Can Scan Data Be Believed?

One of the better features of many USPS mail categories and services is the use of a unique barcode on mailpieces that, when scanned by Postal Service machines or people, can yield information about the item’s location (and delivery status).  Of course, the value and accuracy of that information depend on whether the scan actually was captured at the time and location reported to the customer.

The USPS regularly details on its daily Link the percentage of expected delivery scans that actually occur.  For the week ending October 16, for example, the national score was 97.06%, down 0.14% from the previous week.  The areas and districts that do the best typically score in the mid- to high-ninety percent range, meaning that the others do more poorly; the agency doesn’t disclose them.

The USPS also uses scan data internally to measure service performance and to provide visibility into whether mail is moving through processing and delivery according to workplans and delivery commitments.

The human factor

Naturally, as in any business, managers use available tools, such as scan data, to monitor operational performance and, aware that their superiors are watching as well, strive to ensure that the data reflects well.

However, in cases where timely processing or delivery is impeded by too much volume or too few people, employees feeling the pressure to meet expected service or workhour targets will be tempted to make scans represent the completion of processes (or delivery) that in fact have yet to occur.

For example, a carrier might find it expedient to scan his route’s parcels as delivered as he loads his truck in the morning.  Another might try to finish his route on time by cutting corners, such as by not delivering a parcel, instead scanning it as attempted but not delivered.  A supervisor might allow, or actively encourage, scans that create a more positive report on operational status than is really the case.  The rationale for such actions might be that it’s more important to reduce overtime or maintain scores than reflect reality.

The flaw in any such practice, of course, is the assumption that no-one will notice.  Scan data reported to Mailers Hub has often shown an item arriving repeatedly at a facility, resetting the “clock” each time to mask delays in processing.

The media

Given the current level of media attention being given the Postal Service – primarily in the wake of service issues last summer and subsequent concerns over the prompt delivery of election mail – it’s no surprise that some of that attention is being directed at scan data as a reflector of service.  In turn, if anything questionable about that data can be detected, it can be used to negatively portray USPS service reports and performance more broadly.

For example, an October 15 article in the Washington Post claims that USPS

“employees and supervisors have routinely falsified data on package deliveries, likely so they are not penalized for tardiness. ... Newly obtained data, and interviews with two former postal executives and nine current postal workers, shows that the false codes are far more varied and extensive, and suggest that they are meant to boost the agency’s performance metrics.

“Postal workers said in interviews that they have been pressured by their bosses to perform false scans increasingly over the past six months so that late packages appear to have been delivered – or are delayed because of issues at the delivery address – to ensure they don’t count against the agency’s delivery statistics.”

A Postal Service spokesperson disputed the claims, saying that “the agency maintains ‘end to end visibility’ of its packages as they are accepted, processed and delivered, and that it takes ‘improper package processing very seriously.’”


Obviously, the Postal Service wants its employees to scan mail accurately and consistently.  By its Link reports and other messages, it reminds them that customers expect both accurate tracking data and that the reported service actually is provided.

Unfortunately, just as obvious is the statistical probability that, for a variety of possible reasons, some number of employees – whether clerks, carriers, or supervisors – won’t comply.  In turn, these shortcomings are noticed, by fellow employees or by customers, and serve as the basis for the unattributed comments collected by eager reporters.

As is typically the case, public opinion is skewed by these reports, promoting the erroneous impression that the reported shortcuts and scanning failures are the norm – and the USPS does little publicly to correct that conclusion.

Share this post:

Comments on "Can Scan Data Be Believed?"

Comments 0-5 of 0

Please login to comment