Geosite
Winter 2019AcquiredGeosite is an enterprise SaaS platform for spatial data.
Pressure-test this opportunity
Explore the risks and possibilities with a prompt for ChatGPT, Claude, or your agent.
Geosite is an enterprise SaaS platform for spatial data.
Explore the risks and possibilities with a prompt for ChatGPT, Claude, or your agent.
Geosite began with a blunt observation from defense work: organizations could buy satellite imagery, drone feeds, sensor data, and predictive models, yet the people making decisions still worked from stale maps and disconnected files. Founded by Rachel Olney in 2018, the company built a neutral layer for finding, combining, and delivering geospatial information without forcing a customer to become a geographic-information-systems specialist.
The horizontal platform found buyers in defense, energy, logistics, and infrastructure. Insurance became the clearest repeatable use case. Geosite's Bedrock layer joined outside datasets and models; its Ascend application turned those inputs into property underwriting, catastrophe response, portfolio monitoring, and claims work. The company raised a verified $5 million Series A and attracted carrier partners and investors including MS&AD and QBE.
Descartes Labs acquired all of Geosite's operating assets in April 2024 for undisclosed terms. Olney moved over to lead the buyer's insurance technology business, and EarthDaily retained Ascend after acquiring Descartes Labs later that year. Geosite never reached public markets, and the financial record is too thin to call the deal a large venture return. The acquisition placed its workflow beside proprietary analytics and imagery, where it was more useful than an independent neutral marketplace.
Image 1 / 1
Olney grew up near the United States-Mexico border in New Mexico and studied product design and engineering at Stanford. Her doctoral research at Stanford's Center for Design Research examined how large bureaucracies adopt technology, including work with U.S. Cyber Command. The research put her close to teams that possessed sophisticated systems but struggled to get timely information into everyday decisions.
The failure was consequential. Olney recalled, "People were going into critical missions with outdated geospatial data. The gap in information was causing casualties — and that was untenable." She saw commercial satellites and connected sensors multiplying while the software used to find and combine their output remained fragmented. "I became determined to solve this problem, and I realized I had the toolkit to do it," she later told Threshold Ventures.[1]
Geosite started in defense. "Our first ever contract was with Special Operations Command," Olney said. The job was to make commercial satellite imagery usable by people who were not geospatial specialists.[2] That distinction shaped the company: the hard part was not creating another image or model, but moving the right evidence from many suppliers into the software where a decision was already being made.
A 2019 Forbes profile described Geosite as a subscription product combining satellite, drone, and Internet-of-Things data. It reported $1.6 million in U.S. Air Force contracts and quoted Olney calling the product the "Google Drive, or Switzerland, of geospatial."[3] The Switzerland metaphor captured both the opportunity and the eventual constraint. Geosite could remain neutral across suppliers, but it would depend on those suppliers for the underlying information and analytical advantage.
Geosite's foundation was Bedrock, a data-integration layer that gathered imagery, sensor feeds, property attributes, and analytical outputs from many vendors. It normalized access and let an enterprise deliver selected evidence through a browser application, an API, or bulk data. Customers could change providers without rebuilding an entire internal workflow.
That horizontal engine supported varied early work: defense teams accessing commercial imagery, energy operators examining methane and infrastructure, and logistics or asset-management teams monitoring physical locations. Geosite deliberately avoided producing its own imagery or competing with model vendors. Olney said the company made a "very conscious decision" to stay out of both data production and analysis.[2]
Insurance turned the integration layer into a named application. Ascend accepted a carrier, managing general agent, or reinsurer's property portfolio and attached information relevant to specific decisions. Underwriters could examine a location before binding it. Catastrophe teams could identify exposed properties after an event. Claims staff could triage a book using post-event evidence. Portfolio managers could monitor changing hazards and prepare regulatory work.
Partnerships moved the product into systems buyers already used. Socotra added Geosite to its App MarketPlace in 2022, connecting spatial information to carrier operations.[5] MS&AD companies tested aerial imagery for loss prevention, mitigation, and faster claims handling.[6] QBE Ventures highlighted Geosite's ability to reconcile addresses with multiple geocoders, attach ground truth, and feed both underwriting and claims systems.[7]
Geosite sold to organizations with physical assets, regulated procurement, and expensive location-dependent decisions. The buyer was usually an enterprise team, government program, carrier, MGA, or reinsurer rather than an individual analyst. These customers already had data contracts and core systems; they needed evidence assembled, governed, and delivered without a large internal geospatial staff.
The public sector required a route through procurement as much as a technical integration. In 2021 Carahsoft put Geosite on SEWP V, ITES-SW2, and OMNIA Partners vehicles. Olney described administrative red tape as an obstacle to acquiring the tools needed to use geospatial information.[8] In insurance, carrier investors and software marketplaces served a similar role: they supplied trust, integration access, and a path into operational budgets.
Geosite addressed several large markets, but public evidence does not support one defensible top-down number. Spending on Earth observation, defense intelligence, property insurance software, and catastrophe analytics overlaps; adding those categories would count the same decision more than once. A practical market was the number of portfolios and workflows for which better property evidence changed underwriting, claims, or catastrophe response enough to justify enterprise software and data fees.
Current Ascend documentation shows the depth of that vertical. EarthDaily lists more than 61 vetted partner sources and supports underwriting, natural-catastrophe response, portfolio monitoring, claims, and regulatory work through a user interface, API, and bulk delivery.[12] The continuation does not reveal Geosite's independent revenue, but it shows that the use case survived ownership changes.
Geosite sat between raw data suppliers, analytical vendors, carrier software, and in-house geospatial teams. Its neutrality was useful when no supplier covered every geography, peril, or decision. It was weaker when a source owner could bundle capture, analytics, and workflow.
That bundling is visible today. ICEYE sells near-real-time catastrophe intelligence built on its own synthetic-aperture radar constellation, including building-level flood and damage information.[13] Nearmap pairs its aerial capture with post-disaster imagery and computer vision for property triage.[14] EarthDaily now combines Ascend with a broader observation and analytics business. Core insurance platforms can also integrate selected sources directly.
Geosite's defensible asset was therefore not simple aggregation. It was the accumulated mapping between providers, property identities, user permissions, and operational decisions. That knowledge became more valuable in a vertical application than in a general catalog.
Geosite sold enterprise subscriptions and access to its platform rather than charging consumers. The likely contract combined software, implementation, and usage or data costs, although public materials do not disclose pricing or gross margins. Data-provider licensing was an important economic dependency: a neutral platform had to pay for or pass through outside inputs while supporting integrations and enterprise service.
Channel partnerships reduced the sales burden. Carahsoft enabled government purchases. Socotra offered distribution inside an insurance core platform. Carrier investors supplied domain credibility and potential deployments. The pattern fits complex enterprise infrastructure, where procurement approval and system integration often matter more than self-serve sign-up.
The company raised a verified $5 million Series A in July 2021, co-led by Lavrock Ventures and NextGen Venture Partners, with Bee Partners and MS&AD Ventures participating.[4] QBE Ventures later invested. Funding databases disagree on the cumulative total, so claims between $5.4 million and $7 million should be treated as unresolved rather than averaged.
The clearest early traction was government contracting. Forbes reported $1.6 million in Air Force contracts by late 2019. Geosite later announced its first program of record, although the associated $950 million ceiling covered many awardees and was not Geosite revenue.[9]
Insurance traction is better evidenced through partner actions than disclosed sales. MS&AD ran a joint demonstration, Socotra listed the application, and QBE Ventures invested after studying the product. Olney said general revenue had grown roughly threefold year over year in a 2022 interview, but the company did not publish audited revenue, ARR, customer count, retention, or profitability.
The strongest validation came after the sale: the product continued. Descartes Labs said it acquired all operating assets, combined Bedrock and Ascend with its existing analytics, and placed Olney in charge of insurance technology.[10] Geo Week described the combination as an insurance offering drawing from dozens of geospatial providers.[11] EarthDaily acquired Descartes Labs six months later and explicitly retained Ascend.[15]
Geosite's central choice was to own neither data capture nor analytical models. That made the product credible as a neutral integration layer and let customers choose among suppliers. It also meant the company carried integration, licensing, and quality obligations without owning the upstream margins or unique source data.
Olney stated the choice plainly: "We are not going to do either of those because we made that very conscious decision to stay out of producing and analyzing data."[2] As imagery owners and analytical vendors moved downstream, neutrality alone became less distinctive. The acquisition joined Geosite to Descartes Labs' analytics; EarthDaily then placed it beside a broader observation business. Each step reduced the separation between source, model, and workflow.
Defense supplied the founding problem, but each new industry brought different data, terminology, buying processes, and decisions. Energy methane monitoring, logistics, infrastructure, and insurance could all use the same integration engine; they could not all share one sales motion or application.
Insurance provided a tighter loop. The same portfolio passed through underwriting, catastrophe preparation, event response, claims, and regulation. A carrier could pay for a workflow rather than a general-purpose data layer. Bedrock remained underneath, but Ascend became the customer-facing reason to buy. The company did not abandon horizontal infrastructure; it learned that infrastructure needed a vertical product to earn a durable budget.
Government contract vehicles, carrier investors, and core-system marketplaces carried the software into production. Geosite had to satisfy security, procurement, data rights, and integration requirements before a user could benefit from its interface. That made sales slower and services heavier than a generic SaaS label suggests.
The partnerships were rational, but they also concentrated bargaining power in larger platforms. Socotra controlled a system of record; carriers controlled operating data; imagery and analytics suppliers controlled inputs. A strategic buyer with established enterprise distribution could spread those costs across a larger business.
The counter-case is that Geosite did exactly what a venture-backed infrastructure startup should do: found a valuable vertical, attracted strategic investors, and sold into a platform that could scale it. Product continuity and Olney's leadership role support that reading.
The missing financial record prevents a stronger conclusion. Descartes Labs bought all operating assets rather than announcing a conventional share purchase, and consideration was undisclosed. No public source establishes whether investors earned a gain, employees received meaningful proceeds, or Geosite had a viable path to independent profitability. The defensible outcome is narrower: the company ceased independent operation, its technology survived, and the buyer filled structural gaps in analytics and distribution.