Reprocess interactions overview
The Reprocess Interactions feature enables Genesys Cloud administrators and analysts to retrospectively apply changes to topics across historic interaction data. By reprocessing interactions on demand, users can ensure that newly created or modified topics are reflected across past conversations. This powerful capability enables organizations to unlock deeper, more accurate insights from their historical STA data. This feature is designed to help you maximize the value of their STA investment by unlocking historical understanding. For large enterprises and multinational organizations, this is especially critical in tracking long-term customer experience trends and supporting strategic business decisions.
With Reprocess Interactions you can:
- Complete historical visibility – Retroactively recognize issues, trends, or shifts in customer experiences using newly configured STA topic models.
- Flexible on‑demand updates – Launch reprocessing jobs at any time. There is no need to wait for scheduled analytics runs.
- Targeted scope – Define the set of interactions by filtering media types, date range, and program associations.
- Operational clarity – Monitor job creation, progress status, affected interaction counts, and downloadable identifiers.
- Reprocessing Interactions only applies to topics associated with selected programs.
- When administering updates to topic logic, admins reprocess prior data by selecting those programs.
- Consistent terminology (publish) is used to reflect active job submission.
- One reprocessing interaction job can only process a maximum of 250,000 interactions.
- If any program within an interaction matches a reprocessing job filter, the entire interaction will be reprocessed.
- The topic reprocessing feature cannot delete all topics in a transcript block unless at least one new topic is detected in that block.
- If an interaction contains no detected topics, it will be marked as failed during reprocessing.
- If none of the transcripts in an interaction contain detected topics—even if one transcript block does—the interaction will be considered a failed interaction.
- If the selected program for reprocessing has no topics at the time of reprocessing, the interaction’s existing topics will remain unchanged.
- The reprocessing CSV will show only newly detected topics; it does not display topics that were removed (for example, no before-and-after topic comparison).
Usage example
- Administrator creates a new Topic to capture a known service issue.
- Using Interaction Reprocessing, the administrator selects relevant programs, defines filters (for example, past 6 months), and publishes the job.
- STA retroactively labels historic interactions based on the new Topic.
- Analysts run fresh dashboards and reporting using the refined dataset.
Best practices
| Practice | Description |
|---|---|
| Program Awareness | Reprocess only the relevant programs tied to updated topics. |
| Scope Management | Use date/media filters to limit processing volume and reduce unnecessary costs. |
| Export IDs | Keep the export of Interaction IDs for auditing or re-ingestion support. |
[NEXT] Was this article helpful?
Get user feedback about articles.