The Court of Appeal has held in Marks And Spencer Plc v Interflora Inc & Anor  EWCA Civ 1501 that, in trade mark infringement proceedings, evidence derived from statistically invalid and unreliable surveys (commissioned, for example, to show consumer confusion caused by alleged infringement) may in certain situations be admissible under CPR 32.1 as individual witness evidence, subject to the proviso that such evidence should only be admitted if it will be of value and its usefulness justifies the cost of introducing it.
In this case Interflora Inc (“Interflora”) alleged that Marks & Spencer Plc (“M&S”) had infringed certain trade marks by purchasing Google AdWords for the terms ‘Interflora’ and certain variants. By doing this, M&S caused Google searches for these terms to result in the appearance of a sponsored advert for its own flower delivery service, which was unrelated to Interflora’s business.
The European Court of Justice (“ECJ”) previously addressed the use of Google AdWords as a form of trade mark infringement in the case of Google France SARL v Louis Vuitton Malletier SA (C-236/08, C-237/08 and C-238/08)  ECR I-2417. There it held that AdWords could constitute the ‘use’ of a mark in the course of trade (a pre-requisite for infringement), however, such use will only be unlawful if it adversely affects a mark’s registered proprietor (for example: if the advert suggests an economic link between the AdWord’s sponsor and the proprietor; or else by confusing a ‘reasonably well informed and observant web user’ into believing the sponsored advert relates to the registered proprietor’s goods or services). This approach was affirmed by the ECJ on the facts of the present case (the ECJ’s judgment is M&S v Interflora (C-323/09)  Bus LR 1440).
Once Interflora had established that M&S purchased AdWords relating to its business, the next step was to gather evidence that these actions infringed its rights by causing confusion to web users. To this end, Interflora contracted a market research company to carry out two controlled surveys designed to ascertain whether M&S’s AdWords caused the necessary confusion to a group of selected individuals. At trial Interflora accepted that these surveys were statistically unreliable and flawed and did not seek to rely on them, however it did seek to admit evidence from individual witnesses who had participated in the surveys.
The trial court held that the witnesses’ evidence could be admissible and might be probative even though the surveys themselves were not reliable and contained leading questions. At appeal M&S challenged this ruling, arguing that because the ‘reasonably well informed and observant web user’ was not an actual person, it was inappropriate to hear evidence from individual witnesses unless that witness was acting as a proxy for an admissible survey.
The Court of Appeal accepted that the test for establishing infringement could only be answered by the judge making general observations, and that such matters were ‘not a question of counting heads, but of qualitative assessment’. Furthermore, the fact some persons might suffer confusion was not determinative of the affect that an AdWord would have on the reasonably well informed and observant web user. Instead, the reasonable web user was the internet equivalent of the ‘average consumer’ by whose standards trade mark infringement is judged. On this basis and on the view that AdWords constitute an ‘ordinary consumer service’, the Court held that whether a web user would be confused was a matter for the judge, which could usually be decided without the need for further evidence from experts or actual consumers. The correctness of this approach was reinforced by an examination of previous cases on trade mark infringement.
The Court then examined the nature of evidence derived from surveys which are not statistically valid and reliable, deciding that in submitting evidence derived from these surveys a witness was not standing proxy for a sample of a valid survey, but was instead providing evidence as an individual. It accepted that there were certain situations (such as where goods are not supplied to ordinary consumers and are unlikely to be within the judge’s own knowledge) where a court may wish to hear of informed consumer habits; however as a general rule such evidence was unlikely to be of real use. Nonetheless, there might still be other cases where assistance is required, and therefore such evidence should not be inadmissible as a matter of law. Instead it was up to the Court under CPR 32.1 to use its discretion to decide whether or not the evidence should be admitted. This marks a shift from the previous approach which was to allow admissible evidence unless it was proven to be of no value. Now courts will, even if admissible, only allow evidence if it is: (i) of value; and (ii) its use outweighs the costs of introducing it.
On the facts, the Court of Appeal held that Interflora had failed to demonstrate that the survey evidence was of real value and therefore M&S’s appeal was allowed.
White & Black Legal’s comment: the Court of Appeal’s judgment provides parties with a new test for the admissibility of survey evidence and consequently some much needed clarity. This is an important judgment for future cases where survey evidence is involved and will likely be of particular relevance to actions of alleged trade mark infringement.