The Internal Revenue Service’s (IRS) toll-free telephone system is the cornerstone of its customer service operations. Historically, the IRS has struggled to effectively meet the annual customer demand for its toll-free telephone services. The largest part of that demand occurs during the first three and one-half months of each calendar year, which is generally defined as the "filing season" for individual taxpayers. During the 1999 Filing Season, taxpayers attempted over 57 million calls to the IRS’ three main toll-free telephone lines.

Providing top quality service to each taxpayer is one of the strategic goals of the IRS’ modernization plans. Likewise, Customer Service management has committed to becoming "a World Class Customer Service Organization." In 1998, the IRS began a multi-year plan to modernize its toll-free operations. Some of the planned changes were implemented in 1999. Thus, the audit was conducted to determine whether the IRS’ strategy to improve telephone service resulted in an increased level of quality customer service in an economical manner.

The IRS was not able to satisfactorily handle the level of customer demand on its toll-free telephone lines during the 1999 Filing Season. Although the IRS implemented a major technological change and numerous initiatives that were intended to enhance customer service, far too many taxpayers received busy signals or abandoned their calls after being placed on hold. For the 3 main toll-free telephone lines, for example, about 19.5 million calls resulted in busy signals during the 1999 Filing Season - almost 16 million more than during the previous filing season. Callers abandoned another 8.4 million calls before reaching a Customer Service Representative (CSR). The Level of Service (LOS) provided to taxpayers on these 3 lines for the 1999 Filing Season declined to 51 percent from 73 percent for the 1998 Filing Season.

Management attributed these difficulties to such issues as start-up problems, tax law changes, and new telephone demand. At the same time, the IRS shifted to a balanced measures approach which de-emphasized production and put a greater emphasis on employee satisfaction.

The cost to provide toll-free telephone service during the 1999 Filing Season increased while productivity decreased. The IRS’ labor costs to handle toll-free calls during the first 6 months of Fiscal Year (FY) 1999 were more than $175 million as compared to about $145 million during the same period in FY 1998. During this same 6-month period, the productivity of the individual CSRs decreased by approximately 16 percent from the prior year. As a result, the IRS’ labor costs increased by $1.30 per call (approximately 19 percent). Beyond the Customer Service costs, enforcement resources used to supplement the toll-free program cost the government approximately $314 million in lost revenue that could have been collected.

The service level decline can be attributed, in part, to the start-up problems caused by many technical and operational changes. However, another key factor was that Customer Service management entered the 1999 Filing Season without developing an integrated plan to effectively align the technology and human resource management changes. Further, the technological solution was implemented without effectively addressing fundamental operating problems.

Implementation problems with the IRS’ new national call routing system kept taxpayers from having improved access to the toll-free telephone system. When the problems continued, the IRS reverted to the call routing method used in prior years for the initial routing of calls. However, the problems inherent in this prior method reduced the advantages to be gained by the newer system. In addition, the new technology was implemented without automated controls that would have promptly detected problems.

The forecasting process to determine the staff resources needed to meet estimated customer demand was seriously flawed. Delays and inaccuracies in staffing schedules resulted in an insufficient number of adequately trained CSRs being available when needed. To compensate for these problems, a significant number of Compliance enforcement resources (i.e., Examination and Collection Division staffs) were assigned to answer the telephones. The estimated staff costs alone for these resources were almost $21 million. An additional minimum of nearly $314 million of tax revenue was also uncollected. Further, Customer Service management did not employ sufficient "call avoidance strategies" to reduce its reliance on CSRs.

Several conditions require attention to improve efficiency and reduce costs. These conditions are in the overall organization structure and the work management practices.

The IRS’ ability to effectively and efficiently provide toll-free telephone service is hindered by the large number of call centers it uses. Whereas 1 to 3 call centers is viewed as optimum by businesses recognized as providing world class telephone service, the IRS has 30 toll-free call sites. In addition to the administrative and management overhead associated with so many sites, these sites were generally located in or near large metropolitan areas where labor and space costs tend to be higher.

Some problems exist specifically because of the large number of call sites. First, it makes the complex task of scheduling resources to meet the forecasted call demand even more complicated. Second, with so many labor markets, some sites experience serious recruiting problems. Finally, the lack of standardized equipment among call sites made the technology changes more complicated. Even though substantial site consolidation is the industry trend, the IRS plans to eventually discontinue only 5 of the 30 sites.

CSR productivity in answering taxpayer calls declined 16 percent between FY 1998 and FY 1999. While Management Information System data for determining the cause of this decline are limited, some productivity analyses are possible. For example, six call sites tracked CSR time when the CSRs were actually signed on the system. The data show that these CSRs handled calls only 49 percent of their sign-on time. The other 51 percent was spent in waiting for calls, wrapping up prior calls, taking lunch and other breaks, attending meetings, etc. Thus, if a CSR was signed on the telephone system for a typical eight and one-half hour day, that CSR would actually speak with taxpayers about four hours a day.

Management at several call sites had concerns with the effect that some types of "down" time may have had on the LOS. However, the IRS’ agreement with its national labor union limited the ability of Customer Service management to use certain types of "down" time in evaluating individual CSR performance.

Another factor affecting service levels was expanding from a 96-hour week to a full 168-hour week. Staffing was shifted to cover the expanded hours, so fewer people were available during peak periods. However, less than four percent of the calls to the three main toll-free lines were received during the expanded hours.

When taxpayers did speak with a CSR, the quality of service and customer satisfaction may have been less than satisfactory. The IRS has a new Centralized Quality Review System. Likewise, the IRS has new programs to expand services to some other taxpayers requiring toll-free assistance: the Quality Assurance Program, the Problem Resolution Program (PRP), and the Spanish Language Program. These new systems and programs also encountered problems.

Taxpayers who called the IRS with tax law questions during the 1999 Filing Season received correct answers only 72 percent of the time, as compared to 79 percent last year. The lower accuracy rate could be attributable, in part, to the number of 1999 tax law changes. However, it may also represent an actual decline in quality since managers at nearly half of the 20 call sites surveyed were concerned with the overall ability of their new recruits to professionally and accurately answer telephone calls.

However, even the precision of the 72 percent accuracy rate is questionable. For example, during the 1999 filing season the IRS monitored less than 25 percent of the required number of tax law calls that the sampling plan developed by IRS statisticians required to be monitored. In addition, the IRS performed no corporate monitoring of calls received from Spanish-speaking taxpayers or calls received during the extended service hours that were extensively staffed with new and inexperienced CSRs.

A relatively small percentage of the calls that reached a CSR on the new PRP line involved unresolved tax problems that met PRP criteria. At least one site serviced these non-PRP calls rather than transferring the calls to the appropriate line. This may be causing other callers with legitimate PRP concerns to be unable to reach a CSR.

The IRS did not provide Spanish-speaking taxpayers with effective access to toll-free telephone services. The IRS has a limited number of Spanish-speaking CSRs who often handled English-language calls. Therefore, only a very small percentage of calls from Spanish-speaking taxpayers were successfully routed to Spanish-speaking CSRs. Spanish-speaking callers were frequently told to call back with an interpreter.

The IRS needs different management approaches to provide effective and economical world class toll-free telephone service. A systems management approach throughout the entire Customer Service organization is needed to ensure effective and continuous improvements. Productivity should be improved by evaluating the number of call sites, more timely and accurately conducting resource planning, reducing hours of "live" assistance, and providing managers with tools to monitor CSR productivity. Additionally, quality measurement and customer service can be improved by ensuring that statistically valid samples are taken. Finally, customer service can be improved by providing better access to PRP and Spanish-language services.

Management’s Response: IRS management agreed that the performance of the toll-free telephone system during the 1999 Filing Season fell short of their goals and has agreed to implement some of our recommendations. The Customer Service management structure has been realigned; a study group has been formed to develop criteria for determining which call sites receive more hiring authority than others; program changes have been made to increase the number of refund calls handled by automation; portions of the Union agreement were renegotiated to provide call site managers with more effective tools for monitoring assistor "idle" time; additional quality reviewers are being hired; a bilingual quality review center is being established; and several actions are being taken to make toll-free telephone services more accessible to Spanish-speaking taxpayers.

However, management did not agree to implement several recommendations. They also expressed the opinion that the report did not adequately acknowledge the significant challenges the IRS faced during the 1999 Filing Season, the steps taken during the 1999 Filing Season to correct the problems encountered, and the corrective actions the IRS completed which resulted in improved service during the current filing season.

Management’s complete response to the draft report is included as Appendix VI.

Office of Audit Comment: The Office of Audit concurs that there were significant challenges and changes in 1999. Progress is being made in several areas. At the same time, IRS management did not address, or only partially addressed, some issues and recommendations. In some cases, the Office of Audit will require further clarification and information from IRS management.