Research & Visibility
An open data commons and a funding catalogue
Two independent 2025 mappings of the same ecosystem disagree by a factor of two on Brazil’s share. Build an open regional data commons and a normalized catalogue of public funding programs.
The rationale
There is no auditable count of the region’s deep tech companies, and the disagreement is not marginal. The LADP mapping puts Brazil at 1,048 companies, 40.8% of the region; EMERGE’s 2025 radar puts Brazil at 952, 72.3%. They also disagree on rank order, with Mexico second in one and fourth in the other. Funding totals fare no better: LATAM deep tech funding for 2024 reads as US$138m in one tracker and US$536m in another, cited fifteen pages apart in the same report with no reconciliation. Every downstream figure inherits the ambiguity in its denominator.
What ships first is a definition and a publication rule, and the database follows from them. Publish the inclusion criteria, the vertical taxonomy already in use, the IDB’s twelve categories, and the company-level field list, then let any contributor submit records against them and publish the accepted and rejected changes. Where two sources disagree, the commons carries both with the definitional reason, which turns a contradiction into information. The funding catalogue is the easier half: a scrape and normalisation of public program data into funding type, ticket range, eligibility, deadline and decision time, refreshed on a stated cadence.
The claim that this cuts diligence friction is a hypothesis, not a finding; nobody has measured diligence friction in LATAM. What is measurable is the reproducibility gap, and it is total, since no third party can currently verify a single LADP-original figure. One constraint belongs up front rather than in a footnote: a pan-regional IP registry would run on PCT rails, and only about 14 of 33 LATAM countries are contracting states, with Argentina, Bolivia, Paraguay and Venezuela outside. The registry ships with a known coverage hole instead of a hidden one.
The evidence
What it unlocks
Published definitions and a dated, contributable record set, so a disagreement about Brazil’s company count becomes information, not two unfalsifiable numbers.
It has been done before
WIPO, the Patent Cooperation Treaty
The PCT did not harmonise patent law. It published one filing format and one search standard and let offices recognise each other’s work. In force since 1978, it now has 158 contracting states, and a single international application preserves rights across all of them. The sequence is the lesson for a regional data commons: publish the schema and the definitions first, recognition follows. It is also the limit. Roughly 14 of 33 LATAM countries are contracting states, so any IP layer built on these rails is partial by construction and should say so.
First moves
- 0–6 months
Publish the definition and the schema
Release the inclusion criteria, the mapping to the IDB’s twelve verticals and the company-level field list under an open licence, with the LADP and EMERGE counts side by side and the definitional difference written out. Deliverable: a versioned schema document and one reconciliation table for Brazil’s company count.
LADP coalition secretariat with EMERGE and university research groups - 6–18 months
Ship the funding catalogue
Normalise every open public funding program into funding type, ticket range, eligibility, deadline and decision time, refreshed monthly from agency sources with a public change log. Deliverable: a catalogue covering at least five countries, with a last-verified date on every record and no record older than the stated cadence.
National innovation agencies (FINEP and BNDES in Brazil, CORFO in Chile, ANII in Uruguay) with IDB Lab - 18–36 months
Open contribution with published diffs
Accept record submissions from accelerators, agencies and funds against the schema, publish every accepted and rejected change with its reason, and issue a quarterly reconciliation against at least one independent mapping. Deliverable: four consecutive quarterly diffs and a named maintainer per country.
LADP coalition secretariat with LAVCA and national accelerator networks
How we would know it worked
- Share of records carrying a populated source and last-verified date, published on the dashboard, above 90%
- Residual disagreement on Brazil’s company count between the commons and at least one independent mapping, reported as a number each quarter and falling
- Median staleness of funding-catalogue records, measured as days since last verification and published monthly, below 45
- Number of external contributors submitting accepted records per quarter from outside the secretariat, rising above ten
What could go wrong
A registry nobody updates
Open ecosystem databases decay within two cycles once the launch grant ends, and a stale funding catalogue is worse than none because founders act on closed calls and miss open ones. Containment: a stated refresh cadence with a last-verified date on every record, automatic expiry and visible flagging of records past that cadence, and a named per-country maintainer on a funded mandate rather than volunteer upkeep.
The base data cannot be republished
The mapping rests on a proprietary Tracxn extract that cannot be redistributed, so the open layer risks being aggregates computed over records nobody can inspect, which reproduces the exact problem it was built to fix. Containment: publish the schema, the definitions and the derivation code even where the underlying records stay licensed, and grow an openly licensed record set from contributor submissions so the open share is itself a reported number.
Who acts
Rests on
02 Metrics & Success Stories A deep tech KPI framework
16 Extended Funders Mapping A versioned registry the community can audit
The data behind it
Brazil hosts 41% of the region’s ventures, more than the next four countries combined. The “Big 5” account for 83% of everything mapped.
As published in the report, September 2025.
SourcesLatin American Dynamism Project (LADP), Accelerating Deep Tech in Latin America; Tracxn Technologies, Tracxn, LATAM deep tech company database (LADP extract); EMERGE (Emerge Brasil) & Cubo Itaú, Radar Deep Tech LATAM 2025
Methodology: LADP/Tracxn count and share. ledger #25/#26: EMERGE's independent mapping disagrees sharply (Brazil 72.3% vs 40.8%; different rank order). Neither is "the" number, a definitional gap, not an error.
| Category | Current |
|---|---|
| Brazil | 1,048 |
| Mexico | 358 |
| Argentina | 256 |
| Chile | 251 |
| Colombia | 219 |
| Rest of LATAM | 434 |