View Samples
View Samples

A Comprehensive Guide to Building Footprint Data

What makes a good polygon and how can you tell the difference? In this guide, we’ll dive into what building footprints are, how they are developed and used, and what to look for when sourcing data.

Building footprints in Lusaka
Lusaka, Zambia

What is a building footprint?

Building footprints are polygons representing the extent of individual building structures in the physical world. They are often accompanied by contextual metadata about what the building is, how it relates to nearby places, and what its address is.

A sample of building footprint data extracted by Ecopia AI in Barcelona, Spain
A sample of building footprint data extracted by Ecopia AI in Barcelona, Spain

But wait, what is polygon data?‍

Polygon data in GIS

Geospatial data is typically compiled and stored as one of two main formats: vector or raster. Vector data can be divided further into three types: points, lines, and polygons. All three of these data types can vary in size depending on the geographic extent they represent, and can be applied in a wide variety of use cases. For example, polygons can range anywhere from entire continents to specific units within an office complex depending on the needs of the end user. The general rule of thumb is: if the feature is a collection of lines connected by vertices, resulting in a closed shape on Earth’s surface, it’s a polygon.

Samples of point, line, and polygon vector data extracted by Ecopia AI
Samples of point, line, and polygon vector data extracted by Ecopia AI

Building polygon data

While there are endless entities in the world that can be represented by a polygon, building footprints are commonly used for mapping and analytics workflows that involve specific places. This level of granularity is necessary when understanding how buildings relate to each other, the physical environment, and the accessibility of infrastructure like roads and sidewalks. 

Building footprint data is most valuable when it forms a solid foundation for contextualizing and analyzing a location. A singular building polygon that denotes where on Earth a structure exists can be mapped, but if it is inaccurate or out-of-date, the derived context, analytics, or visualizations are not useful. In fact, it could lead to misinformed decision-making and misallocation of resources. 

How are building footprints created?

There are multiple ways to acquire building footprint data, ranging from creating it yourself, scraping open source databases, and purchasing it from a provider. Each provider has their own methodology for digitizing and curating data for their end users. In this section, we’ll go over the basics and then explain Ecopia’s market-leading methodology.‍

How to create building polygons

As we mentioned earlier, geospatial data can be formatted as either vector or raster. While the final output of creating building footprint data is ultimately vector data (polygons), raster data is needed to extract and digitize real-world features.

Raster data essentially divides the Earth’s surface into a grid and stores geographic information within each grid cell (also commonly referred to as pixels). This is how satellite and aerial imagery data is prepared and stored. There are many ways raster data can be used as-is, but many GIS users ultimately digitize features from imagery data so that they can use them as vector polygons.

Satellite imagery in Lusaka, Zambia
Building footprints in Lusaka, Zambia

Building footprint vector data digitized from imagery raster data in Lusaka, Zambia

To enable this capability, GIS programs have developed sophisticated tools for digitizing imagery into vector features. However, manually digitizing features remains a time-consuming and tedious process that is almost impossible to scale - not to mention expensive. ‍

Ecopia’s building footprint methodology

Our team recognized the challenges organizations encounter when faced with digitizing buildings at scale, so we developed an artificial intelligence (AI)-based mapping system for detecting, extracting, and updating geospatial features from imagery data. Using this methodology, we have been able to create high-precision maps of geographic areas ranging in size from entire continents down to granular cityscapes, in both 2D and 3D.

Developing building polygon data from imagery is complex, but we have been able to help organizations rapidly scale their analysis by taking on the heavy lifting of sourcing and maintaining an accurate database. Our collaborative partner network allows us to digitize the best imagery data, ensuring our AI is constantly updating to reflect dynamic real-world change.‍

Top use cases for building footprints

So what can building polygons be used for? Each day there is a new and innovative use case for building footprint data, but at Ecopia we tend to see them used to solve the following spatial challenges.

Building polygons for insurance risk assessment

When pricing a policy for property insurance, underwriters must calculate the risk of a specific property and structure. Oftentimes, this is done by measuring a building’s proximity to a flood zone, or identifying whether or not a property is adjacent to a high-risk business (such as an industrial plant containing flammable chemicals). But regardless of what underwriters are measuring the risk of, what does not change is the importance of correctly locating the property’s exact extent so as not to under- or over-price a policy.‍

This need for precision is what makes building footprints an integral part of insurance underwriting. While singular latitude and longitude point coordinates can represent the approximate location of a property, they do not provide the level of detail needed to understand how the building itself relates to nearby risks.

Building polygon data (including building-based geocodes) vs parcel centroid data for risk assessment
Building polygon data (including building-based geocodes) vs parcel centroid data for risk assessment

‍For example, an insurance policy priced using only point data derived from a parcel centroid may miscalculate how close the building lies to an eroding beachfront, flood zone, or fire risk area. Without seeing the entire extent of the building footprint and factoring that into their calculations, an underwriter may under-price the policy and open their business up to be liable for millions of dollars in future property damages.‍

Learn more about building data for insurance

Building footprints for property tax assessment

Properties evolve over time - and so does the associated property tax. To keep up, local tax assessment offices have to maintain an up-to-date and accurate record of every property in their jurisdiction. That’s no small feat; tax assessments change with even the slightest modification to a property, which is why assessors are increasingly relying on geospatial building data to support mass appraisal workflows. 

Precise property information enables an assessment office to analyze the location of a property and determine its distance from the nearest house, road, coast, park, or school, providing them with important inputs to their methodology. Detailed building footprint data can also show how large a building is, and help identify where new taxable activity is taking place.

Building footprints in Sacramento, CA, highlighting detected changes by Ecopia AI
Building footprints in Sacramento, CA, highlighting detected changes by Ecopia AI

Detecting change in building and property footprints is critical for tax assessment offices to scale their operations without sacrificing quality. While the industry used to rely on field visits to assess property tax change, the availability of digitized building data enables assessors to more efficiently locate updated properties, reducing the time spent traveling to inspections and ensuring no taxable updates go unnoticed.

Learn more about property change detection

Building footprint polygon data for stormwater management

Understanding climate change and developing resiliency plans requires an accurate representation of how the physical environment relates to man made structures. To achieve this, stormwater management teams rely on building data. Effectively modeling how stormwater will interact with properties is crucial for building infrastructure that can withstand future storm events, as well as for planning emergency response efforts.

As part of these climate resilience strategies, municipalities must comply with federal regulations regarding stormwater infrastructure. To fund infrastructure projects, municipal governments have implemented stormwater utility fees (SUFs), which are essentially taxes based on a property’s impact on water run-off. 

Buildings and other impervious surfaces extracted for the City of Detroit by Ecopia AI
Buildings and other impervious surfaces extracted for the City of Detroit by Ecopia AI

To calculate that impact, municipalities are leveraging building footprint and other impervious surface data. While all buildings are impervious surfaces, there are varying degrees of imperviousness that help determine the SUFs for that property. Seeing how these buildings relate to other impervious surfaces, like driveways or roads, is also critical for effective stormwater management.‍

See how the City of Detroit leverages property data for stormwater management

Building footprint data for network planning

Telecommunications companies are similarly reliant on building polygon data as they plan their network expansion strategy. Network planners require a fabric of different data types to understand current network infrastructure and develop roll out plans. By combining data such as building footprints, building-based geocodes, parcel boundaries, and land use details, telcos can analyze specific structures on a parcel, understand which structures are their own addressable locations (potentially indicating the need for service), and determine the service needs of other buildings located within that parcel.‍

Network planning requires this unique mix of data to identify broadband-serviceable locations (BSLs) and achieve compliance with telecom governing bodies. For instance, not all buildings on a rural parcel are broadband-serviceable; while the main house or office might be, the nearby barn might not be. Similarly, within one parcel there may be one address, but multiple BSLs. This is common in university settings or commercial office parks. Without a comprehensive building footprint dataset, telecoms lack necessary context for parcels, geocodes, and land use data. 

Ecopia AI’s building footprint and building-based geocodes in relation to parcel boundaries for telecommunication network planning
Ecopia AI’s building footprint and building-based geocodes in relation to parcel boundaries for telecommunication network planning

With building polygon data, network planners can best assess which areas are underserved by broadband and how the current level of infrastructure impacts the roll out of expanded coverage. Insight into how buildings are changing is also critical, as new developments pop up and existing infrastructure is improved to facilitate network expansion.

Imagine if a network planner was creating a plan for a broadband internet roll out in a rural community using out-of-date, inaccurate, or incomplete data to determine BSLs. While some structures will remain unchanged, others may have been built in the meantime, or are in the process of being constructed. With accurate, complete building footprints derived from up-to-date imagery, the network planner could factor these new properties and infrastructure into the roll out plan and appropriately determine the scope of work. But without them, the planner runs the risk of leaving out new construction, or rolling out coverage to buildings that no longer exist.

Learn more about building footprints for network planning

Where to get building footprint data

There are a variety of building footprint providers out there, each with its own pros and cons. But as you explore possible sources, there are few things to keep in mind.

What to look for in building polygons

Not all polygons are created equal. Because each building footprint provider is unique in its methodology, they can each vary in quality. We’ve learned a thing or two about what makes a high quality building polygon, so here is a list of what to look out for as you evaluate providers.

  • Coverage: When sourcing building data, make sure you find a provider who has the geographic coverage you need. Some providers only offer polygons for certain geographies (ie. cities but not rural areas), or only certain types of buildings (ie. commercial but not residential locations), so make sure you go with one that has all of the buildings you need to analyze.

  • Accuracy: There are data providers out there who claim to have building polygons, but in reality scrape open source data that is at best an approximation of where the building is. Granularity and accurate representation of features on the Earth’s surface are what make building footprints so useful, so be cognizant of these differences and keep an eye out for true building data.

  • Freshness: The real world is constantly evolving, and quality data reflects this dynamic change. There are two main components of freshness to consider when choosing a building dataset. First, how fresh is the imagery that the buildings are generated from? And second, how fresh is the methodology for digitizing those footprints? Different use cases have varying definitions of freshness in this regard, so when sourcing building footprint data, determine what data freshness means to you and what frequency of updates you need to do your job successfully. 
Ecopia AI’s building footprint and landcover extraction coverage (left) vs another data provider (right)
Ecopia AI’s building footprint and landcover extraction coverage (left) vs another data provider (right)

Get started with Ecopia’s building data

Ready to start mapping with high quality building footprints? At Ecopia, we’ve built the most comprehensive, accurate, and up-to-date database of building polygons in over 100 countries around the world. We work closely with our partner ecosystem to ensure our building data is updated frequently using the best imagery so that you can be confident in the output of your mapping and analytics. Get in touch with a building footprint expert today to get started.

Learn more about Ecopia's building footprints

Ready to get started?

If you're ready to leverage groundbreaking advancements in artificial intelligence, let's chat.

Let's talk