Is PortfolioStat limping toward success?
- By Shawn McCarthy
- Feb 03, 2014
March of this year marks the second anniversary of the PortfolioStat project outlined by Federal CIO Steven VanRoekel and the Office of Management and Budget. Depending on who you ask, PortfolioStat is either a stale idea whose days are numbered or a moderately successful management effort that will gain renewed traction in the years ahead.
Reputation aside, one thing is clear: the idea of taking a portfolio-based approach to how federal IT projects are prioritized and funded will not go away. Enterprisewide prioritization is necessary when an organization is faced with multiple projects and limited funding.
Portfolio management is one of the basic tenets of project management. It’s especially necessary when cheaper solutions are available – often via commodity cloud-based IT services.
The PortfolioStat process has used this concept as its foundation, but in its own unique way. For example, it drew from the 2010 federal TechStat Accountability Sessions, an accountability review of an agency's IT investment. In cases where there are performance problems, it looks for ways that the projects can either be turned around or halted.
So even if the name PortfolioStat doesn't last, the CIO Council's concept of multiproject review and prioritization is likely to remain. What's needed is a way to extract the lessons learned from the current process and use them to make improvements to IT project portfolio management for the long term.
Consider this. There are over 7,000 different IT projects currently tracked by OMB. Roughly 10 years ago OMB launched a process to identify and track common elements across IT projects. As most IT managers know, these common enterprise elements provide the foundation for efforts to identify consolidation targets.
The elements are also part of the conversation during PortfolioStat meetings to drive better management of IT portfolios, increase efficiency (especially in regards to commodity IT) and improve governance of the most essential mission investments.
That can mean the reprioritization of some projects based on potential cost savings or improved citizen services. Think of it as strong encouragement to put your wood behind the arrow that has the greatest financial and performance impact on your department.
But a November report from the Government Accountability Office says that most agencies have fully addressed at least four of seven key requirements of the PortfolioStat process. Just one agency, the Education Department, has met the full set of PortfolioStat deliverables.
In theory, once the full set of requirements are addressed by an agency, PortfolioStat can have a significant impact on how IT projects are prioritized and how some are directed toward cheaper commodity solutions.
The reality is more complex. One comment we hear consistently is that agency CIOs don't have the power to enforce PortfolioStat findings and recommendations – especially across the enterprise. That could be the main reason why some CIOs feel that PortfolioStat has not had the impact that it could.
Another comment we hear is that there is no easy way to measure potential return on investment for proposed IT projects, especially those which seek to merge systems or move to cloud-based solutions.
In some cases, a simple CPA-style audit, plus estimates of current and potential migration costs, should provide enough data to make an informed decision. But the actual process of doing this measurement costs money. Significant time and effort needs to be put into the data gathering.
Perhaps a way to light a new fire under the PortfolioStat effort would be to introduce grant money that could be made available to help agencies conduct these types of up-front audits. That would let them put a stake in the ground for future cost comparisons.
The November GAO report is telling. On the negative side, it highlighted some agencies’ lack of progress. But on a positive note it indicated that it will focus additional attention on PortfolioStat follow-up, real savings related to more efficient software licensing, cloud computing (when it can be measured) and whether there have been real cost reductions related to data center consolidation.
Last year, on the first anniversary of PortfolioStat, OMB said the process already had led to over 100 plans to consolidate systems.
This doesn’t sound like a failed effort that is heading toward oblivion. It's more like a moderate level of success that needs to be expanded, but with an eye for how the tools can be most judiciously applied.