In a recent landmark decision in the dispute between Interflora and Marks & Spencer, the Court of Appeal held that, in the majority of trade mark infringement cases, there is no real value in the evidence of witnesses from the public and that the court should not allow such evidence to be presented. The decision introduces a major overhaul in the way trade mark infringement cases are conducted in the UK and may help to reduce the costs of trade mark enforcement.

Trade mark cases often require the Court to determine how the use of a trade mark in the marketplace affects consumers’ perceptions and purchasing decisions. It is common for parties in litigation to present survey evidence to support their case on questions such as likelihood of confusion or the degree of reputation that a mark acquired or similar issues.

In the English Courts, over the years, it has become standard practice in such cases (either instead of using survey evidence or in addition to it) to present to the Court the witness evidence of members of the public, selected in the process of conducting a survey. The argument for the use of that type of evidence is that, while the selected witnesses may not necessarily be representative of the relevant section of the public who are potential customers for the relevant goods or services, there is still value in the testimony of ‘real’ people who are present at Court and who can be cross-examined, as opposed to relying solely on an anonymous consumer poll presented through the medium of a professional expert witness.


What Interflora had to prove

In the case at hand, Interflora, the flowers delivery network, accuses M&S - one of the UK’s biggest retailers - of infringing its registered mark INTERFLORA by using the mark (and close variants of it) as Google Adwords. When an internet user typed in a Google search using the word INTERFLORA, or one of its variants, a link to the M&S website (where an order could be placed for the delivery of flowers) would come up as one of the ‘sponsored links’ in the search results. That was the alleged infringement.

The test for infringement in these kind of cases was determined by the Court of Justice of the European Union (the “CJEU”) in previous proceedings in connection with the Interflora v. M&S case itself and in a few earlier cases.

The CJEU held that an infringement occurs in such cases if the use of the Google Adword adversely affects one of the functions of the trade mark. To establish that such function is affected, the claimant must show that the advertisement (that is, the ‘sponsored link’ to the defendant’s website) “does not enable a reasonably well-informed and reasonably observant internet user, or enables them only with difficulty, to ascertain whether the goods or services referred to by the advertisement originate from the proprietor of the trade mark or an undertaking economically connected to it or, on the contrary, originate from a third party”.

The test adopted by the CJEU for cases concerning the use of registered marks as Google Adwords is, therefore, a variation on the usual ‘likelihood of confusion’ test that applies in most ordinary infringement cases (although the ‘average consumer’ is replaced by the ‘reasonably well-informed and reasonably observant internet user’ and the burden of proof is given a somewhat differently formulation).

How witnesses from the public are gathered and selected

In order to establish its case for infringement in accordance with the test laid down by the CJEU, Interflora sought to adduce the evidence of members of the public who participated in a consumer survey it conducted. Interflora and its advisers followed a well-trodden path in English trade mark litigation practice.

Interflora’s witnesses were selected through a so called ‘witness gathering exercise’. According to the usual practice, members of the public (after going through a brief screening process) were asked to participate in a survey. In this case, they were asked to enter the word INTERFLORA into a Google search and to consider the search results (these were not ‘live’ internet results but rather a computer mock-up designed to replicate the results that would have been obtained before proceedings were issued). The participants were then asked a series of questions seeking to establish whether or not the ‘sponsored link’ to the M&S website enabled each participant to ascertain (without difficulty) whether the flowers and flower delivery services advertised through the link originate from Interflora or an undertaking economically connected to Interflora or, by contrast, from a third party.

Interflora selected some of these members of the public who participated in the survey as witnesses to give evidence before the Court. Importantly, Interflora did not seek to introduce the survey as a whole to the evidence (nor to rely on an expert opinion relating to the survey). Only the evidence of the witnesses. Nor did it suggest that these witnesses were representative of the wider population surveyed. One can safely surmise that the results of the survey as a whole were not good enough to support Interflora’s case.

The objection to the use of evidence from members of the public

Marks and Spencer objected to the use of the witnesses gathered through the survey on the grounds that the evidence was devoid of any evidentiary value. Particular emphasis was put on the fact that Interflora did not use the survey as a whole in its evidence (which is by no means unusual).

The trial court did not disagree that such evidence may carry limited weight in the circumstances, but felt constrained by the fact that there was no clear rule in English law that such evidence was inadmissible and by the fact that the use of such evidence was so well established in English trade mark infringement cases. Indeed, such evidence was heard and considered by the English courts in numerous cases (even though the Courts would not always give that evidence a great deal of weight).

The rules formulated by the Court of Appeal

The Court of Appeal took a different view. It held that under its case management powers, the trial Court can refuse the evidence, even if it is formally admissible. The Court of Appeal recognised that in most ordinary trade mark infringement cases (including where the test is the usual one of ‘likelihood of confusion’) the evidence of members of the public selected through a ‘witness gathering exercise’ is of virtually no probative value and that the costs involved are wholly disproportionate. The Court of Appeal pointed out that Interflora spent a whopping £250,000 on its survey and witness gathering exercise.

The Court of Appeal held that normally the evidence of witnesses from public should not be allowed as evidence of trade mark infringement, except in the following cases:

  1. where the witnesses testify in relation to their spontaneous reaction to the alleged infringement (that is, where they testify on a ‘real world’ situation rather than their reaction to an artificial scenario); 
  2. where the evidence of such witnesses is called to amplify the results of a reliable survey (that is, where there is a real argument that the witnesses are representative of a wider surveyed population and that the survey evidence itself - which is introduced into evidence - meets the requisite evidentiary requirements); or 
  3. where the goods or services in question are not goods or services supplied to ordinary consumers (therefore the judge cannot easily place himself in the shoes of the consumer to exercise his own judgment and may find the evidence of a specialist consumer helpful).

The Court mentioned further that the evidence of witnesses from the public may be more relevant where the question to be examined is whether the registered mark acquired distinctiveness (a question of validity rather than of infringement) and in passing off cases (where questions of the existence of goodwill and reputation and the issue of deception of consumers are intertwined).

Apart from those cases, the Court of Appeal said, it is difficult to imagine cases where the evidence of members of the public selected by a party could have any evidentiary value. The Court of Appeal emphasised that in considering applications to adduce such evidence the trial court should not only be persuaded that the witness evidence from members of the public has real value but also that the use of such evidence would be cost effective.

Procedural guidelines

It should be noted that the use of survey evidence (and any expert evidence) in trade mark and other intellectual property cases in England and Wales is always subject to the trial court’s discretion and subject to the court’s permission. Surveys are notoriously difficult to construct in a manner that satisfies the evidentiary standards of the Court. One must try to draw relevant (and if possible helpful) comments from consumers without posing leading questions, or being too suggestive and without inviting the consumer to speculate, and it is often difficult to conduct the survey in a ‘real world’ purchasing environment or to create a mock-up scenario sufficiently similar to a ‘real world’ situation to make a survey relevant and of probative value.

The Court of Appeal laid down the procedure that parties must follow in relation to the use of surveys in general as well as to the use of witnesses from the public in trade mark infringement cases. Accordingly, a party wishing to introduce consumer survey evidence must:

  • provide the result of any pilot survey conducted (if any); 
  • establish that the proposed (full) survey will meet the standards laid down by the case law (known as the Whitford guidelines); and 
  • provide the Court with information as to the cost of the pilot survey and an estimated of the costs of the full survey.

A party wishing to introduce the evidence of witnesses from the public who responded to a survey or other experiment, must:

  • provide the witness statements of the witnesses; 
  • demonstrate that their evidence would be of real value to the determination of the issues; 
  • disclose full details of all previous pilot surveys or experiments carried out ahead, or as part of, the witness gathering exercise; 
  • disclose how the witnesses were selected; and 
  • provide information as to the cost of all pilot surveys (including the gathering of witnesses from the public) and an estimate of the costs of any further related work.

According to these guidelines, it can be expected that the Courts will closely scrutinise applications for the use of surveys and witnesses from the public in trade mark cases.

Final note

The decision of the Court of Appeal in Interflora v. Marks and Spencer went against the trade mark owner which was seeking to enforce its registered trade mark. The claimant in that case may be displeased with having spent £250,000 on evidence that will not be allowed to be used. But from a wider perspective the result is by no means a bad one for trade mark proprietors nor indeed is it bad for the effective enforcement of intellectual property rights in the UK.

The decision in this case does not make it more difficult to establish trade mark infringement. There is nothing in this decision that raises the bar for potential claimants. It simply does away with a long established practice which was expensive and wasteful but which did not necessarily help trade mark owners prove their cases.

For many years, trade mark owners were advised - quite correctly - that the Courts will expect the evidence of witnesses from the public to establish likelihood of confusion or similar requirements of infringement. They may have also been advised that such evidence may even have greater weight than the more ‘scientific’ survey evidence - adduced through an expert - in view of the high standards required by the English Court in relation to consumer surveys. The English courts have traditionally had a bias in favour of real witnesses whose evidence can be heard ‘live’ by the Court and tested in cross-examination.

Now that the Court of Appeal finally decided that the evidence of a selected non-representative group of consumers is of no real probative value and that it ought not to be used, except in very specific cases, most ordinary trade mark infringement cases should be less expensive to conduct and both the parties to the litigation and the Courts will be able to focus on more relevant (and probably less expensive) evidence.

Finally, the parties to trade mark litigation may have to accept that in many cases the judge is sufficiently well placed to decide how the trade marks in dispute are perceived by an ‘average consumer’ - without the assistance of consumer surveys and witnesses from the public. Accordingly, the cost of litigation can be substantially reduced for all parties involved (and particularly for claimants) without adversely affecting the prospects of success of an infringement claim or the quality of the judicial decision. The decision of the Court of Appeal should be welcome by trade mark owners and practitioners alike.