Welcome to our latest Data to Decisions project, a series of podcast interviews with federal leaders on how they are using data analytics to prevent and counter tax fraud, improve training, respond to emergencies, protect investors, keep our food supply safe, and more.

Partnership for Public Service

Between 2011 and 2013, the Partnership for Public Service and the IBM Center for The Business of Government collaborated on three reports on using data to improve decision-making.

  • From Data to Decisions: The Power of Analytics - This report set out to study federal agencies’ use of analytics and how it helped them achieve better program results. It focuses on identifying leading practices that illustrate how data informs decisions and drives meaningful and positive program changes.
  • From Data to Decisions II: Building an Analytics Culture - This second report examines what it takes to build analytics into an agency’s decision-making process and culture. The report includes concrete steps for building a disciplined approach to analytics and profiles seven agencies using analytics to achieve better results.
  • From Data to Decisions III:  Lessons From Early Analytics Programs - This third report examines long-standing programs and how they have advanced and evolved over time to be a sustainable component of an operation.  It highlights five analytic efforts that were begun, in one case, more than 25 years ago.  Based on these cases, the authors identify a series of lessons that they saw as important if analytics are to be successfully embedded in an agency’s culture.

The fourth initiative is series of podcasts.  These podcast conversations are designed to broaden the perspective to additional agencies as well as revisit some of those covered in the reports; provide insights into the essential ingredients for a successful analytics program; and offer advice from leaders whose agencies are benefiting from analyzing data.

 


Conversations on Using Analytics to Improve Mission Outcomes

 

Lisa Danzig

Lisa Danzig
Associate Director for Personnel and Performance
Office of Management and Budget

Part One: HUDStat and using analytics for veteran housing (8:29 mp3)

Part Two:  Lessons learned at HUD and aligning analytics with the 2016 budget process (9:23 mp3)

Part Three: Recommendations to agencies on analytics programs (8:43 mp3)

Interview excerpts (pdf)

Q: What is your message to agencies embarking on the budget process and seeking to use analytics?

A: Engage a set of people who think this could be worthwhile and/or already have a problem or goal they’re trying to achieve and that you could apply this to. This helps you avoid that cycle of collecting hundreds of metrics that aren’t relevant to the problem.

 

Malcolm BertoniMalcolm Bertoni
Assistant Commissioner for Planning
Food and Drug Administration

Part One:  FDA and analytics, how it all began (8:35 mp3)

Part Two:  Ingredients for a successful analytics program and a leadership's role (10:39 mp3)

Part Three:  Recommendations to agencies on analytics programs  (9:04 mp3)

Interview excerpts (pdf)

Q: What are the important elements you need for a good analytics program?

A: You have to have some leadership support because it does require an investment. You also need to have champions both on the analytical side and on the program side—some data junkies who really love measuring and understanding and analyzing how an organization ticks, and some program managers on the front lines who get it, who are willing to embrace it and work with the analysts and improve their organization to make that part of their organizational culture.

 

Carter HewgleyCarter Hewgley
Director of Enterprise Analytics
Federal Emergency Management Agency

Part One:  FEMAStat and the "no surprise model" (8:28 mp3)

Part Two:  Analytics and creating a culture of change (9:57 mp3)

Part Three:  Recommendations to agencies on analytics programs (9:37 mp3)

Interview excerpts (pdf)

Q: What changes would you make to FEMA’s analytics program if you had leeway to change anything you wanted to?

A: My wish list is that we could get a common vision across the organization. If I’m solving a problem for my team’s ability to use data, I need to solve it for the entire enterprise, getting everybody to agree on what tool we all need, how do we all get access to the right data, and how do we do that in a secure environment where we trust each other. Once you have that in place, there is no limit. You’ll just unleash the talent of people to do predictive modeling and all kinds of stuff you see happening elsewhere.

 

Dean SilvermanDean Silverman
Senior Advisor to the Commissioner, Office of Compliance Analytics
Internal Revenue Service

Part One:  IRS analytics programs and reducing fraud (11:09 mp3)

Part Two:  Solving the "really big problems" at the IRS with new solutions (9:13 mp3)

Part Three:  Recommendations to agencies on analytics programs (11:22 mp3)

Interview excerpts (pdf)

Q: Why is it so important to use data analytics?

A: Leaders are always better off having data and an increased understanding of how things work, and how they might work more effectively, both to make decisions and to define the problem. The latter is frequently not thought of as a major focus of data analytics, but defining the problem is where it all starts. Einstein once said, if he was given 60 minutes to solve a problem, he’d spend 50 minutes thinking about it and framing the problem and the 10 minutes solving it. None of us are Einsteins, but the principle still applies.

 

Gerald RayGerald Ray
Deputy Executive Director, Office of Appellate Operations
Social Security Administration

Part One:  Analytics and decision making in legal proceedings (10:07 mp3)

Part Two:  Building a successful data analytics program, change management, and customer service (9:10 mp3)

Part Three: Recommendations to agencies on analytics programs (9:09 mp3)

Interview excerpts (pdf)

Q: Why do you think some agencies struggle with their use of analytics?

A: There are a couple of core problems. One is they’re not particularly familiar with it. The other piece of it is that many agencies don’t have enough data scientists, or the data scientists are removed from the operational people, and if they’re not consulting with each other, they simply go off in different tangents, in different directions and you don’t get what you really need.

 

Lori WalshLori Walsh
Chief, Center for Risk and Quantitative Analytics
Securities and Exchange Commission

Part One: Analytics and the SEC's mission and programs (10:20 mp3)

Part Two: Important elements of an analytics program and the role of leadership  (10:17 mp3)

Part Three: Recommendations to agencies on analytics programs (10:06 mp3)

Interview excerpts (pdf)

Q: What role does leadership play in making analytics successful?

A: Leadership is critical. It takes time to get the infrastructure in place so you need leadership who understands and is willing to back you during this time of growth. Additionally, you’re always going to have the people who want to do things the way they’ve always done things and having senior leadership step in and say we’ve got to do things differently is critical.

 

Steve BeltzSteve Beltz
Assistant Director, Recovery Operation Center
Recovery Accountability and Transparency Board

Part One:  Using data to show accountability and transparency with federal funds (9:46 mp3)

Part Two: Important elements of an analytics program, people are critical (8:55 mp3)

Part Three:  Recommendations to agencies on analytics programs (9:37 mp3)

Interview excerpts (pdf)

Q: What advice would you give to others on how to improve their data analytics programs?

A: They have to understand their data. That’s very important. If they can’t get good access to their data or in an understandable format, then you’re not going to get anywhere, and that’s what most agencies are struggling with right now. You need to buy good tools and then the next part is finding good analysts.

 

 

Podcast music: Drops of H2O (The Filtered Water Treatment) by J. Lang featuring Airtone. Copyrighted under a Creative Commons license.