A procurement director at a German engineering firm needs a pump system integrator in Indonesia. She has three candidates. She will spend forty minutes on research before deciding which ones get an email.

Forty minutes. That is your entire window. And she is not starting from your website. She is starting from her own verification tools, databases, and search habits that were built for European and North American companies. Your company needs to be legible within that framework, or you do not exist.

I run three companies in Indonesia and have sat on both sides of this. I have been the company being evaluated, and I have done evaluations for partnerships across borders. The verification sequence is remarkably consistent regardless of the buyer's country. What changes is how much friction each step introduces when the target company is in Southeast Asia.

The verification sequence

International buyers do not verify randomly. There is a sequence, and each step either builds confidence or kills the inquiry. Understanding this sequence is the first step toward surviving it. The full pipeline is mapped in how to pass digital due diligence, but cross-border verification adds specific friction points that domestic deals never encounter.

graph TD A["Step 1: English-language Google search
Company name + country"] --> B["Step 2: International business database
Bureau van Dijk, D&B, Bloomberg"] B --> C["Step 3: Registry verification
Government company registry"] C --> D["Step 4: Structured data check
Schema, Knowledge Graph, Wikidata"] D --> E["Step 5: AI agent query
ChatGPT / Gemini / Perplexity"] E --> F["Step 6: Social proof scan
LinkedIn, press, industry presence"] F --> G["Decision: Contact or Next"] A -->|"No English results"| H["Eliminated"] B -->|"Not in database"| I["Flag: unverifiable"] C -->|"Registry in local language only"| J["Manual effort required"] style A fill:#222221,stroke:#c8a882,color:#ede9e3 style B fill:#222221,stroke:#c8a882,color:#ede9e3 style C fill:#222221,stroke:#c8a882,color:#ede9e3 style D fill:#222221,stroke:#6b8f71,color:#ede9e3 style E fill:#222221,stroke:#6b8f71,color:#ede9e3 style F fill:#222221,stroke:#c8a882,color:#ede9e3 style G fill:#222221,stroke:#6b8f71,color:#ede9e3 style H fill:#222221,stroke:#c47a5a,color:#c47a5a style I fill:#222221,stroke:#c47a5a,color:#c47a5a style J fill:#222221,stroke:#c47a5a,color:#c47a5a

Step 1: The English-language search

The first thing an international buyer does is type your company name into Google. In English. Not in Bahasa Indonesia. Not in your local language. English.

This immediately creates a problem for most Indonesian companies. Your website is in Indonesian. Your press coverage is in Indonesian. Your government filings are in Indonesian. The buyer sees one or two results, maybe your LinkedIn company page and your website homepage. Both with minimal English content. She cannot assess what you actually do.

Compare this to a Singaporean competitor. Full English website. English press mentions. English-language industry directory listings. The buyer can assess the Singaporean company in five minutes. Your company requires effort she may not be willing to invest.

The fix is not translating your entire website. It is having key pages in English: the homepage, the about page with company structure, the services page, and a capabilities document. These four pages handle 80% of what an international buyer needs in the first three minutes.

Step 2: International business databases

Most international procurement teams have access to Bureau van Dijk (Orbis), Dun & Bradstreet, or Bloomberg company databases. These are their primary tools. They type your company name, and they expect to see registration data, director names, financial indicators, and corporate structure.

For Indonesian companies, these databases are often incomplete. Data may be 12 to 24 months stale. The company name might be transliterated differently. The registration number might use the old format rather than the current OSS-based NIB. These are not your fault. But they are your problem.

Companies that actively maintain their profiles in these databases pass this step. Companies that assume the databases will find them automatically get flagged as "insufficient data." That flag rarely gets removed.

Practical step: check if your company exists in D&B and request a DUNS number if you do not have one. Update it annually. This costs nothing and takes one afternoon.

Step 3: Government registry verification

The buyer wants to confirm that your company is legally registered. For Indonesian companies, this means checking AHU (for company deed), OSS (for business licenses), or the DJP tax registration system.

Problem: none of these registries have English interfaces. The buyer cannot navigate them. She might ask a local agent or consultant to check, but that adds time and cost. If she has twenty companies on her long list, she will prioritize the ones that do not require extra effort.

This is where having your registration data on your own website, in English, with the actual registration numbers, becomes critical. Not just "We are registered with the Ministry of Law and Human Rights." But the actual AHU number, the NIB number, the NPWP. Declared on your website, verifiable against the registries for anyone who does navigate them.

Step 4: Structured data and machine readability

Sophisticated buyers and their verification tools now check for structured data. Does the company website have Organization schema? Does it declare its legal name, registration numbers, founding date, and officers in JSON-LD format? Does Google's Knowledge Graph recognize this company as a distinct entity?

For most Indonesian companies, the answer to all three is no. And this matters because structured data is the layer that connects your website to every other verification system. Without it, each verification step is isolated. With it, the verification chain becomes self-reinforcing.

A buyer's due diligence tool can crawl your website, extract your Organization schema, and automatically cross-reference it against registry data. That automation only works if the structured data exists. Without it, everything is manual, slow, and prone to the buyer giving up.

Step 5: AI verification

This step barely existed two years ago. Now it is standard. The buyer asks ChatGPT, Gemini, or Perplexity: "What do you know about [company name]?" or "Is [company name] a reliable pump system integrator in Indonesia?"

If the AI can provide accurate, detailed information about your company, including what you do, who runs it, what certifications you hold, and who your clients are, that is a strong verification signal. It means enough structured, corroborated data exists online for an AI to synthesize a coherent profile.

If the AI confuses you with another company, provides outdated information, or says it cannot find reliable data, that is a red flag. Not a dealbreaker necessarily, but another point of friction that accumulates.

Building the kind of entity infrastructure that makes AI verification possible is not a marketing exercise. It is a structural investment in how your company is represented across the entire digital information ecosystem.

What international buyers care about that domestic buyers do not

Verification pointDomestic buyerInternational buyer
Website languageLocal language fineEnglish required for key pages
Business database presenceRarely checkedD&B / Orbis is standard
Registration number on websiteNice to haveExpected, used for cross-referencing
Structured data (schema.org)Almost never checkedIncreasingly checked by tools
International press coverageNot relevantStrong signal if available
Wikidata entryNobody checksUsed by KG and AI systems
AI agent accuracyNot testedBecoming routine verification
Payment terms documentationDiscussed in personExpected on website or in proposals

The friction tax

Every step that requires extra effort from the buyer is a friction tax. And friction taxes compound. One step that requires manual research is fine. Two steps that require it, the buyer is getting impatient. Three steps, and she is already comparing you unfavorably to the Singaporean competitor whose information was available in five minutes.

International buyers are not biased against Indonesian companies specifically. They are biased against effort. They will choose the company whose verification is easiest, all else being equal. And "all else being equal" is often the case when comparing companies at the long-list stage, because the buyer does not know enough about any of them yet to differentiate on capability.

The companies that win cross-border contracts are not always the best. They are the ones that were verifiable within the buyer's existing process. That is a solvable problem.

Building for international verification

The minimum viable approach for any Indonesian company targeting international buyers:

English versions of four key pages: homepage, about (with company structure and registration numbers), services/capabilities, and contact. Not a full bilingual website. Just the four pages that handle first-contact verification.

A DUNS number and active profile in at least one international business database.

Organization schema in JSON-LD on your homepage, declaring legal name, registration numbers, founding date, address, and key personnel.

Consistent NAP (name, address, phone) data across your website, LinkedIn, Google Business Profile, and any directories you appear in.

A Wikidata entry with basic corporate data, linked to your website and any external references.

This is not a six-month project. It is a focused two-week sprint. And the return on investment for companies pursuing international contracts is disproportionately large, because almost nobody in the Indonesian market has done it yet. For a deeper look at how to approach the full pipeline, the entity infrastructure course material breaks it down step by step.

Frequently Asked Questions

Do international buyers really use AI to verify companies?

Increasingly, yes. A 2024 survey by OMMAX found that 38% of procurement teams in the DACH region reported using AI tools as part of their vendor research process. The usage is highest in technology and professional services procurement. For industrial sectors, it is growing but still secondary to traditional database checks. The point is not that AI is the primary tool. The point is that it is one more layer where your company either appears accurately or does not appear at all.

Is it worth paying for a Bureau van Dijk or D&B listing?

A DUNS number from D&B is free to obtain. Updating your profile is free. The paid premium listings provide additional features, but the free tier handles the core requirement: existing in the database with accurate data. For most Indonesian companies, the free DUNS number is sufficient. The critical thing is that the data is current and matches your website. Stale or inconsistent data is worse than no listing.

How much English content do I need on my website for international buyers?

Four pages minimum: homepage, about/company profile (with registration numbers and leadership), services or capabilities overview, and contact information. These four pages handle roughly 80% of what an international buyer looks for in the first five minutes. A full bilingual site is better, but the four-page minimum delivers most of the value at a fraction of the effort.

References

  1. OMMAX. "Digital Due Diligence: Transaction Advisory Services." OMMAX. Link
  2. Exporteers. "How to Enter the Asian Market: Due Diligence Guide." Exporteers. Link
  3. Dun & Bradstreet. "D-U-N-S Number: Company Verification." D&B. Link
  4. Roland Berger. "A Short Guide to Due Diligence of Digital-Oriented Acquisition Targets." Roland Berger Insights. Link

Related notes

2026-03-28

The companies that show up in ChatGPT are the ones that bothered to be verifiable.