How every number is grounded.

Every metric on a Qravitas dossier carries an evidence marker, and quantified narrative claims are independently checked against the company's own measured evidence before they are published. This page describes the mechanism plainly: what is measured, what is verified, and what the platform does not claim.

How we ground every number

The engine attaches a provenance tier to every metric it computes: measured (taken directly from a source, such as a filing figure or a performance test result), modeled (derived deterministically from measured inputs), or estimated (inferred, and labeled as such). A number without a provenance marker does not reach a dossier. This is enforced by automated sweeps across the full evidence base, not by editorial discipline.

On top of provenance, quantified claims in the narrative layer are independently verified: each claim is checked against the measured evidence for that company using a separate grounding-verification pass, and scored as verified or not. Across the evidence base today, 86.4 percent of quantified claims are grounded in measured evidence, measured across 3,607 claim-verification records covering 16,059 claims (measured 2 Nov 2026). The number is published as measured; claims that fail verification are flagged, not silently kept.

What the numbers count

The evidence base is built entirely from public signals. When a design partner connects internal systems through private connectors, that first-party data is isolated to that customer and used only for their own diagnostics; it never enters the shared evidence base or the cross-company graph.

A company is read when we hold at least one dated reading of it, built from public signals. Every number on this site comes from one measurement of the evidence base, taken on 2 Nov 2026:

• 11,413 companies read, each counted once after duplicate records of the same company are merged.
• 7,000+ in the free public lookup: companies we have read, shown with public facts only.
• 6,298 read in the last 30 days, of which 3,650 were re-reads of a company already on file.
• 122 companies with three or more readings taken on separate days. Only these can show a change over time, so this is the number that limits what we can say about change today. It grows with every re-read.
• 5 changes published. A change is published only when it is larger than the instrument’s own noise across three readings on separate days; anything inside that noise is not a finding, however large it looks.
• 357 board-grade briefs: the multi-pass written analysis, which covers a subset of companies read.
• 3,607 claim-verification records covering 16,059 quantified claims, of which 86.4 percent are grounded in the company’s measured evidence. Grounded is not the same as predictive.
• A median of 13 signal layers per company read. Not every layer is available for every company.

The layers and their sources, in plain terms:

• Technical debt and web performance: direct measurement of the company's public web surface (performance, stability, and rendering metrics), scored into a debt index and its revenue-friction consequences.
• Financial and forensic signals: public filings and disclosures, including year-over-year changes in stated risk language, read forensically rather than summarized.
• Go-to-market and funnel health: the observable marketing and conversion surface: search visibility, pricing pages, hiring signals, and the gaps between them.
• Cultural and narrative signals: public discourse and disclosure, scored deterministically, with language models used only to judge, never to compute.
• Peer grounding: every company is scored against its structural peer cluster in the cross-company knowledge graph, so findings are relative to real comparables, not absolutes.

Deterministic code computes every metric. Language models reason over the evidence and write the narrative, but they do not invent the numbers, and their quantified statements are then verified against the evidence as described above.

What we do not claim

Precision about scope is part of the methodology:

• Distress signals are leading indicators, not outcomes. They are built from disclosure and public signal; they flag elevated risk, they do not predict or confirm specific events.
• Board-grade narrative depth covers a subset of the evidence base today. Companies read carry a median of 13 deterministic signal layers, and not every layer is available for every company; the deepest multi-pass board briefs are being extended across the evidence base and we say so rather than imply otherwise.
• The homepage engine demo is illustrative. It simulates the pipeline for demonstration. The company lookup on the same page is the opposite: real public facts from the evidence base. The findings themselves are shown only in a paid workspace.
• We do not publish a time-per-audit figure yet. The steady-state sample is not large enough to state one honestly. When it is, we will publish it with the sample behind it.
• Verification is a rate, not a halo. 86.4 percent of quantified claims are grounded in measured evidence; the remainder are flagged. We publish the measured rate rather than rounding it to a promise.
• Your first-party data does not build our evidence base. The evidence base and the cross-company graph are public-signal only. Data you connect through private connectors stays isolated to your account and powers your diagnostics, nothing else.