What a Market Survey Cannot Tell You

A survey arrives in a non-use case the way a photograph arrives in a boundary dispute: it looks decisive, and it answers a question nobody asked. The report is thick, the methodology section is confident, and the number at the end is large or small depending on who commissioned it. What it almost never establishes is the fact the court actually has to find.

Article 74 of Law No. 20 of 2016 allows an interested third party to sue in the Commercial Court to delete a registration on the ground that the mark has not been used in trade in the goods or services for which it is registered. Constitutional Court Decision No. 144/PUU-XXI/2023 extended that window from three years to five and recognised force majeure as an excuse for an interruption. The provision was not otherwise altered. The question it poses is unchanged: was this mark used in trade?

A consumer survey asks something else entirely: do people remember this mark? Those two questions have different answers, and the difference is not academic.

A survey is not an item of evidence

Indonesian civil procedure is explicit about what counts. Article 164 HIR, mirrored in Article 1866 of the Civil Code, lists five: written evidence, witness testimony, presumption, admission, and oath. A market survey is not among them, and no provision adds it.

So a survey never enters a case in its own right. It enters as a written document — a report produced by a party — or through an expert who speaks to it. In the first form it is a private document made by someone the party selected and paid. In the second its weight depends on the expert, not the instrument. Either way the court weighs it freely. It carries no presumption, binds nobody, and displaces no other evidence.

What recognition actually measures

Set aside the law for a moment and ask what a recognition survey is physically capable of detecting. It samples a population and records how many recall a mark. That recall is produced by exposure. Exposure is produced by advertising reach, media spend, distribution breadth and time in market.

Every one of those is a function of capital. None of them is trade.

The two can diverge completely, and in Indonesia they routinely do:

• A foreign proprietor runs a global campaign, registers in Indonesia, and sells nothing here for six years. Recognition in an urban sample is high. Use in trade is nil.
• A producer in Garut sells under its mark every working day for a decade through local distribution. Recognition in a national sample is close to zero. Use in trade is continuous, documented and real.

If recognition is treated as the proxy for use, the first registration survives and the second falls. That inverts the provision. Article 1(1) of the Act defines a mark as a sign used to distinguish goods or services in trading activities. Trade is the statutory concept. Fame is not.

Where surveys do belong

None of this makes survey evidence worthless. It makes it evidence of a different fact.

When the question is whether a mark is well known in Indonesia, reputation is the issue, and the knowledge of the relevant sector is one of the criteria applied in practice. There a properly constructed survey is directly probative, and a proprietor who has not built that record before the dispute will struggle.

The error is not using surveys. The error is using them to answer Article 74.

What proves use in trade

Evidence of use is documentary, and it is the kind of material a trading business generates without trying:

• Invoices and receipts tied to the goods or services in the class as registered
• Tax records, including VAT filings that correspond to those sales
• Customs and shipping documents for imported or exported goods
• Distribution, supply and licensing agreements, with evidence that they were performed
• Marketplace transaction histories, which carry dates, quantities and counterparties
• Product registrations and circulation permits where the goods require them
• Packaging, labels and catalogues showing the mark as registered, not as varied

Note what this list does not require: a budget. A small enterprise that genuinely trades has all of it. A large advertiser that does not trade in Indonesia has none of it, however well its mark is recalled.

Challenging a survey on the other side

Where an opponent relies on a survey, the attack is not rhetorical. It is methodological, and it is specific:

• The universe. Who was defined as the relevant public, and why? A survey of the general urban population says nothing about a mark registered for industrial fittings.
• The sampling frame. How were respondents reached? Online panels over-represent the connected and the urban.
• Aided versus unaided recall. Showing the mark and asking "have you seen this?" produces a far higher number than asking what marks the respondent knows in the category. The two are often reported as if interchangeable.
• Question wording. Leading formulations, and questions that supply the answer in the stem.
• Independence. Who commissioned, designed and paid for the instrument, and was the design fixed before the results were seen?
• The inferential leap. Even a flawless survey measures recall. Ask on the record what it is said to prove about trade.

That last point is usually the shortest route. A survey rarely fails because its statistics are wrong. It fails because the conclusion drawn from it does not follow.

Our position

Use in trade is a requirement, not an obstacle. A registration held without trade is a reservation, and the Act does not create a right to reserve. We say that plainly, and it binds our own clients as much as anyone else's.

What we contest is the standard of proof. When use is tested by an instrument that measures advertising reach, the test stops asking whether a business traded and starts asking whether it could afford to be remembered. The recitals to Law No. 20 of 2016 name the protection of micro, small and medium enterprises among the purposes of the Act. A standard only a large advertiser can satisfy defeats that purpose from inside the statute that declares it.

The dividing line is use, not size. A foreign proprietor who genuinely trades in Indonesia stands on the same side of it as a producer in Garut. The only party on the other side is one who registered and never traded.

Primary sources

• Law No. 20 of 2016 on Trademarks and Geographical Indications — Articles 1(1), 35, 74, 76, 77 (accessed 5 October 2026)
• Constitutional Court Decision No. 144/PUU-XXI/2023, pronounced 30 July 2024
• Article 164 Herziene Indonesisch Reglement (HIR) and Article 1866 of the Indonesian Civil Code, on the admissible forms of evidence in civil proceedings (accessed 5 October 2026)
• Minister of Law Regulation No. 5 of 2026 on Trademark Registration, in force 22 February 2026, which revoked Minister of Law and Human Rights Regulation No. 67 of 2016