Forum Discussion
Highlighting in an Azure Map Visual
Hi mikemagill，
Fully understand that.😂
As another alternative solution, you could consider using a Heatmap Instead of Points.
If the goal is to show the general area rather than individual postcodes, switching to a heatmap layer in Azure Maps can help preserve visibility while emphasizing high-value areas.
Or, custom Map Visual (via Power BI Custom Visuals or Python/R)
If you need full control, consider creating a custom map visual using Power BI's Python/R scripting or a custom visual like Mapbox, which may support dynamic radius filtering.Either of above is a workaround, you could consider it based on your customer's need.
In short, precumputing by SQL database is a way fully matched with your need, however, it is huge workload.
Anyway, good luck.😀
Best Regards,
Qi
Community Support TeamIf there is any post helps, then please consider Accept it as the solution to help the other members find it more quickly.
If I misunderstand your needs or you still have problems on it, please feel free to let us know. Thanks a lot!
Hi mikemagill,
You are right that Azure Maps in Power BI only provides "Filter" or "None" under Edit Interactions, so you can't use a "Highlight" effect like you would on a bar chart.
Please check the doc for details:
If you want this feature to be supported, you could raise an idea here:
https://ideas.fabric.microsoft.com/
Since you have mentioned that 90,000 postcodes and working out the distance between each will kill the report performance, I think the only way is to claclualte the distance, however, you could consider precomputing distances between postcodes and storing them in a table (SQL or Power BI), which is not that easy😂
- Precompute distances between postcodes in SQL Server.
- Store only nearby postcodes (e.g., within 5-10 km) for each postcode.
- Create a disconnected table of postcodes in Power BI and use a measure to check if a postcode falls within the selected one's precomputed nearby list.
Here are some related blogs for your reference:
Best Regards,
Qi
Community Support Team
If there is any post helps, then please consider Accept it as the solution to help the other members find it more quickly.
If I misunderstand your needs or you still have problems on it, please feel free to let us know. Thanks a lot!
Many thanks. I've submitted an idea.
I might explore the SQL database solution but I suspect my client might baulk at an 8 billion row table even if it is only three columns. 😉
Hi mikemagill，
Fully understand that.😂
As another alternative solution, you could consider using a Heatmap Instead of Points.
If the goal is to show the general area rather than individual postcodes, switching to a heatmap layer in Azure Maps can help preserve visibility while emphasizing high-value areas.
Or, custom Map Visual (via Power BI Custom Visuals or Python/R)
If you need full control, consider creating a custom map visual using Power BI's Python/R scripting or a custom visual like Mapbox, which may support dynamic radius filtering.Either of above is a workaround, you could consider it based on your customer's need.
In short, precumputing by SQL database is a way fully matched with your need, however, it is huge workload.
Anyway, good luck.😀
Best Regards,
Qi
Community Support TeamIf there is any post helps, then please consider Accept it as the solution to help the other members find it more quickly.
If I misunderstand your needs or you still have problems on it, please feel free to let us know. Thanks a lot!