The key is using the data to actually inform decisions, not just compiling it into reports. Writing on The HUDdle, the Department of Housing and Urban Development's blog, the agency's CIO Jerry Williams describes a project HUD is undertaking with the Department of Veterans Affairs.
The Homelessness Analytics Initiative (HAI) will combine data sets from multiple agencies on homelessness counts, unemployment rates and other pertinent information, with the aim of creating a predictive model for forecasting homelessness patterns, monitoring shelter performance and addressing issues such as homelessness among veterans.
Prior to HAI, Williams said each agency collected its data without worrying about interoperability with data from other agencies. Data integration and data quality obviously will be key factors for HAI, as they are for many enterprise data initiatives.
IT Business Edge's Loraine Lawson cited a recent survey that found those two issues top the list of business intelligence problems. Forty-four percent of organizations surveyed by SearchBusinessAnalytics.com said data integration would be one of their biggest challenges this year, up from 23 percent last year.
If HAI is successful, it will save money, as the integration approach should cost less than building the same intelligence from scratch, Williams writes:
Transformation is about being more efficient and effective, lowering costs, and doing more with less. Optimizing existing data, to help us make more informed decisions, will help us deliver increased value with fewer dollars.
Williams lists five factors he says will be critical to HAI's success:
- Create a solid information architecture, so agencies will know what data they have.
- Introduce policies, procedures, standards and guidelines to link data intelligence efforts with the agencies' overall mission. (He says this won't be tough in this instance, as both HUD and the VA have been tapped by the Obama administration to focus on homelessness.)
- Treat data as a national asset, removing any "ownership" barriers to allow agencies to more easily share data.
- Exploit and reward folks who creatively use data to solve problems.
- Pay close and constant attention to data security and privacy.