Overview

As states adopt PFML programs to support workers and their families, PFML’s unique nature is becoming more apparent.

No two states are exactly alike, which gives states the flexibility to build a program that meets their constituents’ needs, but it also means there is no template to follow. While this comes with several challenges, it also means there are opportunities to stand up PFML programs quickly without an onerous technology burden, resulting in an effective and long-lasting program.

Here are six tips for designing a reliable and resilient PFML program.

Enable human-centric design

PFML isn’t just for employees’ critical life events but also those of their families.

As a result, it becomes difficult to predict who will file a claim and when. A human-centric design philosophy recognizes that users and their situations are unique, and their experiences should be tailored for the best possible outcomes.

States can incorporate this approach into the design of their user portal and ensure the functionality can be as responsive as possible. The system should be easy to navigate, with top functions and important information in prominent locations on the website. Personalized action items will help users find their next steps, locate and provide necessary information, and understand potential outcomes.

This enables government agencies to provide better service to a user base with different experiences, incomes, education, and expertise.

Ensure extensibility and maintainability

For states to dedicate the time and resources to developing a new program, it must come with some promise of longevity.

Ensuring the new system is both configurable and scalable is critical. States need the ability to make small changes or repairs without taking the whole system offline or requiring a dedicated developer. Legislation often requires tax or benefit rate adjustments over the program's life. As the program expands, factors like eligibility, covered family members, and wage replacement rates will also need to be adjusted.

A low-code/no-code platform allows state employees to make these changes in real-time without needing the work of developers. It also enables integrations with other systems for data sharing and validation, reducing data duplication, and creating a single source of truth.

Incorporate predictive analytics

As PFML programs develop, states will be asked to demonstrate how they can maintain solvency over time. 

Traditionally, this involves an actuarial analysis every three to five years, creating a model for the next period’s results. However, PFML is a new program with little historical data or insights. Additionally, given the variation from

state to state, patterns seen in one program may not necessarily transfer to another.

States also have to account for unexpected variations in use. For example, Rhode Island’s program saw significant increases in monthly claims in 2020 and 2021 that aligned with the worst months of the COVID-19 pandemic.

By using predictive analytics instead of actuarial analyses, states can better plan for short-term and long-term solvency, as well as prepare for future pandemics and crises. Predictive analytics also provides continuous input of data, allowing for deeper insights into the program over time and enabling program administrators to access valuable information whenever they need it.

States can also use a related tactic, sentiment analysis, for real-time feedback about user experience. For example, when Netflix experiences a service interruption, users will often take to social media to investigate whether the issue is server-wide or specific to their account before reaching out to customer service. Users of government programs such as PFML are increasingly doing the same.

States can use sentiment analysis to flag such social media posts, bringing service issues to their attention without waiting to receive a support inquiry. Quickly catching and fixing problems means fewer affected users and shorter wait times for assistance.

Prioritize state employees with enhanced automation

Automation can allow for faster and more accurate service delivery for claimants, and better efficiency for employees. 

Predictive analytics can support a high-level view of the program, while other forms of automation, such as robotic process automation (RPA), can streamline data processing and administrative tasks, such as claims processing.

Automation of routine tasks allows state employees to focus on more complex issues that require their expertise. For example, RPA can identify whether an employer needs to register for PFML.

• If they are liable, they can be directed to correct registration forms.

• If they aren’t, they can be referred to resources that can help them decide whether to register voluntarily.

• If the employer has questions, or faces registration issues, a state employee can step in to resolve the issue.

Automation can enhance PFML program performance in multiple areas, such as offsetting limited staffing. Real-time eligibility and liability outcomes are delivered with maximized accuracy through data verification. Standardized criteria also contribute to more predictable outcomes, making it easier to identify suspicious claims.

Given the recent rise of fraud, waste, and abuse in public benefit programs, this visibility is critical for swift resolution of any nefarious activity.

Consider a mixed staffing model

In addition to automation, states have other options for maximizing the potential of their limited resources.

Like any new initiative, standing up a PFML program requires onboarding a large number of employees in a short timeframe. Often, they are learning to work with a system that is still being developed, meaning that the trainers themselves are learning the system shortly before training others.

A mixed staffing model can overcome this challenge, wherein state employees work alongside staff from partners to manage employer registrations, assign tax rates, and process initial claims. This enables state agency staff to learn directly from the system’s creators and partner staff can act as a safety net, maintaining critical processes while state staff learns from the early stages of the program’s launch. This results in a smoother experience for users from the program’s onset, faster and more accurate outcomes, and higher satisfaction.