How to Shape the Social Safety Net

Lessons on integrating user perspectives from Code for America Summit
Terri Ricks of Louisiana’s Department of Child and Family Services speaks about the state’s Integrated Benefits Initiative pilot.

One in five Americans gets their healthcare through Medicaid. One in eight Americans use SNAP benefits to pay for their food. Every year, billions of dollars of assistance for those people are distributed through various safety net benefits programs—and an equally significant amount of money goes into the technology behind them. But despite many best intentions, the reality of our safety net system is that it often forgets to consider the actual needs of the people it is trying to serve. Our work as part of the Integrated Benefits Initiative is focused on creating proof points that a human-centered safety net is possible, and at this year’s Code for America Summit several of our partners took the mainstage to share their perspectives on the importance of keeping clients at the center.

“We believe we must create a safety net that is human-centered, equitable, responsive, and ultimately helps the people in need.”

Stacy Dean from the Center on Budget and Policy Priorities spoke about the issues with different benefits services “staying in their lane.” People who are eligible for one benefits program are often eligible for another, but while each agency may be asking how they can make it easier for a client to access their services, they rarely ask how they can make it easier for a client to access all the services they’re eligible for. Sprawling IT systems compound the problem, though there is a real appetite from states to make improvements in service delivery and technology that serve people more responsively and holistically. And some states are taking small steps that have a big impact.

“Staying in your lane really limits how states envision what the safety net can do…
It’s not always clear within a state who is steering the ship on technology and service delivery.”

Nearly 48% of the population of Louisiana is using a safety net service, or one paycheck away from needing to. Terri Ricks of Louisiana’s Department of Children and Family Services spoke about the collaboration between multiple agencies and the Integrated Benefits Initiative to map both client and staff experience in benefit service delivery in the state. Elevating user voices and tying them to churn data helped create the impetus for a new pilot—one that texts appointment reminders to clients so they don’t lose out on benefits because of procedural denials. And the early results are incredibly encouraging: the Department of Health is seeing appointment show rates for WIC clients increasing by 65%, and text reminders are now starting to be sent to clients in other benefit programs as well.

“This time we changed the conversation. We stopped speaking for our users, and instead we were going out and speaking with them.”

Over the years, Colorado has been able to develop a strong collaboration across the five state agencies that administer benefits. Nina Schwartz of the state’s Office of Information Technology and Patrick Kelly of the Department of Housing and Human Services detailed the challenges Colorado faces in the software development lifecycle, which doesn’t include a formalized way to hear and incorporate user needs. Colorado’s Integrated Benefits Initiative pilot went back to basics and started small: making it easier for clients to report life changes that might affect their benefits eligibility. But even this small scope unearthed key lessons with huge implications, like seeing how the experiences of the client and the county worker are inextricably linked.


To hear more about the Integrated Benefits Initiative’s work, watch the video or read the transcript below.


Laura Ramos:

Hi. So, there are many, many efforts across this country to help people in need of food, healthcare, or a little help covering expenses like utility bills. Programs like these provide client assistance that can make really all the difference in the world to help somebody in a difficult moment in their life.

Now, billions of dollars in benefits get distributed through these programs, and an equally significant amount of money goes into the technology behind them. Despite these many best intentions, we have a safety net system that often forgets to consider the human element, the needs of the people these programs are trying to serve. In the conversations we have had with state and policy leaders, we often get asked, how can we make this better? How can we shift our approach to consider the outcomes we all want? In this case, we believe we must create a safety net that is human-centered, equitable, responsive, and ultimately helps the people in need.

Code for America’s work, as part of the Integrated Benefits Initiative, focuses on creating these proof points that demonstrate how a human-centered, iterative approach delivering safety net benefits is possible. Tonight I’m so excited to bring on stage a few of the partners we have been working with to bring this vision to life. We have been collaborating with the state and policy leaders on pilot projects that take on specific touch points in the process clients and states go through to apply, maintain, and enroll people in benefits.

You’re going to hear three different perspectives about why it is so important to keep clients in the center of how benefit systems are designed and administered. While that can seem like an obvious goal, and I hope that to many of you it does seem obvious, it actually gets really lost in the middle of some very, very complex policy landscapes, bottom-line oriented state legislatures, and some very rigid software development life cycles.

This scenario plays out in every state in this country and hinders our ability to effectively serve people in need. Each of our presenters play a critical role in bringing these programs to life, and I hope you enjoy their stories, because I really enjoy each one of these folks coming up on stage. Together we’re all working to create the safety net that this country deserves.

With that, I’m very excited to bring on stage Stacy Dean from The Center on Budget and Policy Priorities. Thank you.

Stacy Dean:

Good morning everyone. Thanks to Laura and the whole Code for America team for having me here today. So, I’m here to share just two of the reasons that I believe states struggle to build health and human services systems that offer holistic support that’s easy to use and navigate. To be fair to our colleagues in government, and I think it’s really important to start from this place, the safety net does an extraordinary job every day for tens of millions of Americans right now.

One in five Americans gets their health coverage through the Medicaid program, one in eight Americans use SNAP or food stamp benefits to pay for their food every day. Our country is better off, because states deliver these benefits to struggling people today. But we all know that we can do better. Government can lower the cost and difficulty for individuals who need help in order to meet their basic needs, as well as offer that support in a more coordinated and holistic way.

So, what is standing in the way of that? First, there are enormous pressures and institutional norms that encourage federal, state and county officials to think only about their piece of the safety net, rather than considering how the health and human services programs, as a whole, serve families and their states or communities. That’s really true all the way up through senior leadership, including the governor, where the safety net I think is often viewed as a collection of programs and services housed in an agency umbrella, rather than a systems of supports fueled by a vision for service delivery and citizen engagement.

So, let me give you an example of how the piece-by-piece approach can play out. All pregnant women, infants, and toddlers who receive their health coverage through Medicaid automatically qualify for the WIC program. WIC provides supplemental foods and nutrition services and counselings to this group, including breastfeeding support, which is an incredibly popular program amongst eligibles. Virtually all eligible moms enroll themselves and their kids in this program. So they are both in Medicaid and in WIC. The policy is simple, if you get Medicaid you can get WIC.

But states often make these households fill out separate applications, require redocumentation of their circumstances. I’ve actually seen moms stop by the WIC clinic on their way home from the hospital after having delivered a baby in order to update their case and add the baby to the WIC program. That’s something that Medicaid lets them do over the phone from the hospital bed.

So what’s the problem here? It’s not the policy, right? Or the inability to come up with a smart, comprehensive solution. Tech could solve that. The problem, in this particular case, or in many cases, is that the people in one program, Medicaid, ask themselves the question, what can we do to make it easier for a new mom to enroll an infant in health coverage? A reasonable question. But what they didn’t ask is what can our state do to help a new mom connect her infant to all of the services that we offer?

To me, that’s a really a vivid example of where staying in your lane really limits how states envision what the safety net can do. So here’s a second challenge of bringing human-centered design into the health and human service delivery world: It’s not always clear within a state who is steering the ship on technology and service delivery.

The IT systems that support these programs are sprawling, I think is the best word I can come up with. No one owns them, end to end. They’re often old, they’re quilted together with a lot of disorganized governance. New systems and upgrades can be multi-year efforts that span several turnovers and agency leadership further undermining, I think, clear ownership, accountability, and a consistent vision for how technology should be supporting a state’s business model.

So for a variety of reasons, the technology side of service delivery often seems outside of the state’s control. I think there’s also a sense to just be candid that vendors in this space have adapted their business models to monetize this dysfunction, rather than putting forward solutions and ideas that would advance cooperation and a human-centered approach. So for some states, these and other issues just feed into a culture of keeping programs and services apart and split. Considering technology is something that happens to states, rather than something that states lead and shape.

So these are challenges, but they are not insurmountable barriers. They haven’t prevented many states from advancing significant improvements in service delivery and technology. State officials aren’t just creatures of their agency ecosystems, they also live in 21st century America. They see the technology and how it improves their own lives, and they can see the difference between what they get to enjoy and what they’re offering the people that they are trying to serve.

So there’s a real appetite among states to figure out how to bridge from their current service delivery models to a place where they’d be serving citizens more responsively and holistically and respectfully. You’re going to hear from Louisiana and Colorado shortly about small steps they’re taking that really have a big impact on the way that clients experience their programs and services. They’re asking the right questions about how can we do better for the people and build a philosophy of service delivery that helps drive ongoing improvement across the array of programs?

But states need help to make these changes and more changes. I encourage you all to consider how you might equip them with tools and skills and information like model RFP or contract language that helps them spec, procure, project manage and measure human-centered design. They honestly don’t know how to do it, and you all probably have a lot to offer them. Teach them the powerful difference between listening to clients to gain information to inform decisions, rather than listening to reinforce a preexisting plan. Demonstrate that it’s possible to test and try different approaches on a small scale to build evidence for a new direction.

Finally, I think we need to create accountability by asking states, with humility given all that they do accomplish now, questions like, will this plan change actually make it easier for a new mom or easier for you?

So, with the help of people like you in this room, I think we can help states in their quests to build better systems for their citizens. So thank you. With that, let me bring onto the stage Terri Ricks. She is the Deputy Secretary of Louisiana’s Department of Children and Families. Terri.

Terri P. Ricks:

Bienvenue. In Louisiana, our land is diverse, our ideas are diverse, and our people are diverse. From the original native American inhabitants, to the French, Spanish, Acadians, Africans. From the Yugoslavian oyster harvesters to the Sicilian strawberry farmers, nearly every nation on Earth has left his imprint on Louisiana’s culture.

I love that statement from Governor John Bel Edwards. It describes the rich culture and history of Louisiana. We’re a fun people, we’re interesting. We love football, we like fishing, we like festivals, and of course we love Mardi Gras. We really are a strong, resilient, hopeful people, in spite of, and maybe because of, some very difficult circumstances.

In Louisiana, nearly 48% of Louisiana households are using safety net services, or are one paycheck away from needing them. 29% of our population are on SNAP, 75% of Louisiana’s children are covered by Medicaid. That tells me that our clients are not some distant minority, they are our neighbors. The role of agency executives is to be honest. When we came into governing in 2016, there was a $1.9 billion deficit. The safety net program, staff and resources had been cut, and our case loads were soaring.

We also came into very differing views about the role of government in people’s lives. We’ve had to be honest about what role agencies can play in this moment. The truth is, if someone comes to the Department of Children and Family Services hungry, we are not able to feed them today, so we work with our partners like the food banks. So we had to be honest that we’re just one piece of a larger safety net. We had to be honest about our information services system challenges.

The administration has been building a statewide interdepartmental enterprise architecture. We’ve been trying to retire all of the legacy systems, and currently they’re integrating eligibility systems for Medicaid, SNAP and TANF. You can probably guess that this change in staff and support makes everything tricky.

The last thing we had to be honest with is how we treat our people. When they’re coming to us in their most vulnerable state, do they feel helped and respected? We need to really figure out how to meet people where they are. Enter Code for America. We really had to figure out before doing anything else, if Code for America was actually legit. It took us a little while, but once we did, we’ve been extremely excited to work with them. We saw it as an opportunity to further our desire to deliver client-focused services.

The engagement includes four of our agencies, the governor’s office, the Department of Children and Family Services, the Department of Health, and our office of technology services. The first component of the work, I was very excited about, it was to map the client’s and staff’s experiences relative to our service delivering. Code for America brought us their voices. We heard that the SNAP program is really important to families, we heard that many SNAP families have to start over again in six months. We heard that the clients really do want to help people in a really special way.

When I think about a single mom of three saying these words right after saying, “I can’t get by without these benefits, but I have to start over every six months.” It helped me to realize we need to fix this. We heard the voices, we reviewed the data. In one month, more than 12,000 cases were closed because of process reasons. Someone missed an appointment, they failed to give us a document that we needed. Loss of benefits causes household instability, frustration, and a lot of anxiety for those who are already in hardship, they’re already hurting. The other thing is that it’s seriously inefficient for staff.

This is churn. Churn is not new, but when we heard the voices and we put the data together, we realize that we need to fix this. With Code for America’s help, what we realized is there may be an actual new approach to this, that we can fix this. States do need to test out new ways to understand what the client experiences and perspectives are. This engagement allowed Louisiana to recognize our client’s phones as their lifelines.

Everybody’s texting. You guys are texting, some of you are texting now. My kids are texting, our clients are texting. I needed a reminder to get here today. Did you need a reminder to get here today? Our state steering committee decided on the “LA’Message” pilot. The idea is that timely specific text reminders will help clients maintain their much needed benefits and reduce churn. I have to tell you the early results are promising.

For the Department of Health’s WIC clients in the pilot, we’re seeing appointment show rates increasing by 65%. Clients are also proactively rescheduling their appointments at an almost 25% rate. Can you imagine the impact this is having on the clinics managing time and resources?

Well, I’m very excited, because this month some of our SNAP and TANF clients started receiving also reminders through text. We believe the text reminders will really work. The WIC data gives us a lot of hope that this simple reminder could grant some stability to a household that might otherwise lose benefits, and cut down on the casework and reduce the churn.

DCFS employees, I believe, are the best in the business. They are passionate and compassionate. Those that have been a part of the pilot are super excited about what they think it will bring to their mission. They want to improve client’s lives, they want better results, they want to treat the clients with dignity and respect. We speak in our administration about a breeze of hope driving change as mighty as the Mississippi. We are a state government as hopeful as our people are. That’s what they deserve.

We are glad that the breeze blew in Code for America to help us do better by the state. We’re so excited about you guys being a part of it too. Come join us in Louisiana. I’d like to thank you for hearing me today, and I’d also like to ask you to help me introduce my friends from Colorado, Nina and Patrick. Thank you.

Patrick Kelly:

Thank you Terry. The rocky mountains tend to be what people think of when they hear the word Colorado, and they may have a misconceived notion that we’re all the same. But our diversity is much more complex than just whether we are a snowboarder or a hiker in those mountains. There are five state agencies adding to our complexity, as well as a state run healthcare exchange that work together with 64 counties and the governor’s Office of Information and Technology that build our ecosystem of technology to serve our users.

Over the years with much hard work and a little bit of heartache, we’ve been able to develop a strong collaboration amongst all of these agencies. Our goal is to help those people who are seeking assistance from one of our programs. We’re very proud of the fact that we’ve been able to collaborate across agencies.

However, we work everyday to improve the delivery assistance for our clients. However, we do have a huge gap, and that gap is in our software development life cycle, where it takes us roughly six to nine months to make changes to our system. But more importantly, within this structure there’s no formalized way for us to uncover and incorporate the user needs. As a result, we end up doing this and ad hoc ways. We squeeze it in when timing allows, and generally, timing doesn’t always allow for it.

PEAK is Colorado’s client-facing digital tool, where over 1.2 million eligible Coloradans can apply, manage and renew their participation in over 11 programs. We’ve worked very hard to ensure that every agency and program had a say in how the systems worked, all with the intention of serving our clients well. We’re not talking just about the 80% of our state’s population living in the urban corridor along the front range of the Rockies, but also that 20% that are scattered across the rest of our state.

We had the best intentions, but the site we developed isn’t always as simple to use as we’d hoped. We created an elaborate system that attempts to meet all of our regulatory and policy requirements for all of those programs. As a result, our system isn’t always as intuitive as it should be. Mainly because we didn’t ask our clients up front and on an ongoing basis whether or not we were meeting their needs.

We knew we were at a crucial point in PEAK’s development, but didn’t know how best to tackle the issues. We had the commitment to improve the user experience, but we didn’t have a solid plan. As part of the Integrated Benefits Initiative, we have a unique opportunity to rethink exactly how we build our technology. As a neutral third party, Code for America has helped us to assess our needs and prioritize our action steps. They have also continuously guided us on strategies to keep putting our client’s needs first.

Nina Schwartz:

Our collaboration with Code for America has been focused by the guiding question, what is the client experience applying for and maintaining public benefits in Colorado? We have asked this question before, but this time, the way we answered it was different. This time we changed the conversation. We stopped speaking for our users, and instead we were going out and speaking with them.

We asked users of all ages, from all parts of the state, with many different life circumstances, what’s working for you? What are the challenges you are encountering? We had honest conversations with our clients, as well as county frontline workers who engage with our clients every day. This is whom we serve, this is whom PEAK has to work for. One of our key lessons with seeing how the experiences of the client and the county worker are inextricably linked.

We are starting small. One example is improving how clients report a change in their life and making it easier for county workers to process those changes. We could have started by trying to redesign everything, but instead decided to start small by redesigning the output PEAK generates, the receipt in PDF form of the change a client reported. As we’ve gone about redesigning this PDF, we have stayed focused on meeting clients and county workers needs. By continuously getting feedback from county staff and clients, we have been able to arrive at a new design.

Here’s what a prototype of the new PDF looks like. This new PDF meets the following needs. It highlights what client information has changed, only displays the relevant information, and clearly reflects the client’s intentions. This may seem like a small change, but for us it’s tangible proof that you can start with user needs. We also anticipate that this will immediately make it easier to process thousands of the changes we receive every month.

In addition to starting small, we are going back to basics. If we want a user-driven system, we know there are a few things we have to do. Here are our new basics. We have to know and align around the outcomes we want. To achieve those outcomes, we have to develop product roadmaps that will map out the vision and direction. This means meticulous, ruthless prioritization.

We have to always ask ourselves, what are the user needs we’re trying to solve? And, we have to look at our development life cycle and think about how we can always incorporate user research and feedback in every technology project. This is hard work, but we are committed. We’re excited, and we’ve cultivated great partnerships.

There is strong representation from Colorado here at Summit. We’re energized about this work, and we’d love to talk with you more. We work for the people of Colorado. So, if we’re going to serve them, we have to know what they want and need. Thank you.

Related stories