> For the complete documentation index, see [llms.txt](https://doc.duaer.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://doc.duaer.com/advanced-ai/evaluations/tips-and-common-issues.md).

# Fix common evaluation issues in Duaer

In Duaer, when triggers, field maps, or chat output fight the evaluation nodes, check the wiring first.
## Combining multiple triggers <a href="#combining-multiple-triggers" id="combining-multiple-triggers"></a>

1. Run the production or manual trigger once and copy its output JSON shape.
2. After Evaluation Trigger, use a Set node so each test row matches that shape.
3. Merge both branches with No Operation. Later nodes read only that shared output.

While an evaluation runs, Evaluation Trigger should pull the test rows. If another trigger fires at the same time, runs look sometimes like evaluation and sometimes like production. The wiring tip on Evaluation Trigger links to this section.

## The row arrived, but scores are empty

Confirm the test-row field names match what later nodes read. Confirm the metrics were really calculated in [calculate metrics](/advanced-ai/evaluations/metric-based-evaluations.md#2-calculate-metrics). Each digital organization may have only one Evaluation Trigger. To test different parts, split them into [sub-organizations](/flow-logic/subworkflows.md).
## Questions

### Can a Duaer evaluation digital organization have more than one trigger?

Yes, but Evaluation Trigger should pull the test rows while an evaluation runs. Other triggers should not fire at the same time. Merge both branches into one shape before later nodes.

### How many evaluations can one Duaer digital organization have?

One. Each digital organization may have only one Evaluation Trigger. To test different parts, put them in sub-organizations and evaluate each one.