic process automation (RPA).
The U.S. government produces, manipulates and stores tremendous amounts of data every day. The collection, analysis, and movement of routine data are performed by more than 4 million federal employees, including postal employees and military members. Agency staff move and manipulate terabytes of data – processing everything from license renewals, weather data updates, to recalculating global financial reports.
The explosion of digital data and the thirst for more data collection and reporting has caused government agencies to require more staff to perform this robotic work. More people and more data storage are driving the cost of government business higher with no measurable, positive impact on the delivery of government services.
While the president’s call for agency cuts established the goal line, a recent directive from the Office of Management and Budget provides the playbook for agencies to close the gap in achieving these budget reductions while simultaneously delivering better citizen services. The memorandum, titled “Shifting From Low-Value to High-Value Work,” reminds agencies that “each year, federal employees devote tens of thousands of hours to low-value compliance activities from rules and requirements that have built up over decades.” The memorandum also recommends agencies consider “introducing new technologies, such as robotic process automation (RPA), to reduce repetitive administrative tasks and other process-reform initiatives.”
[Twenty-eight agencies have already put their robotic toes in the water. For example, NASA’s George Washington robot is helping the agency distribute funds provided by Congress in hours instead of days. At the General Services Administration, their robot, named Truman, is supporting the agency with pre-negotiation tasks – reducing to seconds what had previously taken an hour per task.
For some agencies responding to the directive, the first hurdle is to understand what RPA is and what it can accomplish for them. RPA is software that is relatively easily trained by business staff and traditional programmers alike to mimic repetitive, transactional and rules-based tasks that consume tens of thousands of hours of agency staff time. Examples of this low-value work might include:
RPA robots helping the Department of Veterans Affairs process medical records being returned to the agency following a new law allowing veterans to seek outside medical services.
RPA robots serving as digital agents for U.S. Customs and Border Protection, to process routine customs documents, allowing agents to shift to higher-value tasks like inspections and interacting with citizens.
The PMA along with the direction of the OMB memo will encourage continued expansion of RPA by early adopters such as GSA, NASA, Treasury, the Defense Logistics Agency, the Defense Information Systems Agency and the National Science Foundation. Lessons learned by these early adopters can serve as an opportunity to accelerate RPA implementation and the broader use of artificial intelligence. Some of those lessons learned include:
Get early wins by picking the right pilot projects first. When you select your first RPA projects, these are some of the criteria you should look for:
A task performed by five-to-10 people. You will get more efficiencies gained with more than just one or two beneficiaries of the robots eliminating their mundane tasks. And save the 500-person task robot for your second robot, after you have lessons learned by going through your first RPA pilot process.
A task that is thoroughly documented, and if a good candidate task isn’t documented already, write down the process at the outset.
A medium-complexity task. Don’t pick a process that is burdened by 50-plus business rules and only used by five staffers. A medium-complexity task will allow you to show success in the six to eight weeks needed to develop, test and deploy the robot. In a few months you’ll be saving staff hours and talking about your next deployment.
Learning the ropes on a low-risk, high-reward project helps both staff and management to get comfortable with the RPA concept.
Involve the chief information officer early. RPA is a business tool operated at the employee-level. But even though it is deployed “at-the-desk” it is still important for CIO involvement and oversight of the RPA work to provide virtual platforms and assure security worthiness. Partner early with the CIO for these foundational services.
Learn more by sharing. GSA and the Defense Department are forming vendor-agnostic internal teams of RPA thought leaders to discuss RPA best practices and to develop playbooks to accelerate the use of RPA. These groups get smart on RPA faster by sharing what they know, which results in more efficient and successful deployment of their RPA program.
Additionally, the incubation period for initial adopters is ending. The growth of these agency RPA projects should begin to scale on par with those experienced in the private sector. Employing attended and unattended robots, agencies will benefit from reduced processing times, eliminated backlogs, greater standardization of data and tighter cybersecurity. The studies by HFS and Deloitte suggest the federal government could save as much as $40-to-$80 billion if artificial intelligence is properly implemented across the government.
An on-demand, versatile, and smart workforce of virtual employees offers to democratize how citizens consume government services. Using the latest technology, we can create a government workforce that is able to adapt and change to the ever-growing demands of a digital-friendly citizenry while reducing costs.
The PMA and the OMB memo enable Cabinet agencies to adopt an automation first posture. By accelerating everything agencies can adopt and expand the use of RPA to more rapidly achieve the goals of the president’s five percent across-the-board reductions while improving government services.
Jim Walker is director of public sector marketing for UiPath and the former RPA Lead at NASA shared services in Stennis, Mississippi.