Exceptions and Selective Data Skew USPS Service Reports

In any public statement about its products or services, it’s to be expected that a private company will present information that frames those products and services as favorably as possible. Any commercial or advertisement is an example.

Whether the Postal Service should engage in such a practice is another matter. Though it’s supposed to operate in a “businesslike manner,” it’s fundamentally a government agency whose vaunted trustworthiness would make the transparency of operations and complete honesty of information no less than core values.

However, under the reign of Postmaster General Louis DeJoy, the USPS has displayed a knack for selecting and spinning facts that would make a Madison Avenue advertising executive proud. Perhaps the best example would be the weekly press releases (and subsequent Industry Alerts) claiming (as in the December 3 release) “solid service performance” and a network that’s “running smoothly.”

Customers might claim otherwise, especially commercial mailers whose experiences don’t align with the Postal Service’s portrayals. As one observer who monitors the daily volume of Presorted First-Class Mail reports, the average days to delivery was recently well over three days, and the volume taking five days is growing.

The choice of facts

The USPS offers its most detailed data in what it’s required to submit every quarter to the Postal Regulatory Commission. Looking at that (summarized in the previous issue of Mailers Hub News) would reveal that actual performance below the highly aggregated national scores is, in many parts of the country – and especially for those types of mail that are less automation-compatible or not destination-entered by the mailer – much less stellar than what the agency claims in its weekly releases.

Nationally aggregated numbers of unspecified formulation (not described as “averaged”) are one thing and may be useful in press releases, but the detailed quarterly figures for letters vs flats, for example, or by point of entry, more accurately reveal the Postal Service’s real performance.

For example, in its October 8 press release, the national “preliminary” figures declared by the Postal Service for the fourth quarter of its fiscal year were 88.0% for First-Class Mail, 92.6% for Marketing Mail, and 82.1% for Periodicals, all “against the [corresponding] USPS service standard.”

However, according to the data for First-Class Mail filed with the PRC, the three-to-five day standard was met only 87.2% of the time for Presorted First-Class Mail, and only 76.1% of the time for single-piece (retail) First-Class Mail. The Postal Service’s claim of “88.0% ... on time” clearly resulted from blending in better numbers from its achievement of the overnight and two-day delivery standards.

Similarly, though the USPS touted delivering “92.6% of Marketing Mail on time against the USPS service standard,” looking at the more detailed information shows that on-time delivery of origin-entered Marketing Mail was only 80.1% for letters and 62.4% for flats. Here again, the USPS used better figures, such as for mail SCF-entered by the mailer (96.4% for letters and 90.6% for flats) to lift the aggregated figure it reported.

The USPS did the same for Periodicals. Though its October 8 release asserted “82.1% of Periodicals [were delivered] on time against the USPS service standard,” the data filed with the PRC revealed that national service achievement for origin-entered inside-county and outside-county Periodicals was only 71.0% and 68.4%, respectively.

Observers from the commercial mailing industry would see through the Postal Service’s numbers, and readily surmise that the agency had homogenized the spectrum of actual scores for each class to produce the rosier numbers is reported. However, the target audience for the USPS press release likely wasn’t industry professionals but the general media and the public, groups that it would expect would take the numbers at face value and, in turn, accept the Postal Service’s claims of ever-improving service.

Understanding the terms

There are actually two separate terms regarding service performance that further muddy understanding the numbers.

As explained by the USPS Office of Inspector General in A Primer on Service Standards, issued September 2

“Simply put, a service standard is the number of days the Postal Service has determined it should take a specific type of mail to be delivered. The Postal Service defines a service standard as a ‘delivery performance goal for a mail class or product.’ ”

However, the USPS also sets “performance targets” which the OIG explained, “are met when a certain percentage of all mail is delivered on time.” Delivery performance against those targets is what’s reported quarterly to the PRC.

To change service standards the Postal Service must first seek an advisory opinion from the PRC but, as was demonstrated earlier this year, it’s not bound by the commission’s opinion and can do what it wants thereafter. The USPS can change performance targets on its own with no more than a decision by the Board of Governors.

Changing the rules

Before October 1, 2021, the service standard for First-Class Mail was one to three days, three to ten days for Marketing Mail, and three to nine days for Periodicals. As of October 1, the standard for First-Class Mail was extended to five days.

For FY 2020, the composite service performance targets were 96.0% for First-Class Mail and 91.8% for Marketing Mail and Periodicals. The targets for FY 2021 were not set until May 6 – over halfway through the fiscal year – and reflected DeJoy’s apparent view that if standards can’t be attained the solution is to lower the standards. As a result, for FY 2021 the composite service target for First-Class Mail is 84.88% and 86.62% for Marketing Mail and Periodicals.

The PMG has stated that, now that the service standards have been relaxed, he expects to hit a 95% performance target in FY 2022, though that’s often conditioned on the completion of his entire 10-year Plan.

While these actions regarding standards and targets draw the most attention, what is often overlooked by commercial mailers – let alone the media and public – is the business rules governing what mail is “in measurement,” i.e., included in the universe of mail on which service performance scores are based, or excluded. In the Postal Quarter IV (July-September 2021) service performance data filed with the PRC, the USPS reported on the volumes of mail that were or weren’t in “measurement” (see the top chart, right).

Generally, pieces eligible to be “in measurement” are those bearing a Full-Service intelligent mail barcode and otherwise physically compatible with automated processing. Some of those pieces are excluded for one of many reasons (see the chart below) as are pieces that are either lack an IMB, aren’t automation compatible, or that bypass automated processing. Logically, excluded mail is likely to be more difficult to process and will take longer to move from origin to delivery and, accordingly, can drag down service scores.

The USPS provided definitions for the reasons “eligible” mail is excluded from measurement:

 The current “census” method being used by the Postal Service to measure service is reliant on a unique intelligent mail barcode on each piece so, by definition, the universe of mail being measured (and reported) consists solely of automation-compatible mail bearing an IMB – the easiest and fastest mail to process.

Conversely, as the charts show, a substantial volume of mail is not “in measurement” and, as a result, the service that mail receives is neither measured nor reflected by the scores reported by the Postal Service.

This circumstance is analogous to a school district report on its students’ achievement in standardized testing when, in fact, the universe of students on which the claim is based doesn’t include those who didn’t complete the test, didn’t mark their answers clearly, or were otherwise excluded.

Accordingly, when the USPS claims any given performance against its service targets, such claims should be (but are not) qualified to note (1) that they refer to only a subset of the total volume of mail and (2) that because that subset is only the “best” mail to process, actual net performance for the class or category of mail is lower.

As noted, while the popular media and the general public accept as fully accurate the numbers in USPS press releases, many senders of commercial mail, especially of flats and Periodicals, find the agency’s claims particularly disingenuous.

Outside observers

While the PMG and his publicists can issue whatever carefully phrased press releases they want, the smoke and mirrors of such messages can’t be used with two other entities: the Postal Regulatory Commission and the Office of Inspector General.

Before the end of calendar 2021, the Postal Service must submit its Annual Compliance Review which, among other topics, includes the agency’s explanations for its service performance. This year, somewhat less than in FY 2020, the impact of the pandemic on employee availability and transportation will supply ample reasons for shortfalls in service. It will fall to the commission to determine the degree to which such factors can rightfully be used to explain service declines, as opposed to the impact of more routine failures to manage operations or adhere to operating plans and transportation schedules. The PRC’s findings will be in its Annual Compliance Determination issued before the end of March.

Meanwhile, the OIG keeps a keen eye on the USPS and regularly issues reports on its audits of the agency’s operations. Previous audits have focused on the USPS during the early months of the pandemic, on facility problems (such as in Cleveland), on failures of managers to use data tools provided, and on failures to manage dispatches and late trips.

Separately from what the Postal Service issues, the OIG also maintains on its site (https://www.uspsoig.gov/service-performance) a graphic dashboard showing USPS service performance under various criteria over six fiscal years.


As those familiar with the Postal Service’s periodic reports may recall, after the PMG reorganized the field structure from seven areas and 67 to districts into two regions, twelve divisions, and fifty districts, the agency decided to change its service reporting format to report by division. Doing so drastically reduced the granularity of the data and further obscured where service was below average.

Fortunately, the PRC quickly reminded the USPS that no such modification had been approved; the agency recast the quarterly data in the old format despite grumbles from the postal lawyers.

Now, in a new fiscal year, it remains to be seen whether the USPS will try its downsized reporting format again and, if it does, what the PRC will do in turn.

If the Postal Service’s current tendencies are any guide, whatever changes it might make likely would be to further polish the appearance of its service, not to better reflect the reality of what that service may be.

Share this post:

Comments on "Exceptions and Selective Data Skew USPS Service Reports"

Comments 0-5 of 0

Please login to comment