Behavior is a System Outcome

The BI Beat is a monthly column written by the Behavioral Insights unit, featuring insights, practical examples, and interviews from the field. Behavioral Insights is funded by the U.S. Department of Labor (USDOL).

A claimant forgets to report earnings. Another documents their work search incorrectly. Someone else misses a deadline because they misunderstood a notice.

These situations are often treated as individual mistakes. The response is usually to explain the rules again, add another reminder, or strengthen enforcement.

But what if these behaviors are telling us something about the system rather than the people using it?

That is where systems thinking and behavioral science become a powerful combination.

From Changing People to Changing Systems

Behavioral science has traditionally focused on understanding why people make the decisions they do. It explains how cognitive load, present bias, friction, and limited attention shape behavior, particularly during stressful moments. These insights have helped improve public programs through clearer communications, better forms, timely reminders, and simpler processes.

Systems thinking expands that perspective. Instead of asking why someone made a mistake, it asks what features of the system made that mistake more likely. It recognizes that behavior is rarely the product of one decision. It emerges from the interaction between policies, technology, communications, organizational processes, and the environments in which people make decisions.

Together, these disciplines shift the conversation from changing people to improving the systems that shape behavior.

A Different Way to Look at UI

This shift is especially relevant for unemployment insurance (UI).

Consider a worker who has recently lost their job. They are suddenly navigating unfamiliar terminology, multiple online systems, requests for documentation, work search requirements, and weekly certifications, all while dealing with financial uncertainty and the emotional impact of unemployment.

If they make an error, it is tempting to conclude that they failed to follow instructions.

A systems perspective asks a different question: Where did the system make success harder than it needed to be?

Following the Error Backward

Take work search reporting as an example.

A traditional approach might conclude that claimants need more education about the requirements.

Behavioral science might find that claimants experience cognitive overload, misunderstand what counts as a valid work search activity, or postpone reporting because immediate priorities take precedence over administrative tasks.

Systems thinking takes the analysis one step further. Why are claimants learning about work search from multiple sources that use different language? Why are examples buried in lengthy handbooks instead of appearing when claimants complete their weekly certification? Why does the system identify reporting errors after benefits have been paid instead of helping prevent mistakes while information is being entered?

These questions move beyond the individual and examine the environment producing the behavior.

This is an important distinction because most integrity issues do not originate from a single interaction. They emerge from a series of connected experiences. A confusing webpage leads to an incorrect assumption. An unclear notice reinforces that misunderstanding. A certification form provides little guidance at the moment a decision is required. Weeks later, the agency discovers an error that could have been prevented much earlier.

Every touchpoint contributes to the outcome.

Designing Systems That Support Better Decisions

This is where systems thinking and behavioral science become especially powerful together.

Systems thinking helps agencies map the claimant journey and identify where complexity, duplication, or poor coordination create unnecessary barriers. Behavioral science explains how those barriers influence decision-making and helps identify opportunities to reduce friction, simplify choices, improve timing, and provide guidance when it matters most.

Often, the solutions are surprisingly small. Embedding examples into an online form, using consistent language across communications, eliminating duplicate requests for information, or providing immediate feedback during certification can influence thousands of claimant interactions. These changes improve behavior not because people suddenly try harder, but because the system makes the desired behavior easier.

This represents an important evolution for behavioral science. Rather than viewing interventions as isolated nudges, we can view the entire claimant experience as an opportunity for behavioral design. Communications, policies, technology, operational workflows, and staff interactions all become part of the intervention.

The System Is the Intervention

Every communication, policy, digital tool, and operational process influences behavior. Whether intentional or not, every system nudges people toward certain decisions and away from others. The question is not whether a system shapes behavior—it already does. The question is whether it has been designed with human behavior in mind.

Behavior is often treated as the problem to solve. But behavior is also feedback. Every misunderstanding, missed deadline, incorrect work search report, or preventable overpayment tells us something about the system that produced it.

If we want different outcomes, we cannot focus solely on changing individual behavior. We also have to redesign the environment in which that behavior occurs.

When agencies combine systems thinking with behavioral science, they move beyond correcting errors after they happen. They begin building systems that help people succeed from the start. That is where lasting improvements in program integrity, administrative efficiency, and claimant experience begin.