The practical answer

Preserve the exact response, identify where the failure occurred and which submissions it affects, then assign the issue to the owner who can fix that layer. Verify the prior filing state before choosing a resend, replacement, or correction process.

“The file failed” is not enough information to repair an AIR submission. An access problem, a stale manifest, an XML schema problem, and an incorrect reporting value require different evidence and different owners. Triage narrows the problem before anyone changes a production record.

This guide follows Publication 5165, revision December 2025, and Publication 5258, revision December 2025. The worklist examples use plain-language issue descriptions, not invented IRS error codes.

Preserve the failure evidence before editing

Save the response, acknowledgment, and error data supplied by the channel, along with the exact transmitted files, environment, send time, packet ID, and available receipt or transmission references. Keep the source packet read-only while investigating so the team can reproduce what failed.

Record whether the observed failure was a local application error, an IRS portal response, or an AIR acknowledgment. A screenshot of a vendor message may omit the actual agency result. Obtain the underlying evidence through the responsible transmitter.

If no final outcome is known, investigate receipt and status retrieval first. An interrupted browser session is not the same event as a confirmed rejection, and its next action should not be chosen as if it were.

Classify the failure layer and owner

Suggested triage categories and evidence
LayerEvidence to inspectLikely investigation owner
Access or initial transportChannel response, environment, authorization contextTransmission administrator
Manifest and file pairingFinal file metadata, references, year selectionsPacket-generation owner
XML structureSchema result and supplied location detailSoftware or integration owner
Submission-level rulesAcknowledgment scope and relevant business rulePreparer with technical support
Underlying employer factsReported value and authoritative sourceEmployer data or tax reviewer

Publication 5165 distinguishes portal faults, manifest and schema validation, and business-rule processing. Treat a prefix as a clue while reading the full message and its scope.

Identify the affected transmission or submission

Determine whether the entire transmission failed or a particular submission was rejected within a partially accepted packet. Link the supplied identifiers and error locations to the internal submission map. Keep unaffected accepted work visible so it does not get swept into a new original batch.

The publication describes XPath information in error detail that can help locate a data element and instance. Use it with the exact sent file. A path applied to a regenerated file may point to a different record if ordering changed.

Count affected submissions and records separately from error messages. Several messages can concern one underlying issue, while one schema problem can prevent processing of a much larger population. The number of messages is not automatically the number of employee returns needing action.

Fictional example: three findings, two repair paths

Fictional Juniper Meridian Company receives a confirmed rejection for a packet with two employer submissions containing 90 and 60 records. Its plain-language triage notes identify a stale metadata value, an invalid date representation in one source mapping, and a second date message caused by the same mapping.

Fictional investigation worklist
IssueOwnerRepair evidence
Stale packet metadataPacket-generation ownerRegenerated metadata for the final file
First date-format findingIntegration ownerCorrected date mapping and representative output
Second date-format findingSame integration ownerReview of the shared mapping's affected population

The original packet contains 150 records: 90 + 60. Three messages do not imply only three failed records. The team establishes the actual rejection scope, repairs two root causes, and uses the confirmed filing history to route the next attempt.

Retest the actual repair and its wider effect

Regenerate the packet from the corrected source or mapping and compare the new output with the sent version. Explain intended differences and investigate unrelated changes. If the employer facts changed, obtain the required renewed substantive review.

Run the applicable schema and business-rule checks against the new exact files using the confirmed versions. Publication 5258 supplies composition guidance, while the current IRS resources identify the relevant specifications. Passing a local check does not predict every later agency validation.

When a shared mapping changes, inspect its full affected population, not just the record named in the first error. Preserve the test result and the new packet version so the operator knows which repair was actually verified.

Route the next attempt from the established history

Have the transmitter apply the relevant rejected-file, replacement, or correction procedure using the prior operation type and exact outcome. Publication 5165, section 7, distinguishes those paths, including rejected correction transmissions. Do not mark every regenerated file as a new original.

Record the chosen process, scope, required original references, actual approval, and release owner. Then retrieve the new acknowledgment and confirm whether the identified problem is resolved. A local retest closes the local repair task, not the agency outcome task.

Use the downloadable worklist to connect each issue to evidence, an owner, a verified repair, and a later outcome. The record should explain why the next attempt was permitted and which previously accepted work was outside its scope.

Triage a rejected packet by failure layer

Triage a rejected packet by failure layer: Preserve the response; Locate the layer and scope; Repair and retest; Route and monitor
The fictional worklist uses descriptive labels, not IRS error codes. The actual response determines scope and routing.
Read the workflow as text
  1. Preserve the response. Keep exact files, environment, references, and agency detail.
  2. Locate the layer and scope. Separate transport, packaging, schema, and data issues.
  3. Repair and retest. Verify the actual change and its affected population.
  4. Route and monitor. Apply the documented next process and retrieve the new outcome.

Put this guide to work

Rejected 1094-C packet triage and retest worklist

Save the editable text worksheet and use it with your own records. Keep completed copies in your secure working files.

Download the worksheet TXT

Common questions

Should we retry immediately after any error?

First identify the actual failure and prior outcome. An unknown send result, a portal fault, and an AIR rejection can require different actions.

Does the error count equal the number of failed returns?

No. Several messages can describe one issue, and a single structural failure can affect a whole submission or transmission.

Can an error location be checked against the newest export?

Use the exact sent file first. Regeneration can change ordering or content, making a location refer to something different.

Does a passing local retest close the issue?

It verifies the local repair tested. Retrieve and review the agency outcome from the authorized next attempt before closing transmission follow-up.

Who decides replacement versus correction?

The responsible transmitter applies the current guidance to the original operation type, scope, and actual outcome. The triage packet should provide those facts.

Official sources and scope

Sources checked September 5, 2026. Use the edition for the tax year and filing method you are working with; later instructions may change thresholds, fields, or procedures.

  1. IRS Publication 5165, revision December 2025

    Failure layers, acknowledgment detail, scope, and correction/replacement distinctions.

  2. IRS Publication 5258, revision December 2025

    Technical composition and validation reference.