15 November 2003
GAO-04-190R FBI Enterprise Architecture
United States General Accounting Office
Washington, DC 20548
November 14, 2003
The Honorable Porter J. Goss
Chairman, Permanent Select Committee on Intelligence
House of Representatives
The Honorable Nancy Pelosi
House of Representatives
The Honorable Bob Graham
United States Senate
The Honorable Richard C. Shelby
United States Senate
Subject: Federal Bureau of Investigations Comments on Recent GAO Report on its Enterprise Architecture Efforts
On September 25, 2003, we issued our report on efforts by the Federal Bureau of Investigation (FBI) to develop a corporate blueprintcommonly called an enterprise architectureto guide and constrain its information technology (IT) systems modernization.1 (This report is available on GAOs Web site at www.gao.gov/cgi-bin/getrpt?GAO-03-959.) We provided the FBI with a draft of this report on August 22, 2003, requesting that comments be provided by September 18. On September 23, the FBI provided us with written comments. However, the comments were not received in time to be analyzed, incorporated, and responded to in the report and still meet our September 25, 2003, reporting commitment to you. As discussed with your offices at that time, we did not extend the reporting date in order to include the FBIs comments and instead are transmitting and responding to them in this follow-up correspondence.
1 U.S. General Accounting Office, Information Technology: FBI Needs an Enterprise Architecture to Guide Its Modernization Activities, GAO-03-959 (Washington, D.C.: Sept. 25, 2003).
In its written comments signed by the Assistant Director, Inspection Division (which are reprinted in their entirety in the enclosure), the FBI made two primary points. First, it expressed its commitment to developing and using an enterprise architecture (EA), including (1) agreeing with our conclusion that it needs an architecture to effectively manage its IT systems modernization; (2) consistent with our recommendations, stating that it recognized the need for immediate attention to its architecture efforts; and (3) noting that it was managing its architecture effort as an IT modernization enabler and priority.
Related to this first point, the FBI also stated that it has efforts currently under way to improve its EA posture, and that substantial and real progress has already been made in doing so. For example, it stated that an executive team had been established to (1) assess the bureaus EA status and resource needs using our EA maturity management framework2 and (2) formulate recommendations for improvement. Although the FBIs comments did not specify when it would complete the assessment, it did state that the necessary resources would be applied to architecture development, maintenance, and implementation following the results of the assessment. To illustrate its progress, the FBI stated that it had
completed and approved what it referred to as an EA foundation document which, according to its comments, contains an architecture approach based on 55 principles spanning 10 categories of bureau activities and operations and acknowledges its largest modernization project (Trilogy) as one enabler for moving from its current architectural state to its target state;
established key IT modernization management structures and processes, such as an investment management process that requires all proposed investments to address EA, a governance board to review investment proposals and architectural decisions, an application integration board to ensure that new applications are consistent with the bureaus IT environment, and change management and control entities to examine and approve changes to its IT infrastructure;
assigned EA resources, including appointing a chief architect, assigning staff, and obtaining private-industry expert assistance, to support its ongoing architecture assessment and development of architecture products;
established a list of existing systems that had completed security certification and accreditation;
begun acquiring an automated tool to serve as an architecture repository, as well as a risk management tool for determining existing system vulnerabilities and cost-effective risk mitigation steps; and
begun conducting outreach with external parties, such as the Justice Department, the federal CIO Council, and its intelligence community partners to, among other things, learn from these entities EA experiences.
2 U.S. General Accounting Office, Information Technology: A Framework for Assessing and Improving Enterprise Architecture Management (Version 1.1), GAO-03-584G (Washington, D.C.: April 2003).
We support the FBIs stated commitment to architecture development and use, including its adoption of our maturity framework. Moreover, we believe that the examples of EA-related activities cited in the bureaus comments, some of which were subsequent to completion of our audit work, are steps in the right direction.
However, the examples that the FBI cites do not alter our reports findings and conclusions about the maturity of the FBIs EA program because they are either already recognized in our report or they do not fully address the EA management maturity core elements that our report cites as not being satisfied. Moreover, the FBI does not currently have a version of an EA to guide and constrain its ongoing and planned IT investments. Our evaluation and response to each of the FBIs examples of progress are provided below.
At the time we completed our audit work, the EA foundation document was in draft form, and our review of this draft showed that while it contained information that would be useful in developing a plan for architecture development, maintenance, and implementation, as well as information that would be useful in developing architecture artifacts or products, it did not satisfy the basic content requirements for either an EA or a plan for developing, implementing, and maintaining one. For example, neither the draft nor the recently approved version specifies the tasks, time frames, or responsible parties for actually developing and completing such architecture products as the business, information/data, services/applications, technology, and performance reference models, as well as the security views that should be part of these models.
At the time we completed our audit work, the bureaus EA governance board did not include all relevant internal stakeholders, such as representatives from its counterterrorism and counterintelligence organizational components. As our framework recognizes, enterprisewide representation and accountability on the architecture governance body is a critical success factor and a recognized best practice. Since we issued our report, FBI officials told us that they now have all relevant stakeholders represented on the board.
Our report recognizes that the bureau had appointed a chief architect and assigned staff as part of its EA efforts. However, the report also points out that it began these efforts over 32 months ago, and the level of commitment and resources devoted to them had neither advanced the FBI beyond stage 1 of our maturity framework nor produced an EA that could effectively support the investment and modernization management processes and structures that the FBI cited as having been established. Moreover, as we state in the report, the then-chief architect characterized the bureaus annual commitment of $1 million in resources to these efforts as limited, and this amount now appears to be an overstatement. Specifically, the FBI stated in its comments that it is actually investing less than this amount in its EA efforts ($285,000 and $500,000 in fiscal years 2003 and 2004, respectively), but that its fiscal year 2005 budget request includes a substantial, but unspecified, increase.
Despite the bureaus progress in establishing a listing of existing systems under security certification and accreditation, which we believe would be a useful source of information in developing an EA, the then-chief architect told us that this listing was incomplete and required management approval before it could serve as a basis for developing the as-is architecture description.
The bureaus comments acknowledge that it is in the process of acquiring automated EA tools, and thus does not yet satisfy core elements of our framework related to establishing an EA management foundation. Further, to augment these tools, the bureau has yet to establish a methodology that it will follow to create its architecture artifacts, which is another management foundation core element.
We support the efforts that the FBI cited for outreach to relevant external stakeholders. Understanding these relationships, and ultimately defining them in architecture artifacts, should be part of an effectively managed EA program.
The FBIs second primary comment was that our report was too narrowly focused and not comprehensive because it was limited to EA and did not include an assessment of the FBIs other IT management controls and capabilities. Because our report focused on EA, the bureau said that the report was premature.
While we agree that the report focuses on the FBIs EA activities, we do not agree that this is either inappropriate or makes the report premature. As agreed with your offices, we are in the process of reviewing a wide range of FBI IT management areas, such as system acquisition capabilities, IT human capital management, IT investment management practices, and architecture development and use. As further agreed, we are to report on these areas incrementally, as appropriate. Our report represents an appropriate and timely first increment for two principal reasons.
Our experience over the last 10 years in evaluating federal agency IT management has shown that providing our congressional clients and the subject agencys leadership team with the results of major segments of our work as they are available permits more timely corrective action, and thus better outcomes.
Reporting first on EA in particular, which can be viewed as an essential link between strategic planning and system investment/implementation, provides the FBI sooner rather than later with a comprehensive set of recommendations for effectively making its architecture efforts more mature in time to influence its ongoing and planned IT investment/implementation efforts. Any delay on our part in reporting on this area of strategic importance would only increase the agencys exposure to modernization risk and postpone your awareness and understanding of this critical issue. This does not, however, mean that the FBI should not be pursuing near-term IT upgrades before it completes and is positioned to use an architecture, nor is it intended to suggest that the bureaus planned and ongoing modernization investments to date are completely unjustified and unreasonable. Rather, it means that these investments and upgrades are being pursued without a blueprint that provides an authoritative, commonly understood frame of reference that translates strategy into implemental actions, which, in turn, increases modernization risk.
- - - - -
We are sending copies of this correspondence to the Chairman and Vice Chairman of the Senate Select Committee on Intelligence and the Ranking Minority Member of the House Permanent Select Committee on Intelligence. We are also sending copies to the Attorney General; the Director, FBI; the Director, Office of Management and Budget; and other interested parties. In addition, this correspondence will be available without charge on GAOs Web site at www.gao.gov.
Should you or your offices have any questions on matters discussed in this correspondence, please contact me at (202) 512-3439 or by e-mail at firstname.lastname@example.org. Key contributors to this response included Katherine I. Chu-Hickman, Barbara Collier, Gregory Donnellon, Michael P. Fruitman, Paula A. Moore, Gary N. Mountjoy, and Megan M. Secrest.
Randolph C. Hite
Director, Information Technology Architecture and Systems Issues
Enclosure: Comments from the Federal Bureau of Investigation