How the platform actually works.

186 sources. 16 markets. Three independent layers.

Government capital signals do not arrive in a single feed. They are scattered across federal procurement portals, parliamentary records, regulatory dockets, ministerial press offices, multilateral approval systems, and sovereign disclosure pages. Each source publishes in its own format on its own cadence with its own taxonomy.

Svatantra ingests from 186 official sources across 16 markets daily. Every record gets normalized into a single signal schema, classified, scored, and tagged. The platform sits on top of three independent evidence layers that, taken together, answer questions no single layer can answer alone.

Capital signals. Capital flows. Confirmed outcomes.

Each layer answers a different question. Each runs on different sources. Each tells you something the others cannot. Read in sequence, the three layers form the complete thesis behind any signal.

Five phases. One automated cycle. Every day.

Every signal that ever lands in a customer dashboard has flowed through the same five-phase pipeline. The pipeline runs daily without manual intervention. New signals are typically visible to customers within hours of original publication.

Every signal classified the same way. No exceptions.

Classification turns unstructured government text into structured records that can be filtered, scored, and compared. Every signal moves through a five-layer classification pipeline: deterministic overrides for known patterns, an ICP relevance filter that strips noise, a two-pass classifier that assigns sector and signal type, schema validation, and confidence routing.

The taxonomy itself was built bottom-up from the actual government corpus rather than top-down from a generic industry classification. It reflects how governments actually structure capital deployment, not how a market-research firm might categorize industries.

53,465 confirmed outcomes. Three independent markets.

Scoring is calibrated against an outcome corpus drawn from three independent markets. The calibration set tells the platform what historically followed signals that looked like the one a customer is reading right now. Every confirmed outcome makes every future score more accurate.

The calibration set is independent by design. Outcomes used for calibration are downstream events (confirmed awards, disbursements, enforcement records), separate from the upstream signals being scored (RFPs, consultations, draft regulations, pre-solicitations). This separation matters because it prevents circular calibration, where the same record produces both signal and outcome.

Every signal carries 18 dimensions of intelligence.

Classification tells you what a signal is. Scoring tells you what it means. Each signal carries 18 dimensions of intelligence covering capital flow likelihood, market structure, regulatory environment, and timing. These dimensions are computed deterministically against the calibration set and presented to customers in a single signal record.

The headline composite score, displayed as a number from 0 to 100, is a weighted blend of nine of the eighteen dimensions. The remaining nine are presented alongside as informational context. Customers see both the composite and the underlying dimensional breakdown for any signal.

The specific dimensions, the weights, the calibration table, and the methodology behind each score are proprietary. They are part of the platform's structural intelligence and are not disclosed publicly. Customers see the values for every signal they touch. Competitors do not.

Every output is sourced. Every claim is traceable.

The platform was built around the principle that intelligence outputs must be defensible. Every signal, score, and brief carries the evidence behind it. Eight evidence principles attach to every customer-facing output:

Intelligence, not advice.

The platform surfaces signals, classifies them, scores them against historical patterns, and presents the evidence. Customers interpret. The platform does not recommend trades. It does not project the future. It does not produce financial advice or regulatory advice or investment recommendations.

This distinction is deliberate. Government capital signals are inputs to a decision, not the decision itself. The customers who get the most out of the platform are the ones who pair it with their own analytical workflow. The platform's job is to make sure the signal reaches them in time, with the evidence they need to act on it.