What Earthdefine does

EarthDefine is a Redmond, Washington-based geospatial AI company that produces high-resolution, nationwide datasets of America's physical infrastructure and natural features. The company applies proprietary Deep Convolutional Neural Networks — trained on NAIP imagery and LiDAR data covering 50+ million acres and 120+ terabytes of aerial imagery — to extract and classify buildings, trees, impervious surfaces, parking infrastructure, and RF clutter at resolutions of 50cm to 1 meter. Its product portfolio includes Building Footprints 3D (191M+ structures with 98%+ accuracy), Tree Map, Canopy Height Model, Tree Locations, Land Cover, Impervious Map, Parking Lots, Clutter3D (RF clutter for telecom), and a Buildings API for programmatic access.

The company sells off-the-shelf data products under a one-time license model and offers custom geospatial AI solutions on a professional-services basis. Pricing is quote-based with multi-year enterprise contracts, and distribution combines direct consultative sales to government and Fortune 500 buyers with a developer-facing API playground. EarthDefine's customers include federal agencies (NOAA, EPA), state agencies (CalFire, Commonwealth of Massachusetts), Fortune 500 enterprises (NRG), research institutions (Oregon State University, UC Davis), and nonprofits (Audubon Society), with telecom carriers targeted through the Clutter3D product and integrations with RF planning tools including Forsk Atoll, EDX Signal Pro, Infovista Planet, and ICS Telecom.

EarthDefine was founded in 2016 by University of Michigan forestry and conservation graduates. The company is privately held, founder-controlled, and bootstrapped, with no disclosed venture capital funding, acquisitions, or leadership changes. It maintains a single Redmond, WA office and operates without disclosed certifications, patents, or formal regulatory compliance credentials.

Name

Earthdefine

Legal name

EarthDefine LLC

Website

https://earthdefine.com

Company type

Private

Founded year

2016

Operating status

Operating

Headcount range

11–50 employees

Short description

EarthDefine is a Redmond-based geospatial AI company producing high-resolution nationwide datasets of buildings, trees, and land cover for federal agencies, telecom carriers, Fortune 500 enterprises, and research institutions using proprietary deep learning models.

Ownership category

akta.pro rank

Product category

Geospatial Data and Analytics

NAICS

Geophysical Surveying and Mapping Services (54136), Geophysical Surveying and Mapping Services (541360)

SIC

Oil & Gas Field Exploration Services (1382)

akta.pro primary industry

Satellite Imagery & Remote Sensing Geospatial Platforms (BPAMANAM)

akta.pro secondary industries

Remote Sensing & Geospatial Vision (satellite/drone imagery) (HDAAAEAL), Remote Sensing & Earth Observation Analytics (Satellite/UAV/LiDAR, Image Processing) (EUAHAKAD)

Keywords

• Geospatial data products
• Building footprint mapping
• Tree canopy analysis
• RF propagation clutter
• LiDAR-based mapping

Headquarters

HQ city

Redmond

HQ country

United States

HQ region

North America

Offices1 record

Markets served

GTM type

B2B

Offering type

Software

Cost components

Technology or R&D, Personnel, Infrastructure, Operations, Marketing or Sales

Revenue model

1. Off-the-shelf Geospatial Data Products: Pre-produced nationwide datasets (Building Footprints 3D, Tree Map, Land Cover, Impervious Map, Parking Lots, Tree Locations, Canopy Height Model, Clutter3D) sold as downloadable data products with accuracy specifications and resolution standards.
2. Custom AI Solutions: Tailored geospatial intelligence for specialized applications beyond standard product offerings, delivered as bespoke solutions for complex geospatial challenges.
3. On-demand Land Cover Mapping: High-resolution (60cm) land cover mapping delivered on-demand for any area in the contiguous US, typically within days of placing an order.

Pricing tiers

Go-to-market motion1 record

Distribution channels2 records

Marketing channels5 records

Core offering

EarthDefine produces and licenses off-the-shelf, nationwide geospatial data products (building footprints, tree maps, canopy heights, impervious surfaces, RF clutter, parking lots, and land cover) generated by proprietary Deep Convolutional Neural Networks trained on high-resolution NAIP imagery and LiDAR data. The company also delivers custom AI geospatial solutions and provides API access to its Building Footprints data for programmatic integration into enterprise and government workflows.

Product overview

EarthDefine is a geospatial AI company offering a platform of eight core data products plus API access for mapping America's infrastructure and natural features. The portfolio includes Building Footprints 3D (191M+ 3D building outlines), Tree Map (60cm tree canopy), Canopy Height Model (LIDAR-derived tree heights), Tree Locations (individual tree vectors), Clutter3D (20-class RF clutter for telecom), Land Cover (7-class classification), Impervious Map (built-up surfaces), and Parking Lots (infrastructure mapping). All products are AI-powered using proprietary neural networks analyzing high-resolution aerial and satellite imagery, delivering >95% accuracy. Data is updated annually and integrates with major RF planning tools (Atoll, Infovista Planet, ICS Telecom) and GIS platforms (ArcGIS, QGIS). The company serves federal agencies, Fortune 500 companies, and universities.

Differentiator

Problem solved

Functional benefit

Brands

• Clutter3D: Nationwide RF clutter dataset at 1-meter resolution with 20-class schema and LiDAR-measured heights, designed for telecom RF propagation modeling and 5G planning

Products and services

• Building Footprints 3D Nationwide 3D building outlines covering 191M+ structures mapped annually with greater than 98% accuracy. Delivers rooftop-precise geocoding and 3D attributes (heights, estimated stories, volume) for 151M+ structures, updated quarterly with new imagery and LiDAR data. Sold to government, enterprise, and telecom customers for broadband mapping, geocoding, 5G planning, insurance underwriting, and city planning.
• Canopy Height Model High-resolution (50cm) raster dataset mapping tree canopy height as a continuous surface, with each pixel representing tree overstory height above ground topography derived from LiDAR data. Covers 85%+ of the United States at 90%+ accuracy, with heights delivered in feet or meters for solar feasibility studies, RF planning, and forestry applications.
• Clutter3D RF clutter dataset with 20-class land use classification and LiDAR-measured height attributes at 1-meter resolution, purpose-built for telecom network planning including 5G, BEAD broadband modeling, and FCC compliance. Delivered as a 5-layer product (clutter classes, DSM/AGL, DEM/AMSL, building vectors, imagery) covering 3,125,942 square miles across 50 states, 3,235 counties, and 32,629 cities, import-ready for Atoll, EDX Signal Pro, Infovista Planet, ICS Telecom, Esri ArcGIS, and QGIS.
• Impervious Map Nationwide impervious surface mapping at 60cm resolution covering buildings, roads, sidewalks, parking lots, and airports. AI model processes over 120 terabytes of color-infrared aerial imagery spanning 3.1 million square miles with 94.9% accuracy in census-defined urban areas. Updated annually to capture changes in the built environment for TMDL compliance, stormwater modeling, and urban growth analysis.
• Land Cover On-demand high-resolution (60cm) land cover mapping with 7 ground cover classes (Herbaceous, Bare, Impervious, Water, Trees, Shrubs, Trees over Impervious). Delivers greater than 95% accurate data typically within days of ordering for stormwater compliance, habitat assessment, and environmental planning applications.
• Parking Lots Nationwide parking infrastructure mapping with AI-derived polygons for outdoor parking spaces over 1,000 square feet. Address-linked with adjoining building footprints for complete spatial context. Updated annually with new imagery for retail site selection, commercial real estate valuation, and urban planning.
• Tree Locations High-resolution vector dataset mapping individual trees as points, polygons, and idealized circular crown polygons with attributes including Tree ID, Maximum Height, Mean Height, and Crown Area. Tree locations and heights derived from LiDAR; covers 85%+ of the U.S. at greater than 95% accuracy and is updated annually for 3D urban modeling, powerline risk assessment, and canopy management.
• Tree Map Nationwide tree cover intelligence at 60cm resolution (2,500 times sharper than NLCD public data), mapping woody vegetation with 96.6% overall accuracy and 97.3% accuracy in urban areas. Updated annually with changes from deforestation, urbanization, land-use change, and natural disasters, supporting carbon accounting, urban forestry decisions, and environmental compliance.
• Buildings API Programmatic API access to Building Footprints 3D data enabling rooftop-precise geocoding, retrieval of 3D building attributes (heights, elevations, estimated stories, volume), and address connectivity for seamless integration into enterprise and government workflows. A demo and playground is available at playground.earthdefine.com/demo.
• Custom AI Geospatial Solutions Tailored geospatial intelligence delivered as bespoke professional services for specialized applications beyond the standard off-the-shelf product offerings, addressing complex geospatial challenges for enterprise and government clients.

Quantifiable outcome

• 98%+ building footprint accuracy (pixel-level) and 94%+ IoU accuracy
• +6 more outcomes

Named customers8 records

Segments4 records

Ideal customer profiles4 records

Technology focussed Yes

API detail

Has API

Yes

API docs

API detail

Core technology

AI maturity

App detail

Integration8 records

AI capability4 records

Feature5 records

Partnerships

One partnership is on record.

• PlanIT GeocoreStrategic or Co-development PartnerEarthDefine has partnered with PlanIT Geo to deliver pre-analyzed tree cover data through TreePlotter Canopy, providing users with easy-to-use interactive maps, tree equity scores, ecosystem benefits analysis, and community engagement tools without requiring specialized GIS staff or software.

Scale indicators7 records

Recent moves6 records

Expansion highlights5 records

Emerging players

• Picterra: Picterra offers a geospatial AI platform for detecting features in aerial and satellite imagery with a self-serve SaaS model. It is comparable to EarthDefine as an AI-driven feature extraction player, though it focuses on a platform/training model approach rather than EarthDefine's pre-built nationwide datasets.
• PlanIT Geo: PlanIT Geo is a strategic partner of EarthDefine (TreePlotter Canopy integration) and a provider of urban forestry management software and analytics. It is comparable to EarthDefine as an adjacent player in tree canopy intelligence for municipal and utility customers, with overlapping go-to-market in urban forestry.

Broad incumbents

• Mapbox: Mapbox provides a mapping and location platform used by enterprise and developer customers, with its own building footprints, boundaries, and movement data. It competes with EarthDefine in supplying authoritative geospatial base layers (especially buildings and places) for downstream applications.
• Esri (ArcGIS): Esri is the dominant GIS platform with extensive partner ecosystem data layers including building footprints, land cover, and imagery-derived products. EarthDefine's Clutter3D and Tree Map are explicitly ArcGIS-compatible, and Esri competes as both a platform and (via partners/data layers) a provider of overlapping geospatial intelligence.
• Maxar Technologies: Maxar is a major satellite imagery and geospatial intelligence provider offering high-resolution earth observation imagery and derived products. It overlaps with EarthDefine as a source of high-resolution aerial/satellite-derived geospatial data for government and enterprise customers, but operates at a much larger scale and across global defense and intelligence markets.
• Planet Labs: Planet Labs operates a large constellation of earth observation satellites delivering daily global imagery and analytics. While broader in scope and global reach than EarthDefine, both sell geospatial intelligence derived from remote sensing imagery to government, agriculture, forestry, and commercial customers.

Direct peers

• Nearmap: Nearmap provides high-resolution aerial imagery and derived geospatial datasets (buildings, vegetation, impervious surfaces) for government, insurance, and telecom use cases. It is one of the closest direct comparables to EarthDefine's off-the-shelf nationwide geospatial product strategy.
• EOS Data Analytics: EOS Data Analytics provides satellite imagery-based analytics for agriculture, forestry, and environmental monitoring, including land cover and tree cover classification. It is comparable to EarthDefine's Tree Map and Land Cover products in deriving thematic geospatial layers from remote sensing imagery.
• CyberCity 3D: CyberCity 3D provides 3D building models and geospatial data products derived from imagery and LiDAR for telecom, simulation, and government markets. It directly overlaps with EarthDefine's Building Footprints 3D and Clutter3D offerings in serving RF planning and 3D urban modeling use cases.
• Descartes Labs: Descartes Labs builds geospatial analytics on satellite imagery using ML/AI for agriculture, energy, and sustainability applications. It is comparable to EarthDefine in applying deep learning to remote sensing imagery to produce consumable analytics products for enterprise and government buyers.

Market position

Strengths4 records

Weaknesses5 records

Competitive moat4 records

Key risks6 records

Key highlights7 records

Customer concentration

• Xx.com/earthdefinellc

Revenue estimate

Valuation estimate

Number of profiles

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Earthdefine