Home › Guides › Latin America company data
Latin American company data: tax registries, partner lists and five million businesses
Latin America's business data lives in tax registries rather than commercial company registers — and that turns out to be an advantage. A tax number is mandatory to operate, so coverage is near-total, and Brazil goes further by publishing the partner list of every company. This guide covers what Brazil, Mexico, Peru and Colombia each publish and which one answers which question.
Tax number as the identity layer
In most of the region the identifier you need is a tax number, not a company number:
| Country | Identifier | Issued by |
|---|---|---|
| Brazil | CNPJ (14 digits) | Federal tax authority |
| Peru | RUC | SUNAT |
| Mexico | RFC (tax) / DENUE unit id (directory) | SAT / INEGI |
| Colombia | NIT | DIAN |
Because operating without one is not possible, coverage is effectively complete — a real advantage over jurisdictions where registration is optional for small traders.
Brazil: the partner list is the story
Brazil's CNPJ registry publishes what most countries treat as sensitive: the QSA — Quadro de Sócios e Administradores, the list of partners and administrators registered against the company. Combined with CNAE activity codes and registered address, it gives a fuller picture of control than many European registers manage.
In our measured sample of Brazilian records, the fields that matter for outreach were unusually complete: company name, registration number, address and activity on every row, and — notably — phone numbers and partner lists on every row too. Among the country registries we measure, Brazil is the strongest for contactability.
Mexico: five million establishments, geocoded
Mexico's DENUE is a different kind of dataset. Maintained by INEGI, the national statistics institute, it is a directory of roughly five million economic units — establishments, not legal entities — with name, activity, size band and a geocoded address.
That distinction matters. DENUE will not tell you whether a company is legally in good standing; it will tell you that there is a hardware retailer of 11–30 employees at a specific coordinate in Guadalajara. For territory planning, field sales routing and market sizing it is exceptional. For legal verification it is the wrong tool.
Peru and Colombia: taxpayer status
Peru's SUNAT RUC registry answers the compliance question directly: is this counterparty a registered, active taxpayer? Name, status and activity are published, and for invoicing in Peru that status field is the one that matters.
Colombia's records similarly centre on entity identity. Neither is a rich firmographic source — treat them as verification layers rather than prospecting databases, and pair them with firmographic enrichment when you need size and sector.
Peru SUNAT data → · Colombia company data →
Choosing by question
| Question | Source |
|---|---|
| Who controls this Brazilian company? | Brazil CNPJ — QSA partner list |
| Where are the businesses in this Mexican city? | Mexico DENUE — geocoded, size-banded |
| Is this Peruvian counterparty an active taxpayer? | Peru SUNAT RUC |
| I need a phone number | Brazil (highest coverage in our measurements) |
| Cross-border group structure | GLEIF LEI data — global entity identity and parents |
Two practical cautions
- Establishment ≠ legal entity. DENUE counts locations. A chain with 200 stores appears 200 times. Deduplicate before you report "number of companies", or you will overstate the market badly.
- Partner lists are personal data. Brazil's QSA names individuals. It is published by the state, but Brazil's LGPD — its GDPR equivalent — still governs how you process it. Company-level use is defensible; building a marketing list of named partners is a different proposition.
The tax number is the company number
The structural fact that defines Latin American company data, and the one that surprises people arriving from Europe: in most of the region the tax identifier is the corporate identifier. There is no separate "company number" issued by a registrar and a tax number issued by a revenue authority. There is one number, issued by the tax administration, and it is how the state knows the business exists.
| Country | Identifier | Shape |
|---|---|---|
| Brazil | CNPJ | 14 digits: 8-digit root (the company) + 4-digit branch + 2 check digits |
| Mexico | RFC | 12 characters for legal entities, tied to name and incorporation date |
| Colombia | NIT | Tax number plus a check digit |
| Peru | RUC | 11 digits, prefixed by entity type |
Brazil's CNPJ deserves a closer look because its shape encodes the group structure directly. The first eight digits are the root — the legal entity. The next four identify the establishment: 0001 is the head office, and each branch increments. So two CNPJs sharing a root are the same company, and you can tell that from the number alone without any matching logic at all.
That is a rare gift in company data. It means deduplication in Brazil is a string operation rather than a fuzzy-match problem — truncate to eight digits and count distinct. Compare that with the French SIREN/SIRET split, which encodes the same idea, or with jurisdictions where establishing that two records are one company requires address and name comparison and a tolerance for being wrong.
The practical rule for the region: never treat the tax ID as sensitive infrastructure data. It is the public identity of the business, printed on invoices and displayed on storefronts. It is also, almost always, the only reliable join key you will get.
Frequently asked questions
What is a CNPJ?
Brazil's 14-digit national registry number for legal entities, mandatory to operate and the join key for Brazilian business data.
What is the QSA?
The published list of a Brazilian company's partners and administrators — effectively an open ownership record.
Is DENUE a company register?
No. It is a statistical directory of about five million establishments, ideal for location and market work, not for legal verification.
Which country gives the best contact coverage?
Brazil, in our measurements — phone numbers and partner lists populated across the sample we ran.