Outbound campaign recording metadata missing in bulk export job

General Discussion
1

Stumbled on a weird bug today with the outbound dialing recording exports. we have a specific requirement for legal discovery where every outbound call needs its full metadata chain preserved. normally the recording api returns the contact_id and campaign_id in the headers. but when we trigger a bulk export job via /api/v2/recordings/bulk-export for calls made by our predictive outbound campaigns, the resulting s3 manifest files are missing the campaign context. the json payload only shows the recording id and file location. no campaign_id. no contact list reference. this breaks our chain of custody process because we cannot map the audio file back to the specific outbound interaction without manual db lookups which is not scalable. we are using the standard export configuration with s3 bucket integration. permissions are set correctly as we can export internal chat transcripts without issue. the error is not a 403 or 500. the job completes successfully. it is just the data quality that is wrong. we tried adding custom metadata fields in the contact list but those do not appear in the export manifest either. is this a known limitation with outbound recordings specifically? or is there a specific scope we need to add to the service account used for the export job? the documentation mentions that digital channel recordings include more metadata but outbound voice seems to be treated differently. we are on the latest platform version. timezone is europe/london. any help would be appreciated. we need to resolve this before our next audit cycle. it feels like the outbound engine strips the campaign association before sending it to the recording service.

2

Have you tried adjusting the export configuration to explicitly request the contact and campaign data groups? The default bulk export often omits these nested objects to reduce payload size.

Here is the typical HCL structure for the export job definition:

resource "genesys_cloud_recording_export" "outbound_metadata" {
 name = "Outbound Legal Discovery Export"
 format = "JSON"
 
 filters {
 type = "OUTBOUND"
 start_date = "2023-10-01T00:00:00Z"
 end_date = "2023-10-31T23:59:59Z"
 }

 data_groups = ["recording", "contact", "campaign"]
}

The data_groups attribute is critical. Without campaign, the manifest will lack the context needed for legal discovery. Also verify that the service account running the export has recording:view and outbound:campaign:view permissions. Missing permissions can silently drop fields in the output.

Note: Large exports with multiple data groups may take significantly longer to process. Monitor the job status via CLI before downloading.

3

This looks like a known limitation with the default export filters. Ensure the include_campaign_details flag is explicitly set to true in the HCL resource. The manifest will fail to parse if the campaign ID is null, so validate the outbound campaign configuration first. The ServiceNow integration usually handles this via webhook, but the bulk API requires explicit inclusion.

4

What’s happening here is that the bulk export job likely omits campaign_id due to API rate limits during high-concurrency ramps. Check your JMeter throughput; if you exceed the WebSocket connection limits, the metadata payload gets truncated. Try adding a Constant Throughput Timer to stay under the 429 threshold.

5

This issue stems from the bulk export defaulting to minimal payload schemas, similar to how Zendesk ticket exports omit custom fields unless explicitly requested.

{
 "include_metadata": {
 "campaign": true,
 "contact": true
 }
}

Adding this ensures the legal discovery chain remains intact during the migration.