Production, not retrieval
Edgar supports the construction of benchmark records before they are searched.
DataAlchemist provides structured benchmark data assets for transfer pricing, IP valuation, and advisory work.
Each asset is built from official source materials, normalized into comparable fields, enriched with context, and linked back to the evidence behind the benchmark.
CoveragePublic filings define the available universe. DataAlchemist's edge is completeness, currency, and substance: how source records are maintained, structured, enriched, and made usable for benchmarking.
Agreements, filings, patents, and market communications are used to construct records that preserve the relationship between the benchmark and the evidence behind it.
The purpose is practical: help professionals identify comparables, understand their context, and support analysis with records suitable for review.
Every benchmark record is produced, not merely collected and extracted. Edgar is DataAlchemist’s in-house, domain-trained AI model, embedded in a multi-stage production process that constructs each record before it is searched. It structures the agreement, classifies the asset by economic use, links evidence around the transaction, and prepares the record for domain-specialist review.
Most AI in professional data markets is applied at the interface: faster ways to search, screen, and select records after the data asset has already been built. Edgar is applied at production, where the benchmark record itself is constructed. A search layer, however capable, can only retrieve the structure, context, and substance already present in the data asset. Each DataAlchemist record may appear as a single row, but it is the output of a layered, multi-flow production system applied across the corpus.
Edgar supports the construction of benchmark records before they are searched.
Asset-level classification, payment structure, FAR, DEMPE indicators, and conduct evidence are built into the record.
A domain specialist confirms each record before it enters a benchmark set.
Edgar is the asset behind the assets.
Structured for transfer pricing, valuation, and advisory workflows.
License-agreement benchmarks built from filings, patents, and regulatory materials. Records include normalized royalty bases, licensed rights, IP type, exclusivity, payment structures, and asset-level industry classification.
Independent-party service fee benchmarks organized by service type, fee type, payment structure, and recipient industry. Used for service fee benchmarking, benefits tests, cost-plus support, management fees, and procurement arrangements.
Lease and rental benchmarks drawn from real estate, equipment, facility, and vehicle disclosures. Structured by asset type, lessee industry, lease term, and payment frequency to support intercompany leases, shared facilities, and cost allocations.
Clause-first search across publicly filed agreements and amendments. Built for research where the answer doesn't fit into a standard benchmark table: unusual fee structures, payment mechanics, rights language, and change-of-control provisions.
Connecting selected benchmarks directly to the materials behind them: agreements, clauses, patents, and filings. Designed to let users move from a record to its source documents, supporting audit review, expert reports, and dispute files.
We work from SEC EDGAR filings, USPTO and EPO patent records, court and regulatory filings, and public-company disclosures across multiple jurisdictions. The objective is to preserve the evidentiary chain, not just accumulate documents.
Terms are normalized to common fields and units, and descriptions enriched where source language is thin, so comparables align across agreements, industries, and jurisdictions.
Each asset is structured according to the analysis it supports. The common discipline is the same across all data: structured fields, source-linked records, contextual enrichment, and a comparability-oriented structure.
DataAlchemist prioritizes traceable, normalized, source-linked benchmark records over undifferentiated document volume.
Review the benchmark data assets behind DataAlchemist: royalty rates, service fees, lease benchmarks, contracts, and evidence packs.
Request a Walkthrough