'Moneyball' strategy works for feds?
Thursday - 12/1/2011, 5:33am EST
Emily Kopp, reporter, Federal News Radio
A's General Manager Billy Beane built a winning baseball team on a shoestring budget by breaking with conventional wisdom. He recruited an Ivy Leaguer to find statistics that really mattered to winning games. Then he used them to build a roster of players on a shoestring budget.
"The thing that you think will help you win isn't necessarily the thing that will help you win," said Centers for Medicare and Medicaid Services Deputy Chief Operating Officer Michelle Snyder at the release of the new report "From Data to Decisions: The Power of Analytics."
Collecting and analyzing the right data is a key theme of the report by the nonprofit Partnership for Public Service and IBM's Public Sector Business Analytics and Optimization. They found a few common keys to success:
- Leaders communicated a clear vision, set expectations and held staff accountable for results.
- Staff had clear line of sight from where they stood to goal
- Agencies invested in technology, tools and talent, although not at great expense.
- Agencies cultivated partnerships with more than the "usual suspects."
Shelley Metzenbaum, associate director for performance and personnel management, Office of Management and Budget (state.gov)
In preparing the report, researchers examined data analytics at eight agencies. They drew on programs at CMS, the departments of Veterans Affairs and Housing and Urban Development, the Federal Aviation Administration, the Social Security Administration, the National Highway Traffic Safety Administration, the Coast Guard and the Navy.
At the VA-HUD program to end veterans' homelessness, the goal was clearly articulated, but the path forward was not.
"There was a lifecycle that we needed to capture," said VA's Performance Improvement Officer David Zlowe.
HUD stopped counting how many housing vouchers they issued to vets, and started counting how many vets were using them. The VA compared that information to vets' medical records.
They found that veterans with substance abuse problems were more likely to drop out of the program than those with mental illnesses.
"Now we can start to manage our interventions and programs even more effectively," Zlowe said.
RELATED STORIES:
Analytics helps agencies turn data into decisions
VA, HUD to use better data to aid homeless vets





