Icon Map Catalog Reference Data

Every layer in the Icon Map catalog publishes a reference file alongside its map tiles: one row per feature, carrying all of that feature's attributes. This source lets you use that data directly - in charts, data tables, slicers and panels - without importing or copying anything.

You browse the catalog exactly as you would to add a layer. Instead of drawing the features, you get their data.

Adding it

  1. In the Data pane, choose Add and pick Icon Map Catalog Reference Data.
  2. Browse or search the catalog and select a layer. The preview shows the columns and some sample rows, so you can see what you are getting before you commit.
  3. If the layer publishes data for several variants (some catalog layers are a group, with a separate dataset per area or theme), choose which one you want under Data variant.
  4. Select Use data.

Adding it at the same time as the layer

If you are already adding the catalog layer, you don't need to come back and find it again. In the layer's Configure step, open the Reference data section and switch on Also add as a data source. One Add gives you both the layer and its data.

The reverse works too: when picking reference data you can switch on Also add as a layer to drop the map layer in at the same time.

What you get

The columns are the layer's own attributes, plus two extras Icon Map derives for you:

  • _center_lon / _center_lat - the centre point of each feature. These let the data drive point layers (icons, circles, heatmaps) straight away.
  • _min_lon, _min_lat, _max_lon, _max_lat - each feature's bounding box. Listed but switched off by default; turn them on if you want to filter or measure by extent.

There is no geometry column. Shapes and lines come from the catalog layer, not from this source - which is why the two work so well together.

The pairing that makes this powerful

Add the catalog layer and its reference data, then bind them on a shared key field. Your charts, tables and slicers now cross-filter the map: click a bar, and the matching features highlight.

Selecting columns matters

The reference file is read across the internet, and only the columns you select are downloaded. On a large catalog layer that is the difference between a couple of megabytes and a couple of hundred.

If you select too many columns from a very large layer, Icon Map will tell you rather than hanging the browser - deselect the ones you don't need. Some of the biggest layers (national property or address datasets, with tens of millions of features) are practical for charts and tables but not for plotting every feature as a point.

Published and embedded maps

Public catalog reference data stays live in a published map: the viewer reads it straight from the Icon Map catalog, so the map reflects the current published data rather than a frozen copy. If it can't be reached, the map falls back to the data captured when you published, so it never goes blank.

Reference data from an organizational catalog is captured at publish time instead, because those files live in your own OneLake and aren't publicly readable.

Limits worth knowing

  • Commercial-tier catalog layers are not offered as a data source. Those layers can still be added to the map as normal - their licensing and usage are handled through the tile service.
  • Only in Fabric. The Power BI visuals don't offer this source.
  • Data is read as published. To edit or extend it, use Add to Lakehouse from the layer's Reference data section and work with your own copy.

Next steps