Power BI Embedded capacity optimisation for concurrent users

The issue you're encountering with Power BI Embedded where reports stop loading and show a spinning bar after multiple users interact with slicers could be due to performance bottlenecks, especially when handling concurrent user requests. Here are a few best practices to improve performance for embedded reports:

1. Optimize your Power BI model: Ensure the data model is optimized. Large or complex models may slow down when multiple users interact with them.

2. Consider direct query mode: Instead of using import mode, you could use DirectQuery to reduce the load on the Power BI report server by querying the data source in real time.

3. Limit visuals and slicers: Too many visuals or slicers can strain the report's performance. Reducing the number or optimizing visuals may help.

4. Use Azure capacity for embedded reports: Make sure you are using an appropriate Azure Embedded Capacity SKU, as this can significantly improve performance under load.

For detailed guidance, refer to the article here: Power BI Best Practices