Federal IT Management Still Has Room for Improvement

Agencies have had trouble completing IT projects on time and on budget since the days of the original UNIVAC computer.

The government has devoted significant attention to determining how to build IT systems more reliably and effectively. Entire methodologies have been conceived to improve project management, including the use of automated tools. And yet, problems persist.

Simple Project Lifecycle View

All projects go through the same simple project stages:

• Planning
• Design
• Building
• Operation

At each stage, agencies must address specific challenges and deal with other factors that often interrelate and create dependencies on each other. Among these factors are complexity and external challenges. Complexity is often introduced into a project in the form of organizational requirements, as well as the technologies and materials involved. Among the external challenges are laws and regulations, labor and materials, and the strength of sponsorship and funding.

Clearly, the project lifecycle is often complex and challenging. That said, some industries — the construction industry, for example — commonly deliver much more reliable results than IT. And while it is perhaps unfair to compare an industry that has existed for thousands of years with one that has existed for a few decades, I think agencies can learn some key points from it.

• The construction industry has developed and uses standards in every aspect of its work. The standards are universal, and designers (architects and engineers) can rely on their materials to meet these standards.
• Capital markets exist to fund building projects, and, as a result, due diligence and quality assurance are highly developed attributes that drive project performance.

Not all construction projects are delivered on time and within budget. Construction projects have their own external factors that can hamper completion, such as weather, material and labor shortages, and local politics. But the construction industry does deliver more reliable results than IT on average and provides a useful model for adoptable practices.

The Federal Analog

An astute observer can examine the federal approach to project management at several levels. We can see numerous examples of comprehensive, disciplined approaches at the project (and sometimes the enterprise) levels, with adequate or superior results. The best examples are project management offices that execute a strategy based on the Project Management Professional (PMP) approach.

At a higher level, the federal approach is driven by the Office of Management and Budget. Where the PMP can be considered an integrated, comprehensive set of processes used to manage IT projects, the federal approach at the OMB level can be more appropriately represented as a collection of tools and best practices.

Most of these best practices have been implemented since the enactment of the Clinger–Cohen Act in the mid-1990s. They include capital planning, investment control, value-measuring methodology, enterprise architecture and portfolio management. They have been part of various management improvement efforts from the administrations in place over this period. The more recent additions have been the Obama administration’s 25-Point IT Reform Plan and PortfolioStat.

The best practices are each useful tools that address an area of concern and, if used thoughtfully, help with the challenges of managing IT. Implementation of these best practices has proved to be a challenge, because each requires development of policies and practices, education, commitment and resources.

It is understandable that OMB would be reluctant to standardize a more comprehensive, integrated approach to project management, but frankly, that is what is missing at the federal level. While some agencies “get it” and have established a project management methodology, far too many have not.

The Role of PortfolioStat

PortfolioStat was implemented with great fanfare as a key process to keep major IT projects on track, improving results and saving money in the process. It is useful to examine the roles of PortfolioStat and what the high-level results have been.

One role of PortfolioStat is to assist and oversee major federal IT projects. Part of the process has been to strengthen IT portfolio governance within agencies, to more effectively plan and manage projects. The focus has been on organization-level investment review boards.

Change does not come quickly to the fed- eral government, and benefits from the current focus certainly may accrue in the future, but the results of PortfolioStat have been underwhelming. While PortfolioStat has pulled the plug on out-of-control projects — a laudable management activity — avoiding problems by planning and managing projects more effectively should be the priority focus. Stopping projects does not implement the IT improvements needed to move agency missions forward. It merely stops the current bleeding of resources.

Another objective of PortfolioStat is to advance OMB initiatives such as cloud computing and data center consolidation. This use is more focused and clear, but there’s still a long way to go before PortfolioStat plays a significant role in improving federal IT management.