DOD's Standard Procurement System: Continued Investment Has Yet  
to Be Justified (07-FEB-02, GAO-02-392T).			 
								 
The Department of Defense (DOD) lacks management control of the  
Standard Procurement System (SPS). Specifically, DOD has not (1) 
ensured that accountability and responsibility for measuring	 
progress against commitments are clearly understood, performed,  
and reported; (2) demonstrated, on the basis of reliable data and
credible analysis, that the proposed system solution will produce
economic benefits commensurate with costs; (3) used data on	 
progress against project cost, schedule, and performance	 
commitments throughout a project's life cycle to make investment 
decisions; and (4) divided this large project into a series of	 
incremental investment decisions to spread the risks over	 
smaller, more manageable components. GAO found that DOD lacks the
basic information needed to make informed decisions on how to	 
proceed with the project. Nevertheless, DOD continues to push	 
forward in acquiring and deploying additional versions of SPS	 
that involve considerable risk. This testimony summarized the	 
July report, GAO-01-682.					 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-02-392T					        
    ACCNO:   A02738						        
  TITLE:     DOD's Standard Procurement System: Continued Investment  
Has Yet to Be Justified 					 
     DATE:   02/07/2002 
  SUBJECT:   Defense procurement				 
	     Defense economic analysis				 
	     Information resources management			 
	     Information systems				 
	     Performance measures				 
	     DOD Standard Procurement System			 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-02-392T
     
United States General Accounting Office

GAO Testimony

Before the Subcommittee on National Security, Veterans Affairs, and
International Relations, Committee on Government Reform, House of
Representatives

For Release on Delivery
Expected at
9:30 a.m.
Thursday,
February 7, 2002

DOD'S STANDARD PROCUREMENT SYSTEM

                            Continued Investment
                          Has Yet to Be Justified

Statement of Joel C. Willemssen
Managing Director, Information Technology Issues

GAO-02-392T

Mr. Chairman and Members of the Subcommittee:

I am pleased to be here today to discuss the Department of Defense's (DOD)
management of its investment in the Standard Procurement System or SPS
program. The department launched this program a little more than 7 years ago
with the laudable goal of replacing 76 existing procurement systems with a
single departmentwide system to more effectively support divergent
contracting processes and procedures across its component organizations.
Through SPS, the department expected to improve efficiency and effectiveness
in how it awarded and managed contracts, and, at that time, estimated
life-cycle costs to be approximately $3 billion over a 10-year period.

The department's goals for SPS are reinforced by the president's recent
management agenda, which emphasizes investing in information technology to
achieve results. The agenda also noted that the federal government has not
produced measurable gains in productivity commensurate with its investment
in information technology, 1 which is now estimated to be more than $50
billion for fiscal year 2003.The agenda reiterates that program performance
and results are what matters most, and that actual program accomplishments,
as well as needs, should be the prerequisite to continued funding. This
emphasis is consistent with information-technology investment management
provisions of federal law and guidance2 and information-technology
management practices of leading public- and private-sector companies.

For the SPS program, we reported in July 2001 that the department had not
met these investment management criteria.3 Specifically:

* The department had not economically justified its investment in the
program because its latest (January 2000) analysis of costs and benefits was
not credible. Further, this flawed analysis showed that the system, as
defined, was not a cost-beneficial investment.

* It had not effectively addressed the inherent risks associated with
investing in a program as large and lengthy as SPS because it had not
divided the program into incremental investment decisions that coincided
with incremental releases of system capabilities.

1The President'sManagementAgenda:Fiscal Year2002, Executive Office of the
President, Office of Management and Budget.

2Clinger-Cohen Act of 1996, Public Law 104-106; Office of Management and
Budget Circular A-130, Managementof FederalInformationResources(November 30,
2000).

3U.S. General Accounting Office, DOD Systems
Modernization:ContinuedInvestmentinthe StandardProcurementSystemHasNotBeen
Justified,GAO-01-682 (Washington, D.C.: July 31, 2001).

* The department had not met key program commitments that were used to
justify the program. For example, the department committed to implementing a
commercially available contract management system; however, because it had
modified so much of the foundational commercial product, SPS evolved into a
customized DOD system. Also, although the department committed to fully
implementing the system by March 31, 2000, this target date had slipped by 3
ï¿½ years to September 30, 2003, and program officials have recently stated
that this date will also not be met.

* It did not know if it was meeting other key program commitments. For
example, the department had not measured whether promised system benefits
were being realized, and the information that was available about system
performance showed that users were not satisfied with the system. Also,
because DOD was not accumulating actual program costs, it did not know the
total amount spent on the program to date, yet life-cycle cost projections
had grown from about $3 billion to $3.7 billion.

Collectively, this meant that the question of whether further investment in
SPS was justified could not be answered with any certainty. Accordingly, we
recommended that investment in future releases or major enhancements to the
system be made conditional on the department first demonstrating that the
system was producing benefits that exceed costs, and that future investment
decisions be based on complete and reliable economic justifications. We also
recommended that program officials clarify organizational accountability and
responsibility for the program, determine the program's current status, and
identify lessons learned from the SPS investment management experience.

In commenting on a draft of our report, the Deputy Chief Information Officer
(CIO) generally disagreed with our recommendations, noting that they would
delay development and deployment of SPS. Since that time, however, the
department has either initiated or stated its intention to initiate steps
that are consistent with our recommendations. It has also taken steps to
address the findings of several department-sponsored studies initiated at
the time of our report. For example, it has (1) clarified organizational
accountability and responsibility for the program, (2) established missing
controls over key acquisition processes such as requirements management and
testing, and (3) begun addressing users' concerns. In addition, department
officials have stated that the department will prepare an economic analysis
before investing beyond already executed contractual commitments and that it
will conduct a productivity study to assess the extent to which the
department is deriving benefits from SPS. These are positive steps that have
advanced the program beyond where it was at the time of our report.

SPS: A Brief Description and History

Nevertheless, much remains to be done before the department will be in a
position to make an informed, data-driven decision about whether further
investment in the system is justified. Namely, although program officials
have stated their intentions to address our recommendations, they have not
yet committed to specific tasks for doing so nor have they established
milestone dates for completing these tasks. Further, the department may
expand the functionality of the current software release to include
requirements previously slated for later releases, which could compound
existing problems and increase costs; and, although intended to be a
standard system for the entire department, not all defense components have
agreed to adopt SPS.

In November 1994, the Office of the Director of Defense Procurement
initiated the SPS program to acquire and deploy a single automated system to
perform all contract-management-related functions for all DOD organizations.
At that time, life-cycle costs were estimated to be about $3 billion over a
10-year period.

From 1994 to 1996, the department defined SPS requirements and solicited
commercially available vendor products for satisfying these requirements.
Subsequently, in April 1997, the department awarded a contract to American
Management Systems (AMS), Incorporated, to (1) use AMS's commercially
available contract management system as the foundation for SPS, (2) modify
this commercial product as necessary to meet DOD requirements, and (3)
perform related services.4 The department also directed the contractor to
deliver functionality for the system in four incremental releases. The
department later increased the number of releases across which this
functionality would be delivered to seven, reduced the size of the
increments, and allowed certain more critical functionality to be delivered
sooner (see table 1 for proposed SPS functionality by increment).

4DOD is not acquiring the source code for SPS and, unless an expanded
license is obtained, is required to obtain sole-source support over the life
of this system from AMS.

             Table 1: Summary of SPS Functionality by Increment

Software

release Increment (subreleases) Functionality

1 3.1 Provide base-level contracting capabilities enabling DOD procurement
personnel to prepare simple contracts, which are generally fixed-price,
1-year contracts that will not be modified.

5.0 Provide more
complex
contracting
capabilities,
enabling DOD
procurement
personnel to
purchase weapons
systems. These
contracts are
generally fewer in number, but are more complicated, consisting of numerous
provisions and contract line-item numbers, and usually undergo extensive
modifications.

5.1 Provide inventory control point (ICP) functionality for ICPs, which are
responsible for the support and acquisition of spare parts and supplies,
enabling workload management to better manage inventories.

Source: DOD.

Since our report of July 2001,5 DOD has revised its plans. According to the
SPS program manager, current plans no longer include increments 6 and 7 or
releases 5.0 and 5.1. Instead, release 4.2 (increment 5) will include at
least three, but not more than seven, subreleases. At this time, only the
first of the potentially seven 4.2 subreleases is under contract. This
subrelease is scheduled for delivery in April 2002, with deployment to the
Army and the Defense Logistics Agency scheduled for June 2002. Based on the
original delivery date, release 4.2 is about one year overdue.

The department reports that it has yet to define the requirements to be
included within the remaining 4.2 subreleases, and has not executed any
contract task orders for these subreleases. According to SPS officials, they
will decide later this year whether to invest in these additional releases.

As of December 2001, the department reported that it had deployed four SPS
releases to over 777 locations.6 The Director of Defense Procurement (DDP)
has responsibility for the SPS program,7 and the CIO is the

5GAO-01-682 (July 31, 2001).

6All DOD components except the Air Force have deployed subrelease 4.1e; the
Air Force has only deployed through subrelease 4.1b. The Air Force is
scheduled to begin deployment of release 4.1e in March 2002.

7DDP is organizationally located within the Office of the Under Secretary of
Defense for Acquisition, Technology and Logistics.

milestone decision authority for SPS because the program is classified as a

8

major Defense acquisition.

Numerous SPS Concerns Have Been Raised by Us and Others

Our July 2001 report detailed program problems and investment management
weaknesses.9 To address these weaknesses, we recommended, among other
things, that the department report on the lessons to be learned from its SPS
experience for the benefit of future system acquisitions. Similarly, other
reviews of the program commissioned by the department in the wake of our
review raised similar concerns and identified other problems and management
weaknesses. The findings from our report are summarized below in two major
categories: lack of economic justification for the program and inability to
meet program commitments. We also summarize the findings of the other
studies.

DOD Had Not Economically Justified Its Investment in SPS

The Clinger-Cohen Act of 1996, OMB guidance, DOD policy, and practices of
leading organizations provide an effective framework for managing
information technology investments, not just when a program is initiated,
but continuously throughout the life of the program. Together, they provide
for

(1) economically justifying proposed projects on the basis of reliable
analyses of expected life-cycle costs, benefits, and risks; and

(2) using these analyses throughout a project's life-cycle as the basis for
investment selection, control, and evaluation decisionmaking, and doing so
for large projects (to the maximum extent practical) by dividing them into a
series of smaller, incremental subprojects or releases and individually
justifying investment in each separate increment on the basis of costs,
benefits, and risks.

The department had not met these investment management tenets for SPS.
First, the latest economic analysis for the program-dated January 2000- was
not based on reliable estimates because most of the cost estimates in the
2000 economic analysis were estimates carried forward from the April 1997
analysis (adjusted for inflation). Only the cost estimates being funded and
managed by the SPS program office, which were 13 percent of the total
estimated life-cycle cost in the analysis, were updated in 2000 to reflect
more current contract estimates and actual expenditures/ obligations for
fiscal years 1995 through 1999. Moreover, the military services, which share
funding responsibility with the SPS program office for implementing the
program, questioned the reliability of these cost

8DOD Regulation 5000.2-R, MandatoryProceduresforMajorDefense Acquisition
Programs and MajorAutomatedInformation SystemAcquisition Programs, specifies
mandatory

policies and procedures for major acquisitions. The policy also specifies
that the DOD CIO is the milestone decision authority, responsible for
program approval, for all major automated information systems, such as SPS.

9GAO-01-682 (July 31, 2001).

estimates. However, this uncertainty was not reflected in the economic
analysis using any type of sensitivity analysis.10 A sensitivity analysis
would have disclosed for decisionmakers the investment risk being assumed by
relying on the estimates presented in the economic analysis.

Moreover, the latest economic analysis (January 2000) was outdated because
it did not reflect the program's current status and known problems and
risks. For instance, this analysis was based on a program scope and
associated costs and benefits that anticipated four software releases.
However, as mentioned previously, the program now consists of five releases,
and subreleases within releases, in order to accommodate changes in SPS
requirements. Estimates of the full costs, benefits, and risks relating to
this additional release and its subreleases were not part of the 2000
economic analysis. Also, this analysis did not fully recognize actual and
expected delays in meeting SPS's full operational capability milestone,
which had been slipped by 3ï¿½ years and DOD officials say that further delays
are currently expected. Such delays not only increase the system acquisition
costs but also postpone, and thus reduce, accrual of system benefits.
Further, several DOD components are now questioning whether they will even
deploy the software, which would further reduce SPS's cost effectiveness
calculations in the 2000 economic analysis.

Second, the department had not used these analyses as the basis for deciding
whether to continue to invest in the program. The latest economic analysis
showed that SPS was not a cost-beneficial investment because the estimated
benefits to be realized did not exceed estimated program costs. In fact, the
2000 analysis showed estimated costs of $3.7 billion and estimated benefits
of $1.4 billion, which was a recovery of only 37 percent of costs. According
to the former SPS program manager, this analysis was not used to manage the
program and there was no DOD requirement for updating an economic analysis
when changes to the program occurred.

Third, DOD had not made its investment decisions incrementally as required
by the Clinger-Cohen Act and OMB guidance. That is, although the department
is planning to acquire and implement SPS as a series of five increments, it
has not made decisions about whether to invest in each release on the basis
of the release's expected return on investment, as well as whether prior
releases were actually achieving return-on-investment expectations. In fact,
for the four increments that have been deployed, the department had not
validated whether the increments were providing promised benefits and was
not accounting for the costs associated with each increment so that it could
even determine actual return on investment.

10That is, an analysis to explicitly present the return-on-investment
implications associated with using estimates whose inherent imprecision
could produce a range of outcomes.

Instead, the department had treated investment in this program as one,
monolithic investment decision, justified by a single, "all-or-nothing"
economic analysis. Our work has shown that it is difficult to estimate, with
any degree of accuracy, cost and schedule estimates for many increments to
be delivered over many years because later increments are not well
understood or defined. Also, these estimates are subject to change based on
actual program experiences and changing requirements. This "all-or-nothing"
approach to investing in large system acquisitions, like SPS, has repeatedly
proven to be ineffective across the federal government, resulting in huge
sums being invested in systems that do not provide commensurate benefits.

DOD Had Not Met or Did Not Know if It Had Met SPS Commitments

Measuring progress against program commitments is closely aligned with
economically justifying information-technology investments, and is equally
important to ensuring effective investment management. The Clinger-Cohen
Act, OMB guidance, DOD policy, 11 and practices of leading organizations
provide for making and using such measurements as part of informed
investment decisionmaking.

DOD had not met key commitments and was uncertain whether it was meeting
other commitments because it was not measuring them. (See table 2 for a
summary of the department's progress against commitments.)

11DOD Interim Regulation 5000.2-R, MandatoryProceduresforMajorDefense
Acquisition Programsand MajorAutomated Information SystemAcquisition
Programs(January 4, 2001).

Table 2: Progress Against SPS Program Commitments

               Key commitments Commitment met? Explanation(s)

System fully operational by March 31, 2000

No Problems were encountered in modifying and testing the commercial product
and in adequately defining requirements. For example, there were no system
performance requirements in the

b

          SPS contract.a The target date had slipped 3-1/2 years.

Contracting community 's needs met  No Approximately 60 percent  of the user
population  recently surveyed   by DOD 's  OIG  were dissatisfied  with  the
system's functionality and

c

                                performance.

Acquire a commercially available software product

No  The commercial  product had  been extensively  modified, resulting  in a
DOD-unique system.

Other commitments

Replace 76 legacy procurement systems and manual processes, thereby reducing
procurement system operations and maintenance costs

? Only 2 legacy systems had been fully retired and 2 partially retired, and
DOD did not know what, if any, associated cost savings had resulted. Also,
DOD now plans to retire only 14 legacy systems as a result of SPS's
implementation.

Increase  user  productivity  ?  DOD  is  unaware  of the  extent  to  which
productivity  may  have   increased  because  it did  not  implement  needed
performance metrics.

Standardize policies, processes, and procedures

?  Each military  service  had or  was planning  to develop  its own  unique
program documentation.

Reduce problem disbursements ? DOD was unable to provide any evidence that
implementing SPS had reduced problem disbursements, nor had it included this
benefit in its latest economic analysis.

 Life-cycle costs of $3.7 billion      DOD was unaware of the amount spent on the program
            over a 10-             ?                                             to date
           year period                     because cost information was being tracked and
                                                                     officially reported
                                     only for the SPS program office. Costs incurred by
                                     all DOD
                                     d component organizations were not accumulated and
                                     reported.

?-DOD was unaware of the extent to which the commitment had been met.

aWhile the former program manager attributed the delay to an increase in
requirements, the SPS Joint Requirements Board chairperson stated that no
additional requirements had been approved. Rather, the board's chairperson
stated that the original requirements had not been well-defined and
clarification was needed to better ensure that user needs would be met.

bAccording to the current program manager, the most recent target date of
September 30, 2003, will not be met. In addition, another target date has
not yet been established for completing the program.

cA user satisfaction manager was recently designated for this program.

dBased on DOD documents we obtained during our current review, at a minimum,
$511.6 million had been spent as of September 30, 2001.

To partially fill the void in knowing progress against SPS commitments, the
program office initiated a study in June 2000 to validate the extent to
which benefits from version 4.1 would be realized. However, the study was
not well planned and executed, and while some useful information was
obtained, the study did not allow DOD to validate whether expected benefits
were actually being realized. For example,

* the sample selected was not statistically valid, meaning that the results
were not projectable to the population as a whole,

* the study was based on the 1997 economic analysis instead of the more
current 2000 economic analysis, despite key differences between the

two analyses, such as the number and dollar value of estimated benefits, and

* the information gathered did not map to the 22 benefit types listed in the
1997 economic analysis. Instead, the study collected subjective judgments
(perceptions) that were not based on predefined performance metrics for SPS
capabilities and impacts. Thus, the department was not measuring SPS against
its promised benefits.

The former program manager told us that knowing whether SPS was producing
value and meeting commitments was not the program office's objective because
there was no departmental requirement to do so. Rather, the objective was
simply to acquire and deploy the system. Similarly, CIO officials told us
that the department was not validating whether deployed releases of SPS were
producing benefits because there was no DOD requirement to do so and no
metrics had been defined for such validation.12 However, the Clinger-Cohen
Act of 1996 and OMB guidance13 emphasize the need to have investment
management processes and information to help ensure that
information-technology projects are being implemented at acceptable costs
and within reasonable and expected time frames and that they are
contributing to tangible, observable improvements in mission performance
(i.e., that projects are meeting the cost, schedule, and performance
commitments upon which their approval was justified). For programs such as
SPS, DOD required this cost, schedule, and performance information to be
reported quarterly to ensure that programs did not deviate significantly
from expectations.14 In effect, these requirements and guidance recognize
that one cannot manage what one cannot measure.

Other Studies Reported Similar Findings and Identified Other Concerns

Shortly after receiving our draft report for comment, the department
initiated several studies to determine the program's current status, assess
program risks, and identify actions to improve the program.15 These studies
focused on such areas as program costs and benefits, planned commitments,
requirements management, program office structure, and

12In January 2001, DOD issued a change to its major system acquisition
policy requiring incremental investment management. Specifically, the policy
notes that a program's milestone decision authority must verify that each
increment meets part of the mission need and delivers a measurable benefit,
independent of future increments.

13Clinger-Cohen Act of 1996, Public Law 104-106, and OMB Circular A-130
(November 30,

2000).

14DOD Interim Regulation 5000.2-R, MandatoryProceduresforMajorDefense
Acquisition

Programsand MajorAutomated Information SystemAcquisition Programs(January 4,

2001).

15See for example, SPS ContractReview: PreliminaryReportand Status,August 1,
2001;

The PresentState oftheSPSProgram, Software Engineering Institute, October
19, 2001;

and Independent Reviewof theStandardProcurementSystemProgram,Gartner

Consulting, November 29, 2001.

systems acceptance testing. Consistent with our findings and
recommendations, these studies identified the need to

* establish performance metrics that will enable the department to measure
the program's performance and tie these metrics to benefits and customer
satisfaction;

* clearly define organizational accountability for the program;

* provide training for all new software releases;

* standardize the underlying business processes and rules that the system is
to support;

* acquire the software source code; and

* address open customer concerns to ensure user satisfaction.

In addition, the department found other program management concerns not
directly within the scope of our review, such as the need to

* appropriately staff the program management office with sufficient
resources and address the current lack of technical expertise in areas such
as contracting, software engineering, testing, and configuration management;

* modify the existing contract to recognize that the system does not employ
a commercial-off-the-shelf software product, but rather is based on
customized software product;

* establish DOD-controlled requirements management and acceptance testing
processes and practices that are rigorous and disciplined; and

* assess the continued viability of the existing contractor.

To address the many weaknesses in the SPS program, we made several
recommendations in our July 2001 report. 16 Specifically, we recommended
that (1) investment in future releases or major enhancements to the system
be made conditional on the department first demonstrating that the system is
producing benefits that exceed costs; (2) future investment decisions,
including those regarding operations and maintenance, be based on complete
and reliable economic justifications; (3) any analysis produced to justify
further investment in the program be validated by the Director, Program
Analysis and Evaluation; (4) the Assistant Secretary of

16GAO-01-682, July 31, 2001.

DOD Has Begun Addressing Problems, But SPS's Future Remains Uncertain

Defense for Command, Control, Communications, and Intelligence (C3I) clarify
organizational accountability and responsibility for measuring SPS program
against commitments and to ensure that these responsibilities are met; (5)
program officials take the necessary actions to determine the current state
of progress against program commitments; and (6) the Assistant Secretary of
Defense for C3I report by October 31, 2001, to the Secretary of Defense and
to DOD's relevant congressional committees on lessons learned from the SPS
investment management experience, including what actions will be taken to
prevent a recurrence of this experience on other system acquisition
programs.

DOD's reaction to our report was mixed. In official comments on a draft of
our report, the Deputy CIO generally disagreed with our recommendations,
noting that they would delay development and deployment of SPS. Since that
time, however, the department has acknowledged its SPS problems and begun
taking steps to address some of them. In particular, it has done the
following.

* The department has established and communicated to applicable DOD
organizations the program's chain-of-command and defined each participating
organization's responsibilities. For example, the Joint Requirements Board
was delegated the responsibility for working with the program users to
define and reach agreement on the needed functionality for each software
release.

* The department has restructured the program office and assigned additional
staff, including individuals with expertise in the areas of contracting,
software engineering, configuration management, and testing. However,
according to the current program manager, additional critical resources are
needed, such as two computer information technology specialists and three
contracting experts.

* It has renegotiated certain contract provisions to assume greater
responsibility and accountability for the requirements management and
testing activities. For example, DOD, rather than the contractor, is now
responsible for writing the test plans. However, additional contract changes
remain to be addressed, such as training, help-desk structure, facilities
support, and system operations and maintenance.

* The department has designated a user-satisfaction manager for the program
and defined forums and approaches intended to better engage users.

* It has established a new testing process, whereby program officials now
develop the test plans and maintain control over all software testing
performed.

In addition, SPS officials have stated their intention to

* prepare analyses for future program activities beyond those already under
contract, such as the acquisition of additional system releases, and use
these analyses in deciding whether to continue to deploy SPS or pursue
another alternative;

* define system performance metrics and use these metrics to assess the
extent to which benefits have been realized from already deployed system
releases; and

* report on lessons learned from its SPS experience to the Secretary of
Defense and relevant congressional committees.

The department's actions and intentions are positive steps and consistent
with our recommendations. However, much remains to be accomplished. In
particular, the department has yet to implement our recommendations aimed at
ensuring that (1) future releases or major enhancements to the system be
made conditional on first demonstrating that the system is producing
benefits that exceed costs and (2) future investment decisions, including
those regarding operations and maintenance, be based on a complete and
reliable economic justification.

We also remain concerned about the future of SPS for several additional
reasons. First, definitive plans for how and when to justify future system
releases or major enhancements to existing releases do not yet exist.
Second, SPS officials told us that release 4.2, which is currently under
contract, may be expanded to include functionality that was envisioned for
releases 5.0 and 5.1. Including such additional functionality could compound
existing problems and increase program costs. Third, not all defense
components have agreed to adopt SPS. For example, the Air Force has not
committed to deploying the software; the National Imagery and Mapping
Agency, the Defense Advanced Research Projects Agency, and the Defense
Intelligence Agency have not yet decided to use SPS; and the DOD Education
Agency has already adopted another system because it deemed SPS too
expensive.

In summary, effective investment in information technology depends on
organizations (1) justifying programs via incremental business cases that
are based on reliable data and sound analysis, (2) making decisions on
investments in programs on an incremental basis, and (3) monitoring actual
return on investment (benefits achieved and costs incurred) for each
increment and using this information to facilitate decisionmaking about
future increments. In the case of SPS, this has not occurred. While DOD has
begun taking steps to strengthen its management of certain aspects of the
program and committed to strengthening its investment

Contact and Acknowledgement

management practices, questions still remain as to what will be done and
when. To increase the chances of program success, the department must
expeditiously follow through on its stated commitments and address each of
our recommendations. If it does not, it risks acquiring and deploying a
procurement system that will not produce business value commensurate with
costs.

This concludes my statement. I would be pleased to answer any questions you
or Members of the Subcommittee may have at this time.

For further information regarding this testimony, please contact Randolph

C. Hite, Director, Information Technology Systems Issues, at (202) 512-3439,
or Cynthia Jackson, Assistant Director, Information Technology Systems
Issues, at (202) 512-5086. You may also contact them by e-mail at
[email protected] [email protected], respectively.

(310221)
*** End of document. ***