Back in 2018, I ran into a friend at the local post office and we struck up a conversation about the news of the day: the explosive devices that had been mailed to national political figures. We were amazed at how fast DHS and USPS identified the person who was sending those packages through the mail. It got us thinking about how that information was put together — and where records management came into play.
Here’s Part of That Conversation
“The United States Postal Service scans and creates digital, bar-coded images that are stored in an electronic repository. These images were instrumental in helping to locate the person who mailed the explosive packages to political figures. What’s remarkable is how efficient the recorded image search must have been in order to quickly identify the origin source of the mail and create a correlation between the two.”
So, What Went Into This Discovery?
First, from a records management perspective, the recorded images must have been captured using optical character recognition software, enabling search within the image object and supporting auditability. That means the associated metadata had to include a date and time stamp, an estimated date range for inventory and retention purposes, and some type of indicator of the sending or receiving postal facility for inbound or outbound processing.
It’s likely that this data collection occurs daily and, when matched with associated video footage, helped provide a more exact date and time stamp for the receipt of the suspicious packages into the outgoing postal facility.
Applying Some Generally Accepted Records Management Principles
Second, information traceability allowed postal service employees to locate the appropriate record index to narrow down the search. This was likely done with a unified set of search parameters across other information repositories for cross-referencing and indexing, along with a metadata profile. The principles of records transparency and availability were exercised.
Finally, the records were available — which suggests the Postal Service has the means to identify, locate, and extract recorded metadata and pull the actual record when an emergency condition arises. The recorded information adhered to the appropriate authorities governing the management and control of postal records through data collection, migration to storage, and final assignment of a retention schedule.
Postal Service Record Keeping
Federal recordkeeping requirements that govern the United States Postal Service do not allow for collection, management, and storage of information that relates to or can be used to identify individuals, except for Postal Service operations and law enforcement purposes (see the USPS general policy overview). Title 39 of the Code of Federal Regulations provides the legal authorization of the postal inspection service that enabled this information to be captured and shared with law enforcement.
The Postal Service routinely scans parcels entering the mail system and stores that data in dedicated repositories. Those scanned images became official records because they are evidentiary of a common Postal Service business process. Federal records may come in any format — text, images, documents, video, or audio.
When I wrote about this in 2018, the takeaway was simple: any electronically captured data can become a business record, given the right context and application of records lifecycle principles. Strong recordkeeping is a tremendous asset in critical decision-making, providing evidence of business functions and activities that can support legal discovery and decisions.
What none of us fully appreciated at the time was how quickly every federal agency would be forced to operate the way USPS did in that story — at industrial scale, fully electronic, with metadata doing the heavy lifting.
Every Agency Is Managing Records at USPS Scale Now
The regulatory ground has shifted. Under OMB/NARA memorandum M-23-07, the National Archives and Records Administration stopped accepting transfers of permanent or temporary records in analog formats as of June 30, 2024. Records are accepted only in electronic format, with appropriate metadata. Agencies must manage permanent records electronically, and permanent records created in analog formats must be digitized in accordance with NARA’s digitization standards under 36 CFR Part 1236, Subpart E before transfer. Electronic recordkeeping is no longer a modernization goal. It is the baseline.
And the baseline keeps growing. The volume of federal information expands every year— an ever-widening deluge of emails, chat messages, text files, and collaborative content. NARA’s Capstone approach, which began as an email retention framework, has been expanded under NARA Bulletin 2023-02 to cover other types of electronic messages — instant messages, texts, and chats — in support of the Electronic Message Preservation Act. Every one of those messages is potentially a federal record.
The old failure mode was a record room with 300 unsorted boxes. The new failure mode is the same room in digital form: shared drives, mailboxes, and collaboration platforms full of content nobody has classified, scheduled, or made findable. The USPS story worked because metadata, indexing, and retention discipline were applied at the moment of capture. Most agencies are nowhere near that — and the volume has outrun what any human team can classify by hand.
Where AI Comes In: Classification, Retention, and Legal Holds
This is why NARA is openly discussing a “zero click” future for records management — records managed without human intervention, with AI and machine learning trained to look at vast volumes of information and answer the fundamental questions: How long should this be kept? Is anything in this collection historically valuable? In many cases, NARA wants to take the burden of digital records management off rank-and-file employees altogether, so the system handles the lifecycle the way USPS’s repository handled those parcel images.
In practice, AI is already earning its keep in three areas of federal records management:
- Classification and retention. AI-assisted classification applies file plans and retention schedules automatically at capture, using content and context rather than relying on each employee to file correctly. This is the single biggest lever against the digital 300-box room.
- Search and retrieval. OCR, natural language processing, and metadata enrichment make records findable the way the USPS images were findable — by date, source, and content — which pays off directly in FOIA response, litigation, and audit.
- Legal holds and disposition. Deleting records on schedule is as important as keeping them, but disposition can’t proceed when litigation is pending or anticipated. Automated legal hold management tracks which records are frozen, why, and when the hold lifts — with an audit trail that makes the whole process defensible.
None of this replaces records managers. It moves them from sorting content item by item to governing the policies, schedules, and exceptions that the system enforces — which is exactly where their expertise belongs.
How Armedia Helps Agencies Get There
Armedia has been doing digital transformation, document conversion, and electronic records management work for federal agencies since 2002. Our digitization services are grounded in the imaging standards recognized by AIIM and ISO 16175, and aligned with NARA’s digitization requirements for permanent records — the work agencies still need done to close out their analog backlogs under M-23-07.
On the platform side, we pair Hyland Alfresco — an enterprise content platform with records management built around federal standards such as DoD 5015.02 — with ArkCase, our FedRAMP® Authorized case management platform, to automate retention and disposition policies inside the workflows where records are actually created. And with ArkCase Illume, smart redaction, AI-assisted transcription, predictive analytics, and natural language processing are built directly into the case lifecycle — putting AI to work inside the workflows where records are created, available to agencies today.
Armedia’s records management solutions are available through the GSA Multiple Award Schedule, 8(a) STARS III, and OASIS+ SDVOSB contract vehicles, so procurement doesn’t have to be the hard part.
The lesson from that post office conversation still holds: when records are captured with the right metadata and managed through their full lifecycle, they answer the important questions fast. The difference in 2026 is that no agency can get there manually. If your agency is ready to put AI and automation to work on its records program, schedule a meeting with Armedia and let’s talk about where to start.
Thanks for the information on the postal service record keeping and i am finally at peace that they do not collect our private profile information or data. thank you so much for sharing.