US Geospatial Industry Shifts From Detection to Execution
The United States is the largest geospatial market in the world. Fortune Business Insights estimates the US market will grow from $141.59 billion in 2025 to $161.23 billion in 2026. About a third of the $575.43 billion global market comes from North America. Geospatial data means all data tied to a location, from maps and satellite images to three-dimensional terrain.
The business of this industry can be summed up in one sentence. Satellites and drones photograph the ground, software analyzes the images, and service companies turn the results into answers that can be used in the field. The question the US industry is trying to answer has changed. It is no longer whether change can be detected in an image, but whether the result can be used on site tomorrow.
The Weight of the US Market
Government demand is the center of gravity in the US market. Defense and intelligence account for 70 to 80 percent of Earth observation revenue, according to an industry assessment from January 2026. Earth observation means collecting images of the planet from satellites. The global geospatial defense applications market is estimated at $15.09 billion in 2026, with North America taking 45 percent of it.
The National Geospatial-Intelligence Agency, known as the NGA, is the Defense Department agency that supplies satellite imagery and mapping information to the military and intelligence agencies. In August 2025 the NGA awarded Lockheed Martin a seven-year, $652 million contract for imagery analysis. The National Reconnaissance Office, which operates US reconnaissance satellites, has ten-year contracts to buy imagery from commercial optical satellite companies. An optical satellite photographs the ground with light, much like a camera.
The commercial market is also growing fast. The GeoAI market, which applies artificial intelligence to geospatial data, is projected to grow from $37.18 billion in 2025 to $85.51 billion in 2033, according to Data Bridge Market Research. GeoAI means technology that lets artificial intelligence read and analyze satellite and aerial images. North America held 40 percent of the GeoAI market in 2025. The drone services market, which rents out drones for mapping and surveying, is projected to grow from $9.9 billion in 2026 to $37.6 billion in 2036, an average of 14.3 percent a year. Mapping and surveying account for 35 percent of it.
Where the Money Flows
The business has three layers. The first is collecting data. Satellites, aircraft, drones and lidar equipment photograph the ground. Lidar is a technology that measures distance with laser pulses and captures the three-dimensional shape of terrain and buildings. The second is software. Mapping software from Esri, the largest mapping software company in the world, and image analysis software such as NV5’s ENVI turn raw images into material that can be analyzed. NV5 is a US engineering and geospatial company. The third is services. Drone service companies, surveying firms and consultancies go to sites, survey them, and sell the analysis as reports and maps.
There is one unusual feature. The most basic map is free. The nationwide elevation data produced by the US Geological Survey and Landsat satellite images are public assets that anyone can use at no cost. Landsat is a series of US observation satellites that has photographed Earth since 1972, and its images are released free of charge. Private companies make their money on top of that. They take the free raw data and turn it into answers for a specific site, selling that work through software and services.
It is also clear where the money concentrates. In the GeoAI market, software held the largest share of 2025 revenue at 46 percent, while services were named the fastest-growing segment. In the drone services market, mapping and surveying are the largest use at 35 percent. Construction and infrastructure account for 23 percent of US commercial drone demand.
The Federal Data Backbone
The foundation of the US geospatial industry is public data made by the federal government. In February 2026, at Geo Week in Denver, the US Geological Survey announced that nationwide baseline data collection for the 3D Elevation Program, known as 3DEP, was complete. 3DEP is the USGS program that surveys the elevation of the entire United States with lidar to build three-dimensional terrain maps. After a twenty-year effort, it was the first time the whole country had been covered with lidar data of consistent precision. Alaska was covered with radar observation data instead. The USGS is now moving from a single national snapshot to a repeat observation system that remeasures areas older than five years. A 2022 joint study by NOAA and the USGS estimated the annual benefit of the next generation of 3DEP at $7.6 billion.
The reference system is changing too. The National Spatial Reference System is the framework that sets the standards for position and height across the country. The modernization replaces the existing NAD 83 and NAVD 88 with NATRF2022 and NAPGD2022. NAD 83 and NAVD 88 are the position and height reference systems the United States has used since the 1980s, and the new standards replace them in the 2020s. In effect, the horizontal and vertical framework that all elevation data hangs on is changing. Partnership funding from federal, state and local government and the private sector covered 72 percent of 3DEP costs between 2015 and 2023.
The Commercial Currents
The first commercial current is the artificial intelligence agent. NV5’s ENVI Agent packages an expert-designed analysis procedure as a tool an agent can run, so that non-experts can use it. The best entries may be considered for actual deployment through GeoAgent, NV5’s AI geospatial platform. If it works as designed, field organizations such as wildfire response teams could run analyses without a remote sensing specialist. This is also NV5’s product strategy, however, and there is no independent verification of the claim.
The second current is the everyday use of drone mapping. As of early 2026, 316,075 commercial drones were registered under FAA Part 107. Part 107 is the Federal Aviation Administration pilot certificate required to fly a drone for commercial purposes in the United States. Turning drone images into maps and three-dimensional models is now used to track progress at construction sites and inspect assets. Infrastructure modernization, wildfire response and precision agriculture support the demand.
The third current is the spread of reality capture. Reality capture means scanning a real space to create three-dimensional data. Lidar and three-dimensional scanning, scan-to-BIM and digital twins are becoming standard work in construction and facility management. Scan-to-BIM turns scanned three-dimensional data into a building information model, and a digital twin is a model that reproduces a real building or facility in virtual space. Three-dimensional models made from airborne and terrestrial lidar and photogrammetry now connect design, construction and maintenance. Photogrammetry is the technique of measuring distances and three-dimensional shapes from overlapping aerial photographs.
Four Bottlenecks
The first bottleneck is the last step. Detecting change is already a mature technology. The bottleneck is turning detection results into clean boundaries a decision maker can use right away. That is why ASPRS set wildfire burn perimeters as the task for this year’s student challenge. ASPRS is the American Society for Photogrammetry and Remote Sensing, the group that sets standards for photogrammetry and remote sensing. Finding the changed area is easy because vegetation reflects less near-infrared light after it burns, but turning a pixel-level change map into a usable line takes binarization, noise removal, polygon generation, boundary refinement and validation. Binarization means splitting an image into two values, black and white. In the challenge’s judging criteria, repeatability carries 35 percent, more than the 25 percent for analytic performance, which shows where the bottleneck sits.
The second bottleneck is repeatability. A workflow tuned to one fire or one image cannot be used in the field. It has to work consistently across images, times and regions.
The third bottleneck is people. The Geo Week program includes a session on a workforce development roadmap for geospatial work, as well as one on modernizing the National Spatial Reference System. Even the best tools reach no field without people to use them. Sixty-one percent of commercial drone operators named regulatory complexity as a major barrier.
The fourth bottleneck is verification. Most performance claims for commercial GeoAI tools come from the companies selling the products. Independent evaluations or comparisons are still scarce. Market forecasts are also hard to compare directly because each firm defines the market differently. Estimates for the 2026 global market range from $487.4 billion to $623.2 billion.
The Outlook
The near-term outlook is bright. The global geospatial solutions market is projected to reach $1,561.61 billion by 2034, growing an average of 13.29 percent a year. GeoAI and drone services are also expected to grow at double-digit rates. Defense and intelligence demand provides the floor, while infrastructure modernization and disaster response grow the commercial demand.
Four issues will decide the outlook. The first is whether funding keeps flowing for repeat observation under the next generation of 3DEP. Building the national map once is done, and the question now is whether older areas keep getting remeasured. The second is whether the National Spatial Reference System replacement reaches field surveying without disruption. The third is how the expansion of beyond-visual-line-of-sight drone flights is settled. Beyond-visual-line-of-sight flight means flying a drone to places the pilot cannot see with the naked eye. The fourth is whether the idea of AI agents running analysis without specialists leads to actual deployment. It could make burn perimeter mapping faster and cheaper for wildfire response, power grid restoration and land management, but it remains a possibility, and no deployment has been decided.
The places to check are set. Geo Week in Salt Lake City in February 2027 and the Esri User Conference in San Diego in July will show how far things have come. In this industry, the winners will not be the companies that detect best, but the ones that turn detection into a product a non-expert can use in the field tomorrow.