Panjiva aggregates and standardizes over one billion global trade transaction records, sourced primarily from customs declarations and supplemented by other government and licensed private sources, into a unified dataset of company-to-company shipments. Records include the buying and selling companies, products (classified using the HS code system), shipment volumes, logistics providers, and trade lanes.

S&P Global applies machine learning and natural language processing to resolve company identities across languages and data sources, standardize units of measure, and impute values (such as USD value or weight) where the underlying source doesn't report them directly. These methods make Panjiva useful for mapping a company's actual suppliers and customers, tracking shipment-level trade flows for a specific product or trade lane, and benchmarking trade activity across countries and industries.