Greater Emphasis on Testing Needed To Make Computer Software More Reliable and Less Costly

GAO evaluated the adequacy and effectiveness of federal agencies' software testing practices.

Federal agencies spend billions of dollars each year to develop and maintain computer programs used for business applications. GAO found that federal agencies generally are not managing the software testing process effectively to help ensure that the software performs its intended functions accurately and reliably. GAO believes that undetected software errors are costing agencies millions of dollars unnecessarily because such errors: (1) cost more to correct after software becomes operational; (2) often require expensive manual processing or other corrective action; and (3) sometimes result in the loss of financial assets. Too often, decisions on the amount and extent of testing depend on the discretion of individuals or on time and staff limitations, not on testing policies and requirements for the agency. Software users do not always participate in the testing process, and user roles and responsibilities in testing are not always clearly defined, which contributes to the development of software that does not meet user needs. Most agencies do not use software problem data to evaluate the overall effectiveness of the testing process in producing quality software. Further, federal agencies have not taken advantage of software technology that could improve the testing process.

Recommendations

Our recommendations from this work are listed below with a Contact for more information. Status will change from "In process" to "Open," "Closed - implemented," or "Closed - not implemented" based on our follow up work.

Director: Howard G. Rhile Jr Team: General Accounting Office: Information Management and Technology Division Phone: (202) 512-6418