Fage UK Ltd & Anor v Chobani UK Ltd & Anor

Survey evidence in passing-off proceedings should be admitted at an interlocutory stage only where it is likely to be of real value and its likely utility justifies the cost. The court must assess the survey’s construction, implementation, reliability and proposed questions, applying a cost-benefit analysis. A survey need not be perfect, but it must be sufficiently reliable to provide useful evidence and avoid producing a skewed collection of witnesses. In passing-off cases, a properly designed survey may assist in determining whether a substantial proportion of the relevant public recognises a sign or expression as indicating the claimant’s goods or their origin.

The claim concerned extended passing off arising from the defendants’ proposed sale in the United Kingdom of yoghurt described as Greek yoghurt. The claimants contended that the expression denoted yoghurt made in Greece from Greek milk. The defendants contended that it described a product with a characteristic taste and consistency, irrespective of origin.

The defendants sought permission to conduct a full survey after completing a pilot survey, and to call evidence from survey respondents. The claimants challenged the survey’s relevance, methodology, questions and reliability. The central issue was whether the proposed evidence was likely to be sufficiently valuable to justify its costs and admission at trial.

1. The defendants were permitted to proceed with a full survey, at their own risk as to costs. The application to adduce evidence from survey respondents was stood over. The court reserved reconsideration of whether the survey results, and any respondent evidence, would ultimately be admitted.

2. Following Marks and Spencer PLC v Interflora Inc and another [2012] EWCA Civ 1501, the court should decide survey admissibility at an interlocutory stage. Permission should be given only if the evidence is likely to be of real value and its likely utility justifies the costs. The court must evaluate the material available, including the survey’s reliability, and conduct a cost-benefit analysis.

3. A survey may be relevant where it addresses a pleaded issue. Here, a properly designed survey could assist on the first hurdle in an extended passing-off claim: whether a substantial proportion of relevant consumers regard “Greek yoghurt” as indicating that the product was made in Greece.

4. The fact that the survey was imperfect did not necessarily make it valueless. A representative sample of 1,000 people could be sufficient if it was calculated to represent a relevant cross-section of the public and respondents were approached and questioned consistently with the Whitford guidelines derived from Imperial Group plc v Philip Morris Ltd [1984] RPC 293.

5. The pilot survey’s single location, the wording of its first question and other methodological criticisms materially affected reliability. Expert input would probably improve the questionnaire. However, in the context of the tight timetable, the pilot results suggested that a full survey might assist the court, and the balance favoured permitting it subject to later reconsideration.

6. Had there been more time, the court would have been minded to decline the survey and require a better version. The defendants had liberty to apply for approval of modifications, and the parties had liberty to apply generally.