Customers have addresses. Sites and sensors have coordinates. Orders ship somewhere. Even data with no location of its own usually belongs to a region or territory that has one. Location is a key that can join datasets that were never designed to join, and most analytics treat it as a label instead of a relationship.

Taking that relationship seriously starts with how location is represented: points, lines, and polygons, and the coordinate system that gives them meaning. From there, a small set of spatial relationships does most of the work. What is near what? What falls within what? What overlaps?

This session walks through that progression inside Microsoft Fabric. We'll see how spatial data sits alongside business data in a lakehouse, how those relationships can be expressed with SQL spatial functions and Python with GeoPandas, and how the results flow into reporting like any other table.

Along the way we'll cover the ideas that trip up newcomers, including why a coordinate system choice can quietly break a distance calculation.

By the end, you'll read your existing data differently, and you'll know what questions are now open to you.