The IRS’s Level of Service Measure Fails to Adequately Show the Experience of Taxpayers Seeking Assistance Over the Phone

Nina Olson, taxpayer, taxpayer advocate service, customer service

In a series of blogs I published earlier this year, Telephone Service in an Omnichannel Environment – The IRS Must Make Communicating with the IRS Over the Phone Easier for Taxpayers and Telephone Service in an Omnichannel Environment – The IRS Must Ensure Taxpayers Are Getting the Assistance They Need Over the Telephone, I explained why telephone communication is still an important way for taxpayers to get assistance from the IRS, even as the IRS expands its online self-help service options. In these blogs, I discussed how the IRS is failing to develop its telephone service as a vital part of an omnichannel communication environment, and thus failing to recognize the needs and preferences of taxpayers. In my 2017 Annual Report to Congress, I identified the limitations of the IRS’s telephone service as one of the Most Serious Problems encountered by taxpayers, and highlighted my concern that the IRS’s operational measures are overly focused on efficiency rather than the taxpayer experience. In this blog, I will further detail my concerns about the IRS’s reliance on the Accounts Management (AM) Customer Service Representative (CSR) Level of Service (LOS) as the benchmark measure to evaluate its phone service, as it can mask the struggles faced by taxpayers seeking assistance.  

As of September 15, 2018, just over two weeks before the end of Fiscal Year (FY) 2018, the IRS has received nearly 96 million calls on its toll-free lines. The IRS has gotten about 2.5 million more calls in FY 2018 than during FY 2017, an increase possibly attributable totaxpayers’ questions about changes to their tax obligations after the Tax Cuts and Jobs Act. (For a discussion of TAS’s early testing of IRS responses to taxpayers’ questions about tax reform, see Taxpayers Need More Guidance and Service to Understand and Comply With the Tax Cuts and Jobs Act.) The IRS evaluates its success in answering these calls using the LOS, a measure of taxpayer access to telephone assistors. In Filing Season 2018,  the IRS reported an LOS of about 80 percent for the first time in over ten years, and reports a fiscal year-to-date LOS of about 76 percent. These numbers seem to indicate that the agency is doing a good job communicating with taxpayers over the phone. However, when we take a closer look at the nuances of what the LOS actually measures – and importantly, does not measure – the picture is far less rosy. As I described in my recent testimony before the Senate Subcommittee on Taxation and IRS Oversight and in my June Report to Congress, the benchmark LOS measure is a very narrow one and does not reflect the taxpayer experience in the following respects:

First, the benchmark LOS publicized by the IRS reflects only the results for the IRS’s “Accounts Management” telephone lines, and does not include calls directed to the IRS’s compliance functions. In Fiscal Year 2018 through September 15, 2018, net attempts to the IRS’s “Accounts Management” lines have made up about 80 percent of the total number of calls received by the IRS. The LOS on other lines not included in the “Accounts Management” umbrella was not uniformly high. Taxpayers calling the IRS’s compliance functions to discuss payment options wait approximately 24 minutes on average to speak with a telephone assistor, and the LOS on these lines is about 53 percent for the fiscal year as of September 15, 2018. Of particular note, the “Installment Agreement/Balance Due” line has an LOS below 50 percent, meaning that more than half of all taxpayers did not receive assistance at the time of the call, and wait times for taxpayers who got through were over 27 minutes. While the “Installment Agreement/ Balance Due” line was formerly grouped in the “Accounts Management umbrella, the IRS moved it to a different group in the fall of 2016. Thus, the IRS has subsequently been able to report a higher benchmark LOS even though this benchmark reflects a smaller portion of overall calls and service remains poor on other lines.

Second, the benchmark LOS generally only reflects calls routed to a customer service representative, not to those directed to automated responses. When taxpayers call the IRS, they are greeted by an automated voice presenting them with a list of routing options. Depending on which option is selected, the taxpayer will be routed either to a live telephone assistor or to a pre-recorded automated system. Taxpayers do not have the option to speak directly to a telephone assistor if they are confused about which routing option is best for them or if they simply prefer to speak with a live human being.Thus, the LOS is based off of data reflecting where taxpayers have been directed by the IRS, not where and how taxpayers need or prefer to be assisted. If taxpayers routed to automation are unable to get the information they need, they cannot be rerouted to another location and must start their journey over. I have previously recommended the IRS institute a system similar to a 311 line, which would provide taxpayers assistance in finding the right location to get help at the inception of a call.

Of the 75.8 million net attempts routed to the IRS’s “Accounts Management” lines in Fiscal Year 2018 through September 15, only 24.5 million were answered by a telephone assistor. That’s just 32 percent of calls received on these lines. The rest were routed to automation or reflected taxpayer hang-ups. As I noted in Volume 2 of my June Report to Congress, the IRS does not have a way for the caller to indicate if the automated message has resolved the caller’s inquiry after the call as a way of measuring the effectiveness of its automated responses. Taxpayers calling the IRS have often already tried to resolve their inquiry through the IRS’s self-help solutions online, as 46 percent of taxpayers calling the IRS have checked these resources and still need assistance. Thus, if we assume that most taxpayers calling the IRS generally want to speak to an employee for live assistance, 32 percent would better reflect the service provided by the IRS than the benchmark 80 percent reported by the IRS. While some taxpayers are adequately served through automation, the stark contrast between the 80 percent LOS reported by the IRS and the 32 percent of calls actually answered by a telephone assistor makes clear the need for more reliable and robust performance measures.

Third, the reported LOS does not provide qualitative information about the assistance a taxpayer receives on a telephone call. In its Strategic Plan, the IRS describes the LOS as “the percentage of toll free callers that successfully speak to a Customer Service Representative.” However, merely being connected with a telephone assistor is not a “successful” outcome in the mind of a taxpayer, just a necessary first step in the experience. Achieving a high LOS does not mean much if the IRS is unable to answer taxpayers’ questions over the phone or guide them to an appropriate resolution of their issues. In my 2017 Annual Report to Congress, I recommended that the IRS measure its First Contact Resolution rate, the single biggest driver of customer satisfaction. TAS research shows that almost 40 percent of taxpayers calling the IRS felt one call did not fully resolve their problems. This result indicates taxpayers are not getting the full assistance they need over the phone, jeopardizing their rights to quality service and to be informed, while potentially undermining voluntary compliance.

The IRS asserts that taxpayers should “expect the same level of service when dealing with the IRS in the future as they have now from their financial institution or a retailer.” However, as long as the IRS continues to rely on the flawed LOS as a measure to assess which programs need priority attention, it will be unable to effectively identify areas to improve and live up to this standard. Continuing to have “good” LOS numbers while telephone service fails to meet IRS goals and taxpayer needs will only cause trust in the IRS to further erode. In next week’s blog, we will take a closer look beyond the LOS at the experience of taxpayers seeking assistance from the IRS, and evaluate how IRS service stacks up against other government agencies and the private sector.

Visit TaxpayerAdvocate.irs.gov or call 1-877-777-4778.

Nina Olson

Nina E. Olson, the National Taxpayer Advocate (NTA), is the voice of the taxpayer within the IRS and before Congress. She leads the Taxpayer Advocate Service (TAS), an independent organization inside the IRS that helps taxpayers resolve problems and works for systemic change to mitigate problems experienced by groups of taxpayers.

Subscribe to TaxConnections Blog

Enter your email address to subscribe to this blog and receive notifications of new posts by email.



2 comments on “The IRS’s Level of Service Measure Fails to Adequately Show the Experience of Taxpayers Seeking Assistance Over the Phone

  • “The IRS asserts that taxpayers should “expect the same level of service when dealing with the IRS in the future as they have now from their financial institution or a retailer.” This is right up there with President Trump telling world leaders at the UN that he is the greatest president the US has ever had. It’s nothing but a joke because what it should say don’t expect much. Whenever I call IRS “customer service” I don’t feel like a customer, I feel like a victim. The service ranges from poor to poorer to we are doing our best to make it worse.

  • I would love to see a study that compares this specific aspect: accuracy of info provided by IRS telephone agents (shown as a percentage) compared to accuracy of info provided by CRA telephone agents (shown as a percentage). In my 20 years practicing US tax in Canada, the offhand informal feedback I’ve received from clients is that CRA is significantly more diligent than IRS in providing accurate info by phone to taxpayers. Would love to see a formal scientifically conducted survey

Comments are closed.