Surveys can seem like an obvious tool to prove the reputation of a trade mark. But they can be hard to conduct to the satisfaction of a judge and easily become very expensive. As a result, their use depends very much on the established practice and attitude of the courts in a particular country. This Q&A guide, which first appeared in Managing IP and is reproduced here with permission, examines when surveys are worth the cost.
Does case law or any other set of guidelines set out how a survey should be conducted? What other advice would you have?
The Federal Court of Australia has issued a practice note in relation to survey evidence to assist in ensuring the admissibility of surveys relied upon in Federal Court proceedings. However, this practice note does not contain detailed guidelines regarding survey methodology or other matters. The Trade Marks Office has no guidelines regarding how a survey should be conducted.
Helpful points to note regarding the conduct of a survey are:
- The survey should be conducted professionally by an independent market research company and should include relevant, unambiguous, non-leading questions;
- The survey sample should be sufficiently large to be statistically significant (for example there were 1,500 respondents in a survey relied upon by Cadbury in support of its application for the colour purple in relation to chocolate);
- The respondents must be selected to be representative of a relevant cross-section of the public;
- Ideally, the survey should be conducted at the appropriate time; that is at the date of filing of the trade mark application if the survey is to be directed to the mark’s capacity to distinguish (although later survey evidence was recently accepted in Mars Australia’s application for the colour purple in relation to cat food); and
- The survey must be directed to the specific mark at issue, for example “canary” yellow for post-it notes, rather than the colour yellow per se, in Minnesota Mining & Manufacturing Company’s application.
How involved are the courts or IP office in controlling the structure of the survey? Has this changed recently?
The IP office and the courts are not actively involved in controlling the structure of the survey aside from the court enforcing compliance with the Federal Court practice note and any specific directions which may be given by a Judge.
Are surveys considered expert evidence by the courts?
In order for a survey to be admitted as evidence in a court proceeding, it is usual to have a marketing expert swear an affidavit exhibiting the survey, explaining the methodology and outlining conclusions that may be drawn from the results.
How much do surveys often cost? How big a factor is this for the decision to conduct a survey?
A professionally conducted survey can be very expensive (more than $100,000) depending upon the number of survey respondents, whether it is national, covers both urban and rural areas, and is conducted face to face (incorporating visual aids), online or by telephone. The complexity of the survey questions and methodology will also increase cost, though being part of an omnibus survey can reduce them.
There is the option of conducting an initial pilot survey involving a small number of respondents and at reduced cost (about $25,000-$30,000). In light of the difficulties in ensuring that survey evidence is not subject to challenge, parties to proceedings before the IP office may be reluctant to commit to the expense, while those involved in Federal Court proceedings may be more inclined to do so with a view to bolstering their prospects of success.
Overall, are surveys considered a valuable tool for trade mark disputes in your jurisdiction?
In theory, surveys may be very persuasive in establishing;
- that a trade mark is capable of distinguishing designated goods or services;
- that use of a trade mark is likely to deceive or cause confusion; or
- the deceptive similarity of competing marks.
The Australian Trade Marks Office Manual states that there is no absolute requirement that survey evidence be provided to support the distinctiveness of a mark. However, the manual further states that survey evidence may be useful for establishing the distinctiveness of non-traditional marks, such as shape and colour marks.
The judge in Chocolaterie Guylian v Registrar of Trade Marks stated that it has become common in trade mark registration appeals to rely on survey evidence. However, there has been judicial comment that market survey evidence or evidence of an independent marketing expert is not required to establish that a mark is sufficiently distinctive for registration. Statistics regarding the effectiveness of surveys in securing acceptance of trade mark applications before the IP office are not available. However, a review of recent trade marks disputes (including registrability disputes) that have been heard and determined by the IP office and the courts reveal a fairly low level of success, presumably on the basis that surveys appear to give rise to extensive challenge, scepticism and criticism.
Even fairly strong survey results may be disregarded or given little weight if the IP office or courts determine that the survey was conducted too late, did not relate to the specific mark at issue, had too small a sample or does not show that consumers understand the mark to be an indicator of the origin of the designated goods.