As the number of profiles directly influence the cost of Customer Insight, it is essential that the data ingested from dataverse, which often is the core of the data, can be filtered properly. For instance many CRM-systems for B2C might have reseller contacts, supplier contacts, or just contacts that do not want marketing, that are of no interest in customer segmentation. Hence there should be an ability to filter out this data.

There are workarounds by using the Dataflows which directly use Dataverse, but this is not suitable for large amounts of data.

Using datalake data that is synchronized via Azure Synapse Link and read via Power Platform Data flows is probably the best workaround, but it is a bit complex.