Data quality and error handling
ZTZenoti Docs Team Updated November 9, 2026 2 min readGuide
Monitor data quality and resolve common issues when working with Analytics β Amazon S3 data delivery.
Before you begin
- Your organization must have an active Analytics β Amazon S3 subscription.
- You need access to your S3 bucket to verify data delivery.
Steps
Monitor data freshness
Track when your data was last refreshed:
- Check your S3 bucket for the most recent GroupID folder.
- Verify that new GroupID folders appear every 1β2 hours during normal operation.
- Query ETL metadata columns in your warehouse tables to confirm data freshness:
`sql SELECT MAX(etlcreateddate) AS last_loaded FROM Bi_FactInvoiceItem WHERE void = 0;
- If no new GroupID has appeared beyond the typical 1β2 hour cadence and no
_processingfolder is present, contact Zenoti Support.
Check for processing errors
- Look for GroupID folders with a
_processingsuffix in your S3 bucket. - A
_processingfolder that persists for several hours may indicate a failed data load. - Do not query data from
_processingfolders β the data is incomplete. - The next scheduled refresh cycle typically processes any data missed by a failed load.
Verify data completeness
- Check for missing table folders β Incremental GroupID loads contain only tables that had changed data. If a table is missing from an incremental GroupID, it means no data changed for that table in that cycle.
- Verify GroupID 1 completeness β The initial full load (GroupID 1) should contain all tables. If a table is absent from GroupID 1, it may not be included in your subscription scope.
- Check row counts β Compare row counts in your warehouse with expected transaction volumes.
Handle the void column correctly
The void column in fact tables controls data consistency:
void = 0β Current, active record. Include in queries.void = 1β Soft-deleted record. Remove from your downstream data.
Failing to filter on the void column leads to duplicate records and inflated aggregations. Always include WHERE void = 0 in fact table queries.
Request a table-level bypass
When specific tables consistently fail and block the pipeline:
- Contact Zenoti Support with the table names, center ID, and organization ID affected.
- Provide error messages and business impact details.
- Once the bypass is configured, other tables continue processing normally.
- Work with support to resolve the underlying issue and remove the bypass.
Handle data freshness discrepancies
If your reports show outdated information:
- Verify the most recent GroupID in your S3 bucket.
- Check whether your pipeline has processed all available GroupIDs.
- Query
etlcreateddateorEtlLastUpdatedOncolumns to find the most recent load timestamp. - Ensure your incremental load logic correctly applies the delete-insert pattern for fact tables.
Tips
- Set up automated monitoring to alert you when expected data files do not arrive within the normal 1β2 hour window.
- Keep a log of recurring error patterns to help identify issues that need investigation.
- Use the ETL metadata columns (
etlcreateddate,EtlLastUpdatedOn) in your queries to filter for the freshest data.
Was this article helpful?
Your feedback helps us improve our documentation.