The Practitioner Priority Service Was Successful in Its
First Year of Operation, but Action to Reduce Caller Wait Time Is Needed
July 2003
Reference
Number: 2003-30-143
This report has cleared the Treasury
Inspector General for Tax Administration disclosure review process and
information determined to be restricted from public release has been redacted
from this document.
July
8, 2003
MEMORANDUM FOR
COMMISSIONER, SMALL BUSINESS/SELF-EMPLOYED DIVISION
FROM: Gordon C. Milbourn III /s/ Gordon C.
Milbourn III
Assistant Inspector General
for Audit (Small Business and
Corporate Programs)
SUBJECT: Final Audit Report - The Practitioner
Priority Service Was Successful in Its First Year of Operation, but Action to
Reduce Caller Wait Time Is Needed (Audit # 200230034)
This
report presents the results of our review of the new Practitioner Priority Service (PPS) that was implemented by the
Internal Revenue Service (IRS) in January 2002 to provide a single nationwide
toll-free telephone number as the first point of contact for tax professionals
with questions about their clients’ tax accounts. Tax professionals are key external stakeholders in the IRS’
efforts to effectively administer the nation’s tax laws. More than 53 percent of the approximately
130 million taxpayers who filed individual income tax returns in Calendar Year
2001 used paid preparers.
The PPS replaced the IRS’
Practitioner Hotline (PHL). The
consolidation of the former PHL services from 37 locations into 5 PPS sites was
expected to provide improved telephone access, reduced wait times, and improved
overall consistency and quality of service.
The overall objective of this review was to evaluate whether the PPS is
providing tax professionals with easy toll-free telephone access to tax account
information needed on behalf of their clients.
We initiated this review as part of our continuing audit coverage of the
IRS’ various toll-free product lines.
In summary, the performance
measures used by the IRS show the PPS performed reasonably well in its first year
of operation. Customer Service
Representatives assisted approximately 83 percent of the callers and achieved a
73 percent accuracy rate. The PPS
customer satisfaction survey results indicate that practitioners gave the
service they received an average overall satisfaction rating of 4.12 on a
5-point scale.
In
June 2002, the IRS initiated an internal review of the PPS program and found
that the geographic routing of calls had created significant imbalances in wait
times, staffing requirements, and call volumes among the five PPS sites. To alleviate these imbalances, the IRS
implemented Intelligent Call Routing (ICR) in February 2003 to route calls to
the next available assistor without regard to the originator’s area code.
Our
analysis of the telephone system data showed that although three of the five
PPS sites had received similar call volumes, the wait time at one was
disproportionately higher than the wait time at the other two centers. While the implementation of ICR should help
to achieve a more even distribution of calls among the PPS sites, it alone will
not ensure that callers are connected to
an assistor in a timely manner.
We recommended that the IRS
determine the length of time that most tax professionals are willing to wait
for assistance and, based on this tolerance level, develop its staffing models
around a service level performance goal that prescribes the percentage of
callers that should receive assistance within that specified time period. In addition, the IRS should pursue the necessary
technology to enable it to provide a projected wait time message for callers
when the call arrives at the call center.
Management’s
Response: Management’s
response was due on June 27, 2003. As
of July 2, 2003, management had not responded to the draft report.
Copies of this report are also being sent to the IRS
managers who are affected by the report recommendations. Please contact me at (202) 622-6510 if
you have questions or Richard J. Dagliolo, Director (Submission Processing), at
(631) 654-6028.
The
Practitioner Priority Service Performed Well During Its First Year of Operation
The Level of Service Goal Should Be Revised to Assure Minimal Wait Time for Practitioners
Appendix I – Detailed Objective, Scope, and Methodology
Appendix II – Major Contributors to This Report
Appendix III – Report Distribution List
Tax professionals are key external stakeholders in the Internal Revenue Service’s (IRS) efforts to effectively administer the nation’s tax laws. Practitioners help to resolve taxpayer account issues timely and disseminate vital tax administration information to their clients and members of their communication network. More than 53 percent of the approximately 130 million taxpayers who filed individual income tax returns in Calendar Year (CY) 2001 used paid preparers.
The IRS implemented the Practitioner Priority Service (PPS) on January 2, 2002, to provide a single nationwide toll-free telephone number as the first point of contact for tax professionals with questions concerning their clients’ tax accounts. The IRS also committed to providing them the best service possible. The PPS line is staffed by specially trained Customer Service Representatives (CSR) that are located at five IRS campuses. These CSRs receive incoming calls to the PPS line from 7:30 a.m. to 5:30 p.m. (local time) Monday through Friday of each week.
The PPS replaces the IRS’ former Practitioner Hotline (PHL) service that was offered at 37 locations through 46 different local (i.e., nontoll-free) telephone numbers. The consolidation into five sites was expected to provide improved telephone accessibility and reduced wait times, as well as improved overall consistency and quality of service. Each of the five sites handled inquiries about both individual taxpayer accounts and business taxpayer accounts.
The IRS’ SB/SE Division manages and operates the PPS. The SB/SE Division’s Strategic Plan for Fiscal Years (FY) 2002 and 2003 identified Improve Business Results by Revamping Our Operational Practices and Processes as one of its strategies for accomplishing its mission. The PPS was one of the improvement projects in support of this strategy.
To perform this audit, we analyzed IRS telephone system data and interviewed the SB/SE Division’s Customer Account Services staff in Oakland, California; Cincinnati, Ohio; Memphis, Tennessee; and Ogden, Utah. In evaluating the accessibility of the PPS, we relied upon the telephone system data provided by the IRS’ Joint Operations Center in Atlanta, Georgia. Because the Aspect telephone system automatically records specific information (e.g., queue time) about how each incoming call was handled and there is no source documentation, we could not establish the reliability of the telephone system data. Except for our inability to validate these data, we conducted this audit in accordance with Government Auditing Standards between June 2002 and March 2003.
Detailed information on our audit objective, scope, and methodology is presented in Appendix I. Major contributors to the report are listed in Appendix II.
During the PPS’ first year of operation, the various measures used by the IRS show that it performed reasonably well in terms of access to assistance, the quality of service provided, and customer satisfaction.
Of the 703,172 call attempts to the PPS toll-free line in CY 2002, the CSRs answered 552,623 calls in providing a CSR Level of Service (LOS) of approximately 83 percent. Most of the remaining calls (more than 140,000) to the PPS line were abandoned. The CSR LOS achieved on the PPS line exceeded the 80 percent goal that was established by IRS management for FY 2002. It also exceeded, by approximately 13 percent, the combined CSR LOS that was achieved on the major IRS toll-free lines during FY 2002.
The IRS monitors the quality of service provided on the PPS line through its Centralized Quality Review System (CQRS) staff in Philadelphia, Pennsylvania. The CQRS data showed that the accuracy rate of the service provided on the PPS line in FY 2002 was 73 percent. While there was no specific quality goal established for the PPS line, the quality rate achieved was comparable to the quality rate achieved on the IRS’ other toll-free lines for taxpayer account issues.
Customer satisfaction surveys, administered for the IRS by an outside vendor, showed that practitioners were generally satisfied with the services they received on the PPS line. For example, the first survey results showed that the PPS customers gave the service they received an average overall satisfaction rating of 4.12 on a 5-point scale. Specifically, 53 percent of the customers surveyed gave the PPS an overall rating of 5 (“completely satisfied”) and 25 percent gave it an overall rating of 4 (“somewhat satisfied”).
During the implementation of the PPS in CY 2002, the practitioner community expressed concerns about the limitations on the scope of the services, the loss of their familiar contacts, long wait times, and assistors with limited knowledge. These concerns prompted the IRS to launch an internal review of the PPS program in June 2002.
The internal review team identified significant imbalances in wait times, staffing requirements, and call volumes among the five PPS sites that were attributable to the geographic routing of calls. To accommodate members of the practitioner community who expressed a desire to maintain the level of personal attention that they had grown accustomed to when using the former PHL, the IRS had initially decided to use a geographic routing plan. Intelligent Call Routing (ICR), which routes a call to the next available assistor without regard to the area code of the originator, was not used in an attempt to provide practitioners with relationship-based assistance with the same team of CSRs.
Our analysis of incoming calls during the 6-month period from April 1 to September 30, 2002, also confirmed that the geographic routing plan had produced an uneven call distribution pattern among the 5 PPS call sites. As illustrated in Figure 1, one call site received substantially more calls and another site received significantly fewer calls than the remaining three call centers.
Figure
1. Distribution of Total PPS Call
Volume by Hour of Day
April
– September 2002
Figure 1 was
removed due to its size. To see Figure
1, please go to the Adobe PDF version of the report on the TIGTA Public Web
Page.
The internal review team recommended instituting ICR across all five sites to alleviate the imbalances in service and staffing. The IRS implemented ICR on February 24, 2003.
The internal review team also recommended developing and implementing a communication plan to inform and educate the practitioner community regarding the scope and availability of the PPS, improving the handling of calls that are outside the scope of the PPS by establishing priority call routing from the PPS line to IRS toll-free lines in Compliance Services, and planning for specialization by type of taxpayer account as soon as possible. Under the last recommendation, three PPS sites would specialize in handling inquiries about individual taxpayer accounts, and two PPS sites would specialize in handling inquiries about business taxpayer accounts.
We concur with each of the above internal review team recommendations.
During 2002, calls to the PPS line were generally routed to designated campus sites based on the caller’s area code. If the prerouting status checks showed that the average queue time at the designated site was above a specified threshold, the caller was given the option to have the call routed to another site that could provide faster service or routed to the original designated site.
The IRS had established an 80
percent CSR LOS goal for the PPS line for FY 2002. However, the IRS goal did not include
a strategy for assuring that the callers were connected to a CSR within a
predefined time period that reflected caller
tolerance or expectation for waiting.
As a result, one-third of the callers to the PPS waited over 5 minutes
for assistance while another 13 percent abandoned their calls before receiving
assistance.
Across all call center industries, the average queue time
is 150 seconds. However, a report
issued as part of the National Performance Review stated that, “Customer
queue time approaching 1 minute is universally associated with unacceptable
levels of customer service, and is strongly correlated to lost business, lost
opportunities and high levels of customer dissatisfaction.”
In the first PPS customer satisfaction survey report issued
in November 2002, the “length of wait to talk to a representative” was the
factor that received the lowest satisfaction rating, scoring 3.46 on the
5-point scale. The report also showed
that 29 percent of the tax professionals surveyed were dissatisfied with their
wait time and that another 18 percent indicated that they were neither
satisfied nor dissatisfied. The survey results also showed that only
one-third of the tax professionals were completely satisfied with the length of
time they waited to talk to a representative.
An additional 21 percent were somewhat satisfied with the wait
time.
However, the survey was not designed to measure customer
expectations for waiting for assistance.
The survey only asked the participants to rate their satisfaction with
the length of their wait to talk to an IRS representative. It did not capture the customer’s perception
of how long they waited nor whether the IRS had exceeded, met, or failed to
meet their expectations.
For the 6-month period from April 1 to September 30, 2002, 419,269
calls were routed to the 5 PPS call centers.
The CSRs handled 349,387 of these calls (83 percent) and another 13,760
calls (3 percent) were handled by an automated voice message system. During this period, 56,122 calls were
abandoned.
In
analyzing the PPS telephone data, we used the “call abandonment rate” as a
guide for estimating a tax professional’s tolerance for waiting for assistance. As
shown in Table 1, 30,981 (55 percent) of the abandoned calls were abandoned in
3 minutes or less. This may indicate
that, for most tax professionals, the tolerance for waiting to be connected to
an assistor does not exceed a threshold of 3 minutes.
Table 1.
Abandoned PPS Calls by Caller Wait Time
April – September 2002
|
|
Number of |
Abandon Rate |
|---|---|---|
|
30
seconds or less |
7,743 |
14% |
|
31
seconds to 1 minute |
11,054 |
20% |
|
Greater
than 1 minute, less than/equal to 2 minutes |
7,691 |
14% |
|
Greater
than 2 minutes, less than/equal to 3 minutes |
4,493 |
8% |
|
Greater
than 3 minutes, less than/equal to 5 minutes |
6,586 |
12% |
|
Greater
than 5 minutes |
18,555 |
33% |
|
Total |
56,122 |
100% |
Source: TIGTA Analysis of the IRS’ Telephone System Data.
Of the calls received at the PPS call centers, CSRs
answered 39 percent of the calls in 3 minutes or less, while the delay for 34
percent of the callers was over 5 minutes.
The average delay before speaking to an assistor fluctuated by as much
as 50 percent or more among the 5 PPS sites.
Depending on the call site, callers making an inquiry about an
individual taxpayer account waited an average of 4 to 8 minutes to talk to an
assistor. Those callers needing
assistance with a business taxpayer account waited an average of 3 to 7
minutes.
A caller’s tolerance for waiting can be influenced by a
variety of factors. When callers are
connected to a PPS call site, for example, there are no announcements made
regarding the expected wait time.
Uncertain waits tend to seem longer than defined waits. Some professional service organizations bill
their clients on an hourly basis in quarter-hour increments. The fact that their time is billable to
their clients may significantly influence their tolerance for waiting. Other factors that can affect a caller’s
tolerance for waiting include their expectations based on prior contacts, their
availability to wait, and who is paying for the call.
Although the use of ICR and the implementation of skill specialization should enable the IRS to achieve a more even distribution of incoming calls among the PPS sites, it will not necessarily ensure that callers are connected to a CSR in a timely manner. We found that among the three PPS centers that received similar call volumes, as previously illustrated in Figure 1, the wait time for individual tax account assistance at one center was disproportionately higher than that at the other two centers.
Connecting tax professionals to
a CSR within a prescribed time period will not happen by chance; it has to be
planned. Call center industry experts consider the selection of an appropriate
service level to be the first step in effective planning, staffing, and
management decisions. In the context of
the industry definition, service level is a performance measure that reflects
the percentage of callers that spoke with an assistor within a specified amount
of time. Across the call center
industry, service level is preferred over the percentage of calls answered and
other measures because it gives the clearest indication of what callers
experience when they attempt to reach the call center. Service level provides senior management and
customers with a gauge for determining how accessible the call center is. It also reflects an organization’s desire to
make it easy for customers to do business with them. A concrete service level objective provides the basis for meeting
or exceeding customer expectations. An
appropriate service level for any organization should:
·
Meet the
callers’ needs and expectations.
·
Keep call
abandonment at an acceptable level.
·
Minimize agent
burnout and errors.
·
Minimize
expenses.
Any interaction with a customer should be measured against inferred or known customer expectations. Inferred customer expectations may be based on best-in-class service for a particular industry or arbitrarily based on cost constraints and management’s best guess at what constitutes world-class service. Known customer expectations are derived from surveys and unsolicited customer feedback.
A poor service level is likely to result in customer frustration and
lead to a high rate of abandoned calls.
Customer frustration can also have a negative impact on employee
satisfaction and call handle time if employees constantly spend time listening
to customer criticisms of their experience and making necessary apologies. When employee satisfaction deteriorates,
burnout and turnover fluctuate, and recruitment and training costs increase.
To determine if
the IRS would need to hire additional staffing to answer 80 percent of the PPS
calls within 180 seconds, we input the average volume of calls received during
the peak calling hours into an Erlang C Call Traffic Model calculator. As shown in Table 2, 80 CSRs were the
maximum number required to answer the hourly call volume.
Table 2. PPS
Staffing Requirements per Hour
to Answer All Calls Within 3 Minutes
Hour
|
Average Volume |
CSRs Required |
|---|---|---|
|
9 a.m. |
205 |
42 |
|
10 a.m. |
300 |
60 |
|
11 a.m. |
370 |
73 |
|
12 p.m. |
349 |
69 |
|
1 p.m. |
370 |
73 |
|
2 p.m. |
406 |
80 |
|
3 p.m. |
374 |
74 |
|
4 p.m. |
342 |
68 |
|
5 p.m. |
230 |
47 |
Source: TIGTA Analysis of the IRS’ Telephone System Data.
The 3 call sites that we
visited employed 131 CSRs on the day shift and 70 CSRs on the night shift. Therefore, the IRS already has the staffing
that it needs to answer 80 percent of the average call volume received on the
PPS line within 180 seconds. Only 1
additional CSR would be required to answer 85 percent of the calls in 180
seconds.
To further honor the IRS’ commitment to provide tax professionals the
best service possible by reducing caller wait time, the Director, Customer
Account Services, SB/SE Division, needs to:
1. Establish a service level goal for the PPS line that more closely mirrors the call center industry definition of service level (i.e., the percentage of incoming calls that are answered within a specified threshold). This can be accomplished by conducting a service-level elasticity study to determine the length of time that most practitioners are willing to wait for a representative to answer and determine if customer satisfaction with wait time affects their rating of how well the CSR handled the call. As part of the study, the SB/SE Division needs to:
a.
Modify the customer satisfaction survey to include
questions that measure perceived wait time and customer expectations. To measure perceived wait time, for example,
the question might ask, “How long would you estimate you waited on the
telephone before speaking with a representative?” The question should be followed with choices such as: (a) 0 to 30
seconds, (b) 31 to 60 seconds, (c) 2 to 3 minutes, (d) 4 to 5 minutes, or (e)
more than 5 minutes. To measure
customer expectations, the question might ask, “Was your wait shorter, longer,
or as long as you expected?”
b. Gain insight on perceived versus actual wait time and whether it affects customer satisfaction by analyzing the telephone system data to determine actual customer wait time and comparing it with the survey results.
Management’s Response: Management’s response was due on June 27, 2003. As of July 2, 2003, management had not responded to the draft report.
2. Include the newly defined service level goal in the formation of assumptions to develop staffing models to deliver the service.
3. Pursue the necessary technology to be able to provide a message to practitioners, when the call arrives at the call center, estimating how many seconds or minutes they are likely to wait before speaking with a CSR.
Appendix I
Detailed Objective, Scope,
and Methodology
The overall objective was to evaluate whether the Internal Revenue Service’s (IRS) new Practitioner Priority Service (PPS), which was launched in 2002, is providing tax professionals with easy toll-free telephone access to tax account information needed on behalf of their clients. We initiated this review as part of our continuing audit coverage of the IRS’ various toll-free product lines. In accomplishing this objective, we:
I. Identified criteria to use in assessing the adequacy of the access provided to those who contacted the PPS.
A. Researched the Internal Revenue Manual and other IRS guidelines to determine the goals established by the IRS for providing accessible telephone service.
B. Reviewed the IRS’ business rules for routing calls to the PPS call centers.
C. Reviewed call center industry publications to identify guidelines for establishing call center objectives and measures related to accessible telephone service.
D. Reviewed customer satisfaction survey results for caller concerns and expectations.
II. Evaluated whether the IRS consistently delivered access to practitioners whose calls to the PPS were geographically routed to one of five call centers.
A. Analyzed the Aspect Call Detail Table data and calculated the average amount of time (including ring, queue, and delay times) callers waited to be connected to a Customer Service Representative (CSR) at each call center.
B. Evaluated whether the IRS’ CSR Level of Service (LOS) goal for the PPS reflected caller tolerance and/or expectation for waiting.
1. Analyzed the Aspect Call Detail Table data and stratified the volume and percentage of calls abandoned by selected time intervals to estimate caller tolerance for waiting.
2. Reviewed the results of the PPS customer satisfaction survey to determine if practitioners were satisfied with the amount of time they waited for assistance.
C. Evaluated whether the CSR LOS calculation logically represented the percentage of calls to the PPS that assistors answered.
1. Reviewed the Aspect Threshold Record for each of the three call centers visited to determine if the threshold settings were consistent.
2. Interviewed system analysts at the call centers and/or the Joint Operations Center (JOC) to determine whether periodic changes were made to threshold settings, the reason for any changes, and whether a record was kept of those changes.
3. Interviewed system analysts at the call centers and/or the JOC to determine the threshold for counting a call as a primary abandon. Determined the data sources for the network and site-level primary abandon calls.
4. Compared selected call information reported on the IRS Snapshot Report to telephone system reports and the Workload Inventory Tracking System report. Determined the reason for any significant differences in the reported call information.
D. Determined if the PPS call centers consistently handled calls in the queue.
1. Interviewed system analysts to determine their method for managing incoming call traffic and for determining necessary staffing adjustments.
2. Determined the specific informational/noninformational messages that are played while callers are in the queue and determined the timing of these messages.
3. Compared copies of the Call Control Tables for the three call centers visited to determine whether there were any inconsistencies in the placement of messages played to callers in the queue.
E. Evaluated whether the IRS could answer 80 to 85 percent of its average hourly call volume without hiring additional staff.
1. Analyzed the Aspect Call Detail Table data to determine the time period during which the highest volume of calls arrived at the call centers.
2. Analyzed the Aspect Call Detail Table data and calculated the average call volume for each hour of the peak calling period.
3. Analyzed the Aspect Call Detail Table data and calculated the average time CSRs spent handling an inquiry.
4. Computed staffing requirements using an Erlang C Call Traffic Model calculator and selected assumptions for answering 80 to 85 percent of the average call volume in 180 seconds.
5. Obtained the number of CSRs used to staff the PPS at the three centers we visited.
Appendix II
Major Contributors to This Report
Richard J. Dagliolo, Director
Philip Shropshire, Director
Frank J. Dunleavy, Acting Director
William E. Stewart, Audit Manager
Una K. Smith, Senior Auditor
Denise M. Gladson, Auditor
Marjorie A. Stephenson, Auditor
Appendix III
Commissioner N:C
Deputy Commissioner for Services and Enforcement N:SE
Acting Deputy Commissioner, Small Business/Self-Employed Division S
Director, Customer Account Services, Small Business/Self-Employed Division S:CAS
Chief Counsel CC
National Taxpayer Advocate
TA
Director, Legislative Affairs CL:LA
Director,
Office of Program Evaluation and Risk Analysis
N:ADC:R:O
Office of Management Controls N:CFO:AR:M
Audit Liaisons:
Commissioner, Small Business/Self-Employed Division S
Director, Customer Account Services, Small Business/Self-Employed Division S:CAS