Decision-grade satellite intelligence, organized by the question you actually need to close.
Five operations-led verticals where Spica's sub-meter revisit and 90-minute change-detection change the answer — illegal extraction caught mid-act, crop loss sized before harvest, methane plumes confirmed before the next quarterly disclosure.
One constellation, one latency floor, one accuracy number.
Whether the question is crop loss, methane, encroachment, claims fraud, or maritime incursion, the underlying service is the same: an 11-satellite Spica constellation delivers 0.7 m native imagery with a 4-hour median revisit across 94% of Earth's landmass, a fixed 78-minute tasking-to-delivery latency benchmarked by the Open Geospatial Consortium in Q1 2025, and a 97.3% change-detection accuracy on the SpaceNet-7 flood and building-damage benchmark.
Raw pixels are filtered onboard by a neural inference stack that reduces downlink volume 22× compared to raw raster shipping — cutting customer bandwidth costs in published 2024 benchmarks — then fused into verified events: illegal dredging, crop loss, methane plumes, infrastructure encroachment. The verticals below are the operational questions we are built to answer; the underlying physics of acquisition, the latency floor, and the change-detection model are identical across them.
- Sub-meter0.7 m native PAN, 4-hour median revisit over 94% of landmass
- 78 minutesmedian tasking-to-delivery, OGC-benchmarked Q1 2025
- 22×onboard inference cuts downlink vs. raw raster shipping
- FedRAMP ModerateISO 27001, SOC 2 Type II — sole EO provider authorized
The operational question each vertical is built to close.
Five cards. Five questions. Pick the one that matches the decision on your desk this quarter.
Which of our 1,400 fields failed this week — and which are about to?
Sub-meter NDVI deltas, irrigation failure alerts, and yield-loss sizing within hours of overpass — for cooperatives, reinsurers, and input retailers running continuous acreage monitoring.
Read the agriculture brief →
Where is methane venting right now, and what does it cost us in the next disclosure?
Plume confirmation, source attribution, and quantification across upstream, midstream, and LNG operations — fed into LDAR programs and ESG reporting pipelines.
Read the energy brief →
Who is dredging, anchoring, or trans-shipping in waters we are responsible for?
Vessel detection, dark-vessel activity, and sediment plume verification for port authorities, navies, and conservation agencies — delivered within 78 minutes of acquisition.
Read the maritime brief →
Which claims are real, which are inflated, and which happened somewhere else entirely?
Per-parcel damage classification, fraud signal detection, and CAT response imagery for property, crop, and parametric reinsurance — sized before the adjuster is dispatched.
Read the insurance brief →
What changed on the ground since the last reporting cycle, and can we prove it?
Continuous monitoring of land use, deforestation, illegal mining, and critical infrastructure — through FedRAMP Moderate-authorized infrastructure and a published Microsoft Planetary Computer partnership covering 3.8 million km² of tropical forest.
Read the government brief →We used to commission a manual image review every quarter and wait six days for the report. Now events arrive in the inbox before the next operations stand-up, with a verified raster chain we can hand straight to legal. The shift was less about resolution and more about cadence — continuous algorithmic monitoring changed what counts as a surprise.
What buyers ask before they book.
Short answers, drawn from the published benchmarks and certifications referenced throughout this site. If anything below is unclear for your environment, the demo will resolve it.
How does tasking actually work, and what is the realistic latency from order to analyzed tile?
Tasking is submitted via the Spica Insight API, the web console, or an ArcGIS / QGIS plugin. A standard new-collection task is confirmed within minutes; the analyzed tile (raw imagery, atmospheric correction, and change-detection overlays) is delivered at a median of 78 minutes after acquisition, benchmarked independently by the Open Geospatial Consortium in Q1 2025.
What uptime and service-level commitments back the platform?
Spica publishes 99.9% platform uptime. Across the trailing 12 months the measured uptime has been 99.97% across 4.2 million analyzed tiles served daily to enterprise customers. Constellation health and acquisition timing are visible in the public mission status endpoint and in every customer's console.
How does the platform integrate with our existing GIS and analytics stack?
Outputs are delivered as Cloud-Optimized GeoTIFFs, STAC items, and change-event JSON over a documented REST and streaming API. Native plugins exist for ArcGIS Pro, QGIS, and Snowflake; an open-source Python SDK and the Spica-Change PyTorch library (cited in 140+ peer-reviewed papers) cover custom model integration.
Can government and regulated customers meet their data residency requirements?
Yes. Spica is SOC 2 Type II and ISO 27001 certified, and as of November 2024 is the sole Earth observation provider with FedRAMP Moderate authorization. Data residency options include U.S.-only, EU-only, and Singapore-only processing regions; raw and analyzed tiles can be confined to a chosen region for the lifetime of the subscription.
How is pricing structured, and what drives the number we see in a quote?
Pricing is published in three tiers: per-area continuous monitoring subscriptions, per-task ad hoc collection, and platform API consumption for teams integrating Spica change detection into their own product. The drivers are AOI size, revisit frequency, and the depth of change-detection overlays required. Specific figures are scoped during a demo against the buyer's actual AOI.
Ready to see Spica on your own area of interest?
Thirty minutes. One of your sites or assets. A live walk-through of the tasking, change-detection, and integration paths above, scoped to a question you actually need to close this quarter.