I am focused on the maps that travel with us on our phones; they can show us where we are in space; they show us what and, sometimes, who is around us; they show us how to navigate between points in a network of locations that they know about.These maps are much more than their interfaces and front end functionalities, they draw on elaborate stacks of web technology, backend systems and internet infrastructures.

And while maps have always been the visual outputs of attendant spatial, technological, economic, and political systems, these mobile maps are unprecedented in the way they simultaneously deliver, represent and capture spatial data. The scope of their use is also unprecedented – more people, loading more map views, updated with ever greater frequency by a few corporations with the same or similar cartographic techniques, than ever before. These maps are active and animated sites of global information display, assembly, and collection.

Given the volume of information moving through– and being generated by– these mapping systems, a relatively new kind of platform company has emerged to mind the space between maps, their use, and the spatial data they rely upon.

These platforms seemed to have converged around the term “location intelligence” to describe what they produce. Location intelligence companies define their product as “the insight gained from visualizing and analyzing geospatial data. Layering location-specific data—such as demographics, traffic, environment, economics, and weather—on a map or dashboard that reveals unique insights. ” This is a very general description – but these companies have developed ways to structure and synthesize vast amounts of spatial data, integrate them within a web-based geographical information system, and deploy spatial analysis tools and models within a visual mapping environment for display and decision-making.It is often assisted by machine learning and increasingly connected to large language models.

It seems there is more value to be extracted in deploying the information gleaned from map use in a perpetual feedback loop, towards further digital product development and restructuring and investing in physical space. This is the profitable and powerful tail-end of the data captured and synthesized from everyday map use.

Location intelligence companies have adopted the lexicon of human geography, even if their products have a very thin relationship to the full breadth of geographic inquiry. One prominent product, and the center of my work, are Points or Places of Interest data or POI. At a minimum, POI are locations, mappable to coordinates, that map users can generally visit. They are most often retail stores, restaurants, parks or institutions, but can also represent events. Information about these locations can vary between datasets, but they generally reflect location, contact information, and function or place category (store, restaurant, landmark, atm, commercial or public place).

This is a very limited lens through which to view and represent the complexity and many scales of place. And those limitations are cause for concern given the prevalence, influence, and power that location intelligence products that produce POIs have in structuring the physical world and our relationships to it. The discourse on the differences between space, place, and representations of the two in geography is complex – what I am most interested in is recovering the experiential and difficult-to-map aspects of place within our digital systems, and trying to create alternate systems for relating to places, and representing them, at scale.