Planck

Planck is a Tel Aviv-based company (founded 2016) that builds an AI-first data enrichment platform focused on deriving structured, normalized intelligence about commercial properties and businesses from sparse inputs (name, address, tax ID). At its core the product ingests multi-resolution satellite and aerial imagery, street-level imagery where available, and structured/unstructured third-party datasets; applies computer vision, geospatial analysis, and NLP; and outputs standardized property attributes, risk indicators, and evidence assets (imagery snippets, change-detection timelines). This enables insurers and enterprise risk teams to replace or triage manual inspections and desk research with automated, repeatable intelligence products.

Commercially, Planck sits in the InsurTech/geospatial analytics stack as an upstream data and feature provider to underwriting platforms, portfolio risk managers, and claims teams. Sales cycles in this space tend to be long and procurement-driven, but the addressable market is large: property and casualty insurers, commercial real-estate risk management, and enterprise asset owners all value standardized, updateable property intelligence. Planck's traction signals reported publicly focus on platform integrations and deployments with insurance customers rather than rapid viral adoption; that is consistent with mid-stage enterprise software dynamics.

On the competitive landscape, Planck competes with companies that provide imagery, property analytics, or automated feature extraction. Competitors and substitutes vary by focus: some specialize in very-high-resolution rooftop and building geometry (EagleView, Maxar), others emphasize rapid urban imagery capture (Nearmap) or insurance-focused derived features (Cape Analytics). Planck's commercial differentiation is its multi-source fusion pipeline and the way it maps raw signals into insurer-relevant attributes and scoring — effectively products rather than raw imagery — which matters for procurement teams that want directly usable underwriting inputs.

From a technical perspective the platform combines matured components (CV models for object/structure detection, geospatial tiling and mosaicking, OLAP-style feature stores) with engineering to operationalize frequent re-analysis at portfolio scale. The engineering investment to keep models current, manage imagery licensing/costs, and provide explainable evidence for underwriting decisions is material and represents both value and an ongoing cost center.

For national security and defense audiences, the same property-characterization, change-detection, and multi-source fusion capabilities map onto GEOINT use cases: baseline mapping of commercial infrastructure, automated detection of new construction or damage, and rapid prioritization for human analyst review. That mapping is credible but not automatic: defense use typically requires higher-end imagery, classified integrations, and programmatic procurement that differ from commercial deployments. Planck's architecture and product focus make it a logical adjoint to GEOINT toolchains when imagery and access match requirements.