Unable to relate two dimension models to two fact models in dataset

Hi Shahid,

If I understand correctly, your dataset relationship looks like this.

This would not work out of the box because of the ambiguous paths between the 2 fact models. We would usually suggest breaking down the dimension models tailored for each fact model and removing the ambiguity. However, in your scenario, it is not feasible as you need to visualize metrics from both fact models with essentially a many-to-many relationship between the fact models.

In this case, can you help check if a set-up like this is feasible for your use case? We propose a potential solution by creating a separate dimension model that will contain all the common and relevant dimensions shared by both fact models. With this, you can create the necessary relationships between the dimension model and the two fact models to allow you to visualize desired metrics.

This solution is not ideal, but we are continuously working on ways to address this use case better. We recommend keeping an eye on the upcoming revamp to our metric language ([Upcoming] Our brand new Analytics Query Language, AQL!), which may provide a better solution.

Regards,

1 Like