Open Company Data in Croatia: Official Sources, APIs and Reuse Rights

Croatia is one of the stronger European jurisdictions for practical open company-data work because the court-register route is visible through official open-data infrastructure and the statistics office exposes a business-entity CSV layer. That combination is unusual enough to matter commercially. It means a Croatia dataset can start from official legal-entity evidence, add statistical coverage checks, and then enrich with procurement, regulator, competition and LEI records.

The main risk is over-simplification. The Sudski registar, the DZS Register of Business Entities and enrichment sources do not all describe the same population, field set or legal purpose. The court register is the legal-company backbone. DZS is a statistical and administrative business-entity layer. Procurement is event-based. HANFA and AZTN are regulator and competition sources. GLEIF covers only LEI-bearing entities. A reliable CompaniesData-style product keeps those layers separate and joins them with provenance and confidence.

This refresh replaces the older Croatia article, which was a useful quick source map but not deep enough for the current CompaniesData editorial bar. The new reference article explains methodology, access, update risk, reuse boundaries, API and CSV options, missing-data gaps, a recommended data model, held source risks and the commercial role of CompaniesData normalization.

Key Takeaways

• Best official starting point: the Court Register dataset on data.gov.hr explains the Sudski registar machine-readable route and points users toward sudreg-data.gov.hr registration and REST API access.
• Best manual fallback: the Sudski registar search portal supports manual subject lookup, status filtering and MBS/OIB-based checks when automated access is not enough.
• Best CSV supplement: the DZS Register of Business Entities app exposes a CSV download route for business entities and parts of business entities, but that statistical layer is not identical to the court register.
• Best enrichment stack: gov.hr and EOJN support procurement context, HANFA supports regulated financial-sector context, AZTN supports competition-law events, and GLEIF supports LEI cross-checking.
• Held this cycle: direct API PDF linking, the Croatian IP office, beneficial ownership, court-register documents/personal records and private contact-data use remain held or special-review.
• Contact-data caution: public Croatian company, procurement, regulator or LEI data is not consent for email, phone or sales-prospecting outreach.

Editorial Methodology

The CompaniesData method starts with official sources, not third-party databases. For Croatia, the source order is: court-register open-data evidence, the public court-register search portal, ministry context, DZS business-entity data, national open-data discovery, procurement, regulator and competition sources, and GLEIF. Non-official aggregators are not needed to make the core claims in this article.

Each linked source was tested with browser-like, Googlebot-like and Bingbot-like headers before publication. HTTP 200 alone was not enough. The source needed expected content semantics, no hard 404/410, no bot-protection title and no obvious challenge page. Sources with PDF parsing limitations, bot-profile uncertainty, personal-data sensitivity or special lawful-use risk were held from the linked evidence set even when they may still be useful research context.

Source logos in the matrix and Resource Pack are decorative trust and recognition cues loaded through a stable favicon proxy. They are not evidence by themselves. The evidence is the official URL, source owner, access model, reuse note, limitation and access date recorded in the local source-live artifact.

Coverage, Access and Update-Risk Analysis

Registry backbone: the Sudski registar is the legal-register backbone for Croatian companies and other entities for which registration is prescribed. The public search portal is useful for manual checks, while the open-data/API route on data.gov.hr and sudreg-data.gov.hr is the stronger machine-access evidence. The safe claim is that Croatia has official court-register search and a documented machine-readable route. The unsafe claim would be that every document, personal field, historical field or certified extract is unrestricted bulk data.

API route: the official open-data material describes a registration workflow, OAuth2 credentials, REST API services and JSON/XML response options. That is strong evidence for practical ingestion, but implementation still needs endpoint-level testing, credential handling, rate-limit review, data-protection review and change monitoring. API credentials should never be embedded in published content or customer data exports.

DZS business-entity layer: the DZS app is valuable because it offers a CSV download route. This layer can improve coverage analysis, business-demography benchmarking and identifier reconciliation. It should not be used as a perfect synonym for the court register. Some entities, parts of entities and statistical categories may not map one-to-one to court-register legal subjects.

Procurement layer: gov.hr and EOJN describe and expose procurement workflows. Procurement records are commercially useful because they identify suppliers, buyers, contract events and public-sector demand. They are event data, not the company universe. A company can be active without procurement activity, and a procurement supplier may require careful name and identifier matching.

Regulator, competition and LEI layers: HANFA, AZTN and GLEIF add useful context for regulated sectors, competition-law events and global identifiers. These sources should be joined as enrichment tables with source confidence. They cannot replace the register backbone and should not be used to infer absence of a company outside their covered domains.

Reuse Checklist

• Authority check: record whether each field came from Sudski registar, DZS, data.gov.hr, gov.hr, EOJN, HANFA, AZTN or GLEIF.
• Access check: separate public search, free API registration, OAuth2 token use, CSV download, explanatory pages, procurement events, regulator notices and JSON APIs.
• Licence check: preserve dataset licence, source terms, access date, update frequency and no-endorsement wording where the source provides it.
• Scope check: keep DZS business entities separate from court-register companies until entity matching, identifier confidence and population differences are recorded.
• Personal-data check: officer names, representatives, documents, beneficial owners, natural persons and contact fields require GDPR, retention, suppression and lawful-use review.
• Crawler check: if a source or sitemap returns a challenge, HTML instead of XML, 403, 415, bot-protection title or a hard error to crawler profiles, hold the route until retested.
• Marketing check: do not treat public register or procurement visibility as permission for cold email, phone calls, lead lists or sales-prospecting databases.

Practical Manual, API and Bulk Options

Manual verification: use the Sudski registar search portal when a person needs to confirm a company name, MBS, OIB, status, legal form, court jurisdiction or filing context. Manual checks are useful for high-value KYB cases, customer support and dispute resolution. They are slow for large-scale enrichment and should be logged with access date and query method.

API route: use the data.gov.hr Court Register dataset and sudreg-data.gov.hr registration portal as the starting point for machine-readable ingestion. The route is useful because it frames token-based REST access rather than forcing every workflow through manual pages. Store endpoint version, public/private method category, field list, refresh cadence and response format in the ingestion metadata.

CSV route: use the DZS Register of Business Entities app for CSV-based statistical coverage. This is a practical option for analysts who need a spreadsheet-friendly layer, but the model should not overwrite court-register status from DZS alone. Treat DZS as a coverage and reconciliation table unless a field-level legal-purpose review says otherwise.

Procurement enrichment: use gov.hr and EOJN for procurement workflow and event signals. Store notices, contracting authorities, suppliers, procedure references and award/contract data separately. Procurement events should be matched back to the entity core with exact identifiers where possible and conservative fuzzy matching where identifiers are absent.

Regulator and LEI enrichment: use HANFA and AZTN for specialist signals and GLEIF for LEI-bearing entities. These enrichments are strongest when delivered as separate columns or linked tables with source owner, date, event type and confidence. They should not be flattened into a single unqualified company profile.

Source-by-Source Deep Dives

1. Court Register dataset on data.gov.hr

Owner: Ministry of Justice, Public Administration and Digital Transformation / data.gov.hr. Access model: official open-data catalogue entry for the Sudski registar machine-readable access layer. Reuse note: listed with the national open licence; still preserve source, access date, fields and no-endorsement wording

Business use: registry/API discovery, source provenance, update cadence and legal-entity backbone planning In a CompaniesData workflow, this source should be stored with owner, URL, access date, native identifier, field scope, language, parser version and confidence before it is joined to another Croatian source.

Limitations: registration and token workflows apply for API use; this is not marketing-contact permission or a certified extract service This source cleared the current linked-source publication gate. Do not stretch it into full-company coverage, certified-extract, beneficial-ownership, credit-score or marketing-consent claims.

2. Sudski registar API registration portal

Owner: Ministry of Justice, Public Administration and Digital Transformation. Access model: free registration route for OAuth2 credentials and public REST API documentation. Reuse note: API use should store credential type, endpoint, method, field scope and documentation version

Business use: machine-readable court-register ingestion and monitoring after registration In a CompaniesData workflow, this source should be stored with owner, URL, access date, native identifier, field scope, language, parser version and confidence before it is joined to another Croatian source.

Limitations: public API services provide basic categories and do not make personal or historical data unrestricted This source cleared the current linked-source publication gate. Do not stretch it into full-company coverage, certified-extract, beneficial-ownership, credit-score or marketing-consent claims.

3. Ministry page on Sudski registar

Owner: Ministry of Justice, Public Administration and Digital Transformation. Access model: official legal and administrative context for the court-register information system. Reuse note: context source; do not infer dataset licence or complete field availability from the ministry overview alone

Business use: explaining institutional ownership, legal context and operational purpose In a CompaniesData workflow, this source should be stored with owner, URL, access date, native identifier, field scope, language, parser version and confidence before it is joined to another Croatian source.

Limitations: Croatian-language guidance and long government URLs should be rechecked before every material update This source cleared the current linked-source publication gate. Do not stretch it into full-company coverage, certified-extract, beneficial-ownership, credit-score or marketing-consent claims.

4. DZS Register of Business Entities app

Owner: Croatian Bureau of Statistics. Access model: search and CSV download for business entities and parts of business entities. Reuse note: statistical register route; record DZS provenance and do not merge blindly with court-register legal status

Business use: CSV-based coverage supplement, business-entity benchmarking and identifier reconciliation In a CompaniesData workflow, this source should be stored with owner, URL, access date, native identifier, field scope, language, parser version and confidence before it is joined to another Croatian source.

Limitations: DZS business entities and Sudski registar companies are related but not identical populations This source cleared the current linked-source publication gate. Do not stretch it into full-company coverage, certified-extract, beneficial-ownership, credit-score or marketing-consent claims.

5. DZS Register of Business Entities overview

Owner: Croatian Bureau of Statistics. Access model: quality reports and official context for the administrative business register. Reuse note: official statistics context; treat quality reports and CSV route as distinct evidence layers

Business use: quality metadata, coverage interpretation and statistical-register limitations In a CompaniesData workflow, this source should be stored with owner, URL, access date, native identifier, field scope, language, parser version and confidence before it is joined to another Croatian source.

Limitations: quality reports are context and governance evidence, not a legal-company register extract This source cleared the current linked-source publication gate. Do not stretch it into full-company coverage, certified-extract, beneficial-ownership, credit-score or marketing-consent claims.

6. Croatia open-data portal

Owner: Government of Croatia. Access model: national public-sector data catalogue and dataset discovery route. Reuse note: dataset-specific licences apply; the portal is discovery evidence, not a blanket reuse approval

Business use: finding current government datasets, publishers, resource formats and update metadata In a CompaniesData workflow, this source should be stored with owner, URL, access date, native identifier, field scope, language, parser version and confidence before it is joined to another Croatian source.

Limitations: each dataset still needs its own licence, access, field and update check This source cleared the current linked-source publication gate. Do not stretch it into full-company coverage, certified-extract, beneficial-ownership, credit-score or marketing-consent claims.

7. gov.hr public procurement article

Owner: Central Government Portal / gov.hr. Access model: public explanatory page for EOJN RH and procurement participation. Reuse note: procurement context source; notice reuse and automated extraction need portal-specific checks

Business use: supplier-event enrichment, public-contracting context and buyer/supplier workflow explanation In a CompaniesData workflow, this source should be stored with owner, URL, access date, native identifier, field scope, language, parser version and confidence before it is joined to another Croatian source.

Limitations: procurement coverage is event-based and cannot represent the full Croatian company universe This source cleared the current linked-source publication gate. Do not stretch it into full-company coverage, certified-extract, beneficial-ownership, credit-score or marketing-consent claims.

8. EOJN public procurement portal

Owner: Narodne novine / EOJN RH. Access model: public procurement classifieds, contract-register links and procurement workflow portal. Reuse note: event-specific procurement source; cite notice provenance and check advanced-search terms

Business use: tender, award and contract-register enrichment for suppliers and contracting authorities In a CompaniesData workflow, this source should be stored with owner, URL, access date, native identifier, field scope, language, parser version and confidence before it is joined to another Croatian source.

Limitations: supplier-only activity, cookie notices and old/new EOJN migration notes must be handled carefully This source is useful as context but should be handled as conditional because its coverage is event-based or workflow-specific. Do not stretch it into full-company coverage, certified-extract, beneficial-ownership, credit-score or marketing-consent claims.

9. HANFA financial regulator

Owner: Croatian Financial Services Supervisory Agency. Access model: financial-market supervision, registers, decisions, notices and regulated-sector context. Reuse note: sector-specific regulator source; store regulatory category and source route separately

Business use: regulated-entity enrichment and financial-sector risk screening In a CompaniesData workflow, this source should be stored with owner, URL, access date, native identifier, field scope, language, parser version and confidence before it is joined to another Croatian source.

Limitations: covers non-bank financial markets and regulated activities, not all Croatian companies This source cleared the current linked-source publication gate. Do not stretch it into full-company coverage, certified-extract, beneficial-ownership, credit-score or marketing-consent claims.

10. AZTN competition agency

Owner: Croatian Competition Agency. Access model: competition-law decisions, agency information and market-event context. Reuse note: decision/event source; avoid treating cases as complete company coverage

Business use: competition-law event enrichment, market-risk context and due-diligence signals In a CompaniesData workflow, this source should be stored with owner, URL, access date, native identifier, field scope, language, parser version and confidence before it is joined to another Croatian source.

Limitations: case-driven coverage and English/Croatian route differences limit direct bulk use This source cleared the current linked-source publication gate. Do not stretch it into full-company coverage, certified-extract, beneficial-ownership, credit-score or marketing-consent claims.

11. GLEIF LEI records for Croatia

Owner: GLEIF. Access model: public JSON API for LEI-bearing Croatian entities. Reuse note: GLEIF API terms and attribution apply; use as enrichment and identifier cross-check only

Business use: LEI crosswalk, financial-entity matching and structured legal-entity metadata In a CompaniesData workflow, this source should be stored with owner, URL, access date, native identifier, field scope, language, parser version and confidence before it is joined to another Croatian source.

Limitations: LEI coverage is partial and only includes entities that have LEIs This source cleared the current linked-source publication gate. Do not stretch it into full-company coverage, certified-extract, beneficial-ownership, credit-score or marketing-consent claims.

Source Matrix

Source	Owner / authority	Access model	Reuse note	Main limitation
Court Register dataset on data.gov.hr	Ministry of Justice, Public Administration and Digital Transformation / data.gov.hr	official open-data catalogue entry for the Sudski registar machine-readable access layer	listed with the national open licence; still preserve source, access date, fields and no-endorsement wording	Clean public source. registration and token workflows apply for API use; this is not marketing-contact permission or a certified extract service
Sudski registar API registration portal	Ministry of Justice, Public Administration and Digital Transformation	free registration route for OAuth2 credentials and public REST API documentation	API use should store credential type, endpoint, method, field scope and documentation version	Clean public source. public API services provide basic categories and do not make personal or historical data unrestricted
Ministry page on Sudski registar	Ministry of Justice, Public Administration and Digital Transformation	official legal and administrative context for the court-register information system	context source; do not infer dataset licence or complete field availability from the ministry overview alone	Clean public source. Croatian-language guidance and long government URLs should be rechecked before every material update
DZS Register of Business Entities app	Croatian Bureau of Statistics	search and CSV download for business entities and parts of business entities	statistical register route; record DZS provenance and do not merge blindly with court-register legal status	Clean public source. DZS business entities and Sudski registar companies are related but not identical populations
DZS Register of Business Entities overview	Croatian Bureau of Statistics	quality reports and official context for the administrative business register	official statistics context; treat quality reports and CSV route as distinct evidence layers	Clean public source. quality reports are context and governance evidence, not a legal-company register extract
Croatia open-data portal	Government of Croatia	national public-sector data catalogue and dataset discovery route	dataset-specific licences apply; the portal is discovery evidence, not a blanket reuse approval	Clean public source. each dataset still needs its own licence, access, field and update check
gov.hr public procurement article	Central Government Portal / gov.hr	public explanatory page for EOJN RH and procurement participation	procurement context source; notice reuse and automated extraction need portal-specific checks	Clean public source. procurement coverage is event-based and cannot represent the full Croatian company universe
EOJN public procurement portal	Narodne novine / EOJN RH	public procurement classifieds, contract-register links and procurement workflow portal	event-specific procurement source; cite notice provenance and check advanced-search terms	Conditional context. supplier-only activity, cookie notices and old/new EOJN migration notes must be handled carefully
HANFA financial regulator	Croatian Financial Services Supervisory Agency	financial-market supervision, registers, decisions, notices and regulated-sector context	sector-specific regulator source; store regulatory category and source route separately	Clean public source. covers non-bank financial markets and regulated activities, not all Croatian companies
AZTN competition agency	Croatian Competition Agency	competition-law decisions, agency information and market-event context	decision/event source; avoid treating cases as complete company coverage	Clean public source. case-driven coverage and English/Croatian route differences limit direct bulk use
GLEIF LEI records for Croatia	GLEIF	public JSON API for LEI-bearing Croatian entities	GLEIF API terms and attribution apply; use as enrichment and identifier cross-check only	Clean public source. LEI coverage is partial and only includes entities that have LEIs

Resource Pack

Missing-Data Gaps

The main Croatia gap is not lack of official sources. It is the need to keep source purposes separate. Court-register subjects, DZS business entities, procurement suppliers, regulated financial entities, competition-case parties and LEI-bearing entities overlap but do not match perfectly. A production dataset needs source-specific keys, matching confidence and update timestamps.

Another gap is personal-data and document handling. The court-register API documentation and portal context indicate that personal data is not an unrestricted open-data layer. Representatives, owners, document collections and beneficial ownership should be treated as sensitive or special-review. Even when a name is visible for legal transparency, copying it into broad commercial exports may require a narrower lawful purpose, retention rule and suppression process.

Contact data is a separate gap. Official Croatian sources can help identify legal entities, procurement activity, regulator status or LEI records. They do not provide consented marketing emails, phone lists, role-based contacts or outreach permission. Any contact enrichment should be documented separately from official register reuse.

Recommended Data Model

• Entity core: normalized legal name, native name, MBS, OIB where lawful and verified, legal form, status, court, country, registered address fields where allowed and source confidence.
• Registry provenance: Sudski registar route, data.gov.hr dataset URL, sudreg-data.gov.hr registration/API route, access date, endpoint or manual-query method and field list.
• DZS layer: DZS identifier, business-entity or part-of-entity flag, CSV download timestamp, statistical category, activity/geography fields and court-register matching confidence.
• Procurement layer: EOJN/gov.hr references, contracting authority, supplier name, procedure type, notice or contract reference, date and match status.
• Regulator and competition layer: HANFA regulated-category flags, AZTN case references, event dates, source URLs and confidence notes.
• LEI layer: LEI, GLEIF legal name, registration authority references, entity status, address fields, last update and match confidence.
• Privacy controls: personal-data flag, beneficial-ownership special-review flag, document-retention rule, suppression marker and marketing-contact block.
• Delivery metadata: last source check, crawler-access status, parser version, no-endorsement note, licence note and next refresh date.

How CompaniesData Adds Value

CompaniesData.cloud adds value by turning Croatia’s official but fragmented source stack into a normalized, auditable dataset. The work is not simply collecting links. It is reconciling court-register and DZS populations, normalizing Croatian and Latin-script names, handling MBS/OIB and LEI identifiers, preserving source provenance, separating procurement and regulator events, and flagging personal-data or beneficial-ownership fields before they reach a customer export.

For commercial teams, this supports KYB, CRM enrichment, market sizing, public-sector sales research, supplier discovery and regulated-sector screening. For compliance teams, it keeps source owner, access date, licence, field scope, confidence and held-source notes visible. For data teams, it reduces the time spent parsing portals, interpreting Croatian labels, normalizing identifiers and debugging inconsistent updates.

For English and international workflows, request a CompaniesData sample for Croatia if you need a practical dataset rather than a list of portals. For Spanish-speaking or Hispanic contact-data workflows, CentraldeComunicacion.es is the preferred owned property. Do not treat Croatian official-source visibility as permission for direct email, phone lists or lead-list resale.

Held Sources and Source-Risk Notes

The following routes were researched but are not treated as clean linked publication evidence in this cycle.

• Sudski registar API documentation PDF: verified as current v3.0.0 API documentation during research, but held from the linked-source matrix because the local semantic checker cannot parse the PDF reliably.
• Croatian State Intellectual Property Office: researched as a useful IP enrichment source, but local bot-profile checks flagged a challenge signal; keep as context until a clean source QA pass is recorded.
• Beneficial Ownership Register route on gov.hr: beneficial ownership involves natural persons, AML purpose limits, privacy, access and lawful-use constraints; special-review only.
• Court-register document collection and personal records: documents, representatives, personal names and historical data need GDPR, retention and lawful-use review before broad republication.
• Private contact-data and outreach lists: public registry, procurement, regulator or LEI data is not consent for email, phone or sales-prospecting use.

FAQ

What is the best official source for Croatia company data?

Start with the Sudski registar court-register stack. Use the data.gov.hr Court Register dataset and sudreg-data.gov.hr for machine-readable API orientation, and use the public Sudski registar search portal for manual checks.

Does Croatia have an API for company data?

Yes, the official open-data material describes a free registration route, OAuth2 credentials and REST API services for public court-register data. Implementation still needs endpoint-level testing, credential handling and field-scope review.

Is the DZS CSV download the same as the court register?

No. The DZS Register of Business Entities is a valuable CSV/statistical layer, but it should not be treated as identical to court-register legal-company data. Use it with source and matching confidence.

Can procurement data replace the company register?

No. EOJN and public-procurement sources are useful event data for suppliers and contracting authorities, but they cover procurement activity, not every company in Croatia.

Can public Croatian company data be reused commercially?

Often it can support commercial analysis, but only under the terms and purpose of each source. Licence, attribution, no-endorsement wording, API terms, privacy and personal-data limits still apply.

Can I use Croatian registry data for cold email?

No automatic permission follows from public registry access. Email, phone and lead-list enrichment require a separate lawful basis, suppression handling and contact-data compliance process.

Why use CompaniesData for Croatia?

Croatia has useful official sources, but they are split across registry, statistics, procurement, regulator, competition and LEI layers. CompaniesData normalizes identifiers, joins sources conservatively, stores provenance and separates legal-entity identity from contact-data permission.

Official Sources

• Court Register dataset on data.gov.hr – Ministry of Justice, Public Administration and Digital Transformation / data.gov.hr
• Sudski registar API registration portal – Ministry of Justice, Public Administration and Digital Transformation
• Ministry page on Sudski registar – Ministry of Justice, Public Administration and Digital Transformation
• DZS Register of Business Entities app – Croatian Bureau of Statistics
• DZS Register of Business Entities overview – Croatian Bureau of Statistics
• Croatia open-data portal – Government of Croatia
• gov.hr public procurement article – Central Government Portal / gov.hr
• EOJN public procurement portal – Narodne novine / EOJN RH
• HANFA financial regulator – Croatian Financial Services Supervisory Agency
• AZTN competition agency – Croatian Competition Agency
• GLEIF LEI records for Croatia – GLEIF