Cross-Dataset Filtering in QuickSight: Filtering Visuals Across Related Datasets with a 1:M Relationship

Hi QuickSight Community,

I’m running into a cross-dataset filtering challenge and wanted to start a dedicated thread since I couldn’t find a complete solution in existing posts.

My Setup:

Let’s say I have two separate datasets that mirror a parent-child table relationship:

  • Dataset 1 (Customer): One row per customer — customer_id, customer attributes, etc.
  • Dataset 2 (Customer Reviews): Multiple rows per customer — customer_id, review_type, review content, ratings, etc.

The relationship between them is 1:Many — one customer can have many reviews. If I were writing this in SQL, it would look like:

SELECT *
FROM customer c
JOIN customer_reviews cr ON c.customer_id = cr.customer_id

What I’m Trying to Do:

I have visuals on a single QuickSight sheet drawing from both datasets. I want to add a filter control for Review Type (a field that only exists in the (many) Customer Reviews dataset) and have that filter apply across all visuals on the sheet — including those sourced from the Customer dataset.

In other words, when a user selects a Review Type, I want:

  1. The Customer Reviews visuals to filter to matching reviews
  2. The Customer visuals to filter to only the customers who have reviews matching that selected type

The Core Challenge:

Review Type does not exist in the Customer dataset. The only shared key between the two datasets is customer_id. So I need the filter on one dataset to effectively pass a list of matching customer_id values to the other dataset — similar to the cross-dataset filtering challenge described in Cindy’s post above.

Has anyone found a working approach for this in QuickSight? Happy to hear solutions involving calculated fields, dataset joins, or any other workaround.

Thanks in advance!

Hi @Rakhipbhatia,

Just to confirm, does your dashboard need to utilize a cross-dataset solution? If not, I would definitely recommend to do a left join of both datasets via customer_id. With this, you can utilize Review Type while having all of the data connected. If not, please let me know and I can see if there are any cross-dataset solutions for your use case.

Thank you!