Geospatial indicates data with a geographic meaning to it. Data is therefore tied to a location. Geographic locations differ in their level of detail (accuracy of locations). Examples for the level of details are i.a. continent, country, city, street, address, coordinates, but are not limited to those.

I have been working on a B2B geospatial tool for over two years in risk management. Such a web-application aims to provide domain experts a tool that enables them to plan with potential risks. Together with the developers' team, we started on a greenfield that allowed us to build a comprehensive experience. In collaboration with experts, I lead the entire user experience for this product. That included developing use cases and stories, designing the interface, defining a scalable visual language, shipping, refining, and maintaining features. As this project is under a non-disclosure agreement, I cannot go further into details.

Following, I share a few abstracted thoughts.

A tool like the geospatial app I have been involved in allows people to upload their own datasets. That also means that you will need to find a good balance between defining constraints and leaving enough room for exploration.

As an example, a simplified and a rather generic scenario should help understand some basics of a geospatial tool. Imagine the following:

  1. You have a dataset that you would like to understand better. → Visualizing Data
  2. You then would like to understand the potential exposure to or with your data. → Combined visualization of your data with external data (layers)
  3. You would like to focus on a specific area to understand the necessary (details) → Filtering of data and therefore the visualization itself

Based on the gained knowledge, actions are taken (within the tool and/or outside).

Visualizing Data

Visualized data is an abstraction of datasets to make these accessible and understandable to people. As in any data visualization, it is crucial to understand the different information a dataset consists of. They affect the way your visualization works and how people can interact with it.

Data accuracy is crucial, no doubt. However, when working with complex data sets, it is also important to understand which level of detail people should receive information to actually process these depending on their background and goal(s). So, if people are not there to do scientific analysis with data, simplification is valued higher than accuracy — at least initially.

The complexity increases drastically through more variables. It is important to understand the most valuable information you would like to visualize and how it makes the most sense.

Example visualization

Example visualization

In the above example, two different zoom levels are represented. The Data consists of only two variables: location (e.g., coordinates) and value. In the higher zoom-level, locations are visualized as single pins, making it easy and accurate to locate. However, other locations are quickly lost (out of viewport). Therefore, a lower zoom-level helps to get the bigger picture to understand the distribution of locations. These locations here often need to be clustered due to accessibility and performance (large datasets) reasons. Try out / explore: Different visualization modes bring different perspectives. Examples are grids, regions, heatmaps, etc.