May 6, 2020 | By Robin Carnahan and Waldo Jaquith

Only 13% of major government software projects succeed, and the successful and failed ones alike cost 5–10 times more than they should. When those projects fail, so too do the public policy initiatives that depend on them: unemployment insurance, small business loans, paid family and medical leave, SNAP, Medicaid, etc.

As we’ve seen in the response to the COVID-19 crisis even if lawmakers move quickly to pass legislation to get money to laid-off workers, small businesses, and hospitals, those policies can’t be implemented effectively when the technology tools used to apply for, distribute, and track funds can’t be easily modified or don’t work.

This is an egregious state of affairs. But we know it doesn’t have to be this way. At a time when technology allows us to order a new pair of shoes on our phone and have them delivered the next day, it’s increasingly clear that technology isn’t the problem, but instead how government currently procures technology and uses it to deliver service to the public.

Today, in an effort to begin solving this problem, we are starting the State Software Collaborative at the Beeck Center, in partnership with The Rockefeller Foundation.

States need to take back control of the systems they rely on to fulfill their mission. Our goal is to help them do that, through a combination of teaching legislative staff about best practices to budget and provide oversight for major software projects, coaching agencies through using modern procurement practices, and teaching states how to center all of that work in modern software development practices (Agile software development, user-centered design, product thinking, DevOps, etc.)

By knitting together states’ agencies based on common needs, we can help states collaboratively procure, develop, and maintain the software they depend on, so instead of 50 states buying 50 versions of near-identical, overpriced software, they can procure high-quality, fair-priced software just once, and share it among themselves.

States’ needs differ substantially — because of different policies, laws, cultural norms, and technical environments — so it would be a mistake to expect as-is reuse of monolithic software projects. We expect the resulting software to be something like 80% complete, leaving room for the customization necessary to serve each state. We’ll coach states through procuring and managing scrum teams to complete the final 20%, documenting emergent best practices for other states to follow.

State governments have the subject-matter expertise, the funding, the technical knowledge, and the digital infrastructure that is necessary to deliver high-quality, technology-intermediated services to the public. They just need a little help bringing together that expertise from across states and establishing the processes and governance structure to execute on that promise, and that’s where the State Software Collaborative comes in.

We come to this project not as an academic exercise, but as practitioners with decades of experience in this subject. Robin is deeply familiar with government procurement processes from her time as Missouri’s Secretary of State and knows that states are the linchpin to our nation’s COVID-19 response, but as we’ve seen in the past few weeks, too often that work is made harder by old, hard to update and maintain legacy technology systems.

For the past four years we have helped state and local governments through our work at 18F, a tech consultancy inside the federal government General Services Administration, developing and promoting best practices for government procurement of custom software. At a time when states are on the front lines of the government’s COVID-19 response, they must take back control of systems they rely on to fulfill their mission.

The current crisis has shown how important it is for states to both learn from each other and work together in procuring critical supplies. We’ll continue to build on that collaborative spirit and states get the tools they need to support the country as we recover.

Robin Carnahan and Waldo Jaquith joined the Beeck Center as fellows in Spring 2020. They will support the State Software Collaborative project as part of the Data + Digital portfolio. Follow Robin on Twitter at @robincarnahan and Waldo at @waldojaquith.

Photos by Mackenzie Weber & Shahadat Rahman on Unsplash.

December 20, 2019 | By Margarita Arguello

As anyone choosing a college knows, finding the right balance of academic strength and affordability is incredibly difficult. In September 2015, in the midst of rising college costs and a changing U.S economy, the Obama administration launched the College Scorecard — a tool to help American families make decisions about college attendance by providing them “clear, reliable, open data on college affordability and value.” 

By pooling together data from various government agencies and conducting hundreds of interviews with potential users, government teams from the Department of Education, 18F, and the United States Digital Service (USDS) created something so useful, that within a year, 1.5 million unique users had accessed it and Google integrated its data so that it appears front and center when users conduct college-related searches. 

screen capture of google search for Georgetown University
In this screen capture from a Google search, the data fields including “Average cost after aid,” “Graduation rate,” and “Acceptance rate” draw from data collected by the U.S. Department of Education that is made available as open data through the College Scorecard project.

The College Scorecard is one example of how, throughout the past decade, the U.S. government has worked to leverage the tools of technology, data, and design to create digital products and services that directly impact the lives of people across the country. 

As evidenced by the College Scorecard, data and digital are already proving to be powerful drivers of social impact. However, there is still much to figure out in terms of how to best use these tools to create solutions that actually work for people, while also being mindful of user privacy and safety concerns. This is what the Beeck Center’s Digital Service Collaborative and the Georgetown Technology Policy Initiative sought to explore this fall through its Data, Digital + Social Impact Seminar Series. Over five seminars, experts in the field joined Georgetown students to deconstruct the nuances behind digital transformations in government  and delved deep into the topics of open data, privacy, ethics, and design for social good. Providing students an opportunity to learn practical solutions for delivering positive social impact is a key element of the Center’s mission, and one way it generates impact at scale.

The kick-off seminar provided students with a basic understanding of the building-blocks of data and examples of how social impact leaders have worked to improve the lives of people around the world through the use of open data initiatives that is, by releasing data to the public in digestible formats that allows anyone to access the data, use it and republish it. For example, Beeck Center Fellow Denice Ross, who was previously a White House Presidential Innovation Fellow and before that led open data for the City of New Orleans, talked about how emergency planners in the city of New Orleans now use open data from various federal and state government agencies to better coordinate their hurricane response efforts to prevent disastrous scenarios like those seen in the aftermath of Hurricane Katrina. 

Similarly Erica Hagen, co-founder and director at GroundTruth Initiative,  spoke about her experience leading Map Kibera — a project that aimed to create a map of Kibera, a previously unmapped slum in the Kenyan city of Nairobi. By using GIS mapping technology and enlisting the help of the people of Kibera, Hagen and her team were able to produce a map giving residents of this neighborhood access to accurate and useful information, such as the location of clinics and pharmacies within the community.

These conversations raised the important question of how to use data to design policy solutions that actually work for the people they are intended to serve. During later seminars, the expert speakers weighed in on this issue and discussed basic frameworks for effective policy making in an era of rapidly changing digital technologies. 

 

Students sit around a conference table
Beeck Center Fellow Natalie Evans Harris discusses privacy implications of data use on Oct. 8, 2019.

The most prominent ideas that the experts proposed for solving this problem involved adopting some of the popular practices of the tech sector into the policy-making process. These include conducting thorough rounds of user research in the early stages of the policy-making process, designing experimental solutions that can eventually be brought to scale, creating a “Minimum Viable Product” and then iterating to improve it, and using tech-enabled methods to incorporate the views, habits and feedback of the people that their products and services are intended to help. 

Throughout every session of the series, the experts walked students through case studies that exemplify how these concepts have been applied in real life. For example, USDS founding member Erie Meyer, who is now a tech advisor at the Federal Trade Commission, talked about how, in leading the College Scorecard project, her team took a user-centered approach that involved conducting extensive user research and collecting feedback along every step of the process. 

Likewise, Rebellion Defense co-founder Chris Lynch spoke about his experience working to improve the lives of U.S. troops in Afghanistan through his work leading the Defense Digital Service at the U.S. Department of Defense. In this role, Lynch and his team revamped and reinvented the tools and practices that the DoD uses on the ground.

Although the conversations in these five seminars shed light on the complex processes that have made digital solutions like the College Scorecard a promising reality in the United States, much work remains to be done as policy makers and scholars alike navigate the process of digitizing social services. For this reason, the Beeck Center continues to explore the topics of data governance, digital service delivery, and tech policy through its Data + Digital portfolio. 

 

Margarita Arguello is a student analyst at the Beeck Center for Social Impact + Innovation. She is a master’s student in the McCourt School of Public Policy and will graduate in May 2020.

November 22, 2019 | By Tyler Kleykamp

In an era where states often compete with one another for jobs, economic development, or federal funding, the spirit of cooperation was at the forefront last week as the Beeck Center gathered the nation’s state Chief Data Officers (CDOs) together for the first time. States are increasingly naming CDOs to serve in executive roles to influence the way states collect, maintain, and operationalize data to improve services for the public as well as to build efficiencies and effectiveness into internal systems. 

At Georgetown, the CDOs spent two days sharing challenges and successes and identifying opportunities to collaborate with one another. While they had been meeting monthly via conference call for years, this is the first time they had all been together in person. Sixteen CDOs attended, representing all but one state that has a formally established CDO role, and including one CDO who had only been on the job for five days. 

Group shot of state CDOs
State CDOs met in person for the first time at Georgetown on Nov. 12-13, 2019.

“You have to lead with value.”

The CDOs opened the meeting by sharing successes from the previous year. Examples included technology implementations like data integration platforms and improving public availability of data as well as cultural changes such as creating agency data officers and enhancing data literacy. They also discussed broad issues related to their core responsibilities and how they’ve overcome some of the challenges associated with the job, resulting in some central themes. First, relationships matter. It was clear that building trust with their partner agencies is critical and CDOs must demonstrate that they are enabling the agencies and personnel they work with to better leverage data. As one CDO said, “You have to lead with value.” Mandates and more heavy-handed approaches to data sharing can easily make their job more challenging, but they must also demonstrate the ability to handle data in responsible and legally compliant ways. 

A second theme was the amount of friction that occurs in sharing and using data. Much of this is necessary to ensure that sensitive data remains protected and that its use is legally compliant, while other issues are related to the data itself. Issues like data quality, lack of standards, or common identifiers present challenges. CDOs shared how they’ve streamlined or standardized legal agreements and ways they’ve dealt with more technical challenges — but it was clear that more can be done. One solution was having dedicated legal support for CDOs. “Having an attorney who ‘gets it’ and can speak the language of other attorneys has been critical,” shared one CDO.

A primary goal of the convening was to strengthen personal connections among CDOs in additional to building their professional relationships. A rapid-fire question-and-answer session allowed CDOs to explore emerging data issues like artificial intelligence, geospatial data, and new data protection laws. This matchmaking exercise quickly identified common issues they are thinking about so they could follow up afterward. Another mechanism to encourage more personal relationships was through “speed networking,” pairing people for short one-on-one meetings, providing an opportunity for individual conversations in a less formal setting, and setting them up to feel more comfortable working collaboratively in the future.

woman and man standing in conversation
CDOs Ed Kelly (Texas) and Julia Fischer (Maryland) engaged in conversation during an afternoon break.

A CDO’s job is big, but they consistently expressed a need to start small. They discussed surfacing discrete use cases and immediate steps they can take in their states to equip them with techniques they can apply immediately. For example, pulling together public data into a GIS map to better understand risks to children, or conducting an inventory of data that can inform an issue are great places to start. 

Another approach was to break the CDOs into smaller groups to identify specific policy areas important to their states to identify common priorities which surfaced issues like workforce opportunities, reducing opioid-related deaths, and improving child welfare. The groups established ways they can leverage their role or data to make a positive impact immediately, and identified opportunities that may be more challenging where additional support and facilitation from partners like the Beeck Center could move the needle further. 

Woman sits between two men at a round table in conversation
CDOs John Correllus (North Carolina) and Rhonda Lehman (Delaware) discuss public policy priorities with Eric Sweden of the National Association of Chief Information Officers.

Working collaboratively with the Beeck Center team, CDOs identified a broad array of high impact items that can boost their work including: 

  • templates for data sharing agreements or data inventories
  • repositories of documents or information
  • more in-depth research on technology solutions and best practices 

The exercise demonstrated the value that the Beeck Center can bring to this emerging field and allows us to prioritize the creation of high impact tools.

Sticky notes on easel paper to represent a matrix
At the convening, CDOs identified tools that could support their efforts and ranked them based on the impact they could have and the effort necessary to produce them.

Another approach with this group was to include expert lightning talks to demonstrate big ideas to use data for social impact including better serving the most vulnerable individuals in emergencies, reducing recidivism, preventing lead poisoning, improving elections, and enhancing child welfare. These plug-and-play solutions have shown promise in some jurisdictions and are ready for wider adoption by other states. Additionally, a panel of city, state, and federal government officials explored opportunities for collaboration and discussed where states can make a unique impact with data including through improvements in educational outcomes and access to jobs.

State CDOs are doing incredible work in their states, and at the Beeck Center we are committed to highlighting those successes so they can be replicated across the country. We’re getting started immediately on building a CDO toolkit that will include resources both highlighted at this convening and activated elsewhere and will support CDOs in their commitments to sharing resources such as legal agreements with the network. 

The Beeck Center will continue to convene this network both in person and online through webinars and other touchpoints and we look forward to welcoming in the next wave of CDOs as seven states recruit for these roles and hopefully join the network. The role of the State CDO Network will remain focused on keeping the momentum going until we meet again, which one CDO declared, “…is absolutely the one event I have to go to.”

 

Tyler Kleykamp is the Director of the State Chief Data Officers Network at the Beeck Center. He was previously the CDO for the State of Connecticut from 2014-2019 and led many of the early calls and connections among state CDOs.


Related Stories