Each component selects its own Analysis model. A report can therefore present results from several models or data sources on the same page.

1. Add the first component and bind its model and fields.
2. Add the second component and choose its model independently.
3. Give measures clear names and units. Align the comparison period and business definitions.
4. If a filter or chart selection should affect both components, configure Actions → Interactions → Linked components on the source.
5. Test a known value in each target model and verify the result separately.

A click on a chart filters a component on another model only by the clicked fields that also exist in that model. If none of them exist there, the component is not re-queried, and its Conditions list shows no link.

Displaying two models together does not join their underlying rows. Cross-model interaction also needs compatible field semantics and values. Matching captions alone is not evidence that two fields mean the same thing; check geographic levels, date granularity, codes, and units.

If a target does not respond as expected, first verify the link selection and each component’s own filters. Then inspect the corresponding fields in the models. See linked components.