How AI can help federal agencies make the most of limited budgets

How AI can help federal agencies make the most of limited budgets

In today’s budget environment, agency leaders are often asked to do more with less.

In today’s budget environment, agency leaders are often asked to do more with less.

Take the Department of Veteran Affairs, for example. In July of this year, the VA told the House committee that it was anticipating a $15 billion budget shortfall for the rest of this fiscal year. VA leaders say this is largely because of a need to hire more employees to help with the growing demand for VA services.

While this particular example is also deeply intertwined with the rapidly rising cost to hire a skilled healthcare workforce, the story is the same everywhere you look. In the Social Security Administration, the Senate Appropriations Committee voted to approve an approximately $500 million budget increase earlier this month for fiscal year 2025, citing the rising customer service costs associated with social security benefits. Meanwhile, the House has already proposed a nearly $500 million SSA budget cut.

The fluidity of agency budgets year to year in the current appropriations environment underscores a need for federal agencies to stretch every dollar further than ever before — especially when it comes to some of the most cost-intensive program line items like program enrollment and customer service.

This is where artificial intelligence could begin to help agency leaders do just that — although not necessarily in the ways you might think.

While generative AI chatbots have been one of the most visible tools in the AI revolution over the past few years, these tools are not without risks. When it comes to critical government resources like veteran healthcare and retirement benefits, inaccurate information or AI hallucinations are simply not an option. Not to mention, chatbots may not always fit within the experience citizens are expecting.

Instead, the best initial application for AI in federal program service delivery is actually behind the scenes, through a customer experience refinement process called conversation intelligence.

Using generative AI to analyze customer service conversations in aggregate across channels, conversation intelligence extracts deep engagement insights an agency can use to address pain points like citizen frustration, operational inefficiencies and customer service representative knowledge gaps.

More importantly, when used well, it can directly reduce workload for customer service teams – while simultaneously increasing experience satisfaction for citizens.

How conversation intelligence works

Conversation intelligence starts with terabytes of unstructured interaction data. Then — by deploying generative AI to apply algorithms for topic modeling, entity extraction, sentiment analysis, complexity and emerging topics at scale — organizations can begin to track and monitor the sources of experience inefficiency.

The results are deep insights these organizations can use to drive big changes in process and operations that target proactive ways to restructure the citizen experience to better align with citizen expectations.

Let’s take a look at a hypothetical VA example. Currently, veterans must take the initiative to locate enrollment forms and fill them out once they have verified that they are eligible for VA services. It’s not hard to imagine this process presents some opportunities for confusion as veterans move from one step to another.

However, with conversation intelligence, the VA could begin to figure out which specific friction points are the most common, as well as why those friction points occur in the first place.

If finding the form was the hardest part, for example, is there a way to proactively send the form to eligible veterans? Or, if customer service representatives tend to encounter the same set of clarifying questions about one portion of the enrollment form repeatedly, could that information be turned into an FAQ document that streamlines enrollment and saves both customer service reps and veterans from unnecessary, costly support interactions?

In the private sector, innovative process design driven by these sorts of conversation analytics are helping to right-size customer service teams and create customer experiences that are more logical to the end user.

As the Office of Management and Budget continues to charge agencies with reimagining federal service design and delivery through life experiences — such as approaching retirement, recovering from a disaster, and more — conversation intelligence offers a powerful listening strategy to better understand how citizens think about and interact with these major milestones in their lives. By beginning to think about citizen experiences from a citizen-centric perspective, new and improved processes can begin to rewrite the way agencies plan and budget for program enrollment and delivery at a time when budget concerns are becoming a critical roadblock to achieving OMB’s lofty citizen experience goals.

Aaron Mosby is vice president of digital sales, public sector at TTEC Digital.

Copyright © 2024 Federal News Network. All rights reserved. This website is not intended for users located within the European Economic Area.