The Court of Appeal today handed down judgment in the case of Interflora Inc. & others v Marks and Spencer PLC [2013] EWCA Civ 319, with Lewison LJ delivering the leading judgment.  This is the second judgment from the Court of Appeal in this case on the admissibility of evidence from witnesses in trade mark disputes identified through surveys.  We reported on the first decision from the Court of Appeal in November 2012 ("Interflora I") (see our newsflash here).


This is the latest interim battle between the parties in a trade mark infringement case about the use of sponsored keywords for online searches which has not yet come to trial in the UK.  Following the Court of Appeal's decision in Interflora I, Interflora was not permitted to adduce evidence from witnesses obtained through a witness collection exercise or survey.  However, the Court of Appeal caveated its decision by saying that different considerations may be relevant, if the evidence in question consists of the "spontaneous reactions of members of the public to the allegedly infringing sign or advertisement".  

Interflora returned to the High Court seeking permission to adduce witness evidence from 13 consumers on the basis that these witnesses gave evidence consisting of a spontaneous reaction to the allegedly infringing sign which demonstrated "confusion in the real world".  Mr. Justice Arnold granted Interflora permission to adduce evidence from these witnesses on 21 February 2013.  Marks and Spencer appealed this decision which was heard by the Court of Appeal on an expedited basis, given the short window before the trial of this case in April 2013. The Court of Appeal stated that it would not normally have heard an appeal on a case such as this, given that it appeared to be a discretionary case management decision, but it recognised that the appeal raised a question of principle.

Key points:

  • Marks and Spencer's appeal is allowed and Mr Justice Arnold's decision granting permission to adduce witness evidence from consumers is overturned on the basis that he had exercised his discretion to grant permission on a flawed basis.
  • Interflora is not permitted to rely upon consumer witness evidence from witnesses identified in a survey.
  • Lewison LJ re-iterates the test laid down in Interflora I in even stronger terms, saying that in hindsight, he did not make his message clear enough, and "Let me say it again, more loudly. A judge should not let in evidence of this kind unless the party seeking to call that evidence satisfies him (a) that it is likely to be of REAL value; and (b) that the likely value of the evidence justifies the cost".
  • It is not enough for a party to show that survey evidence might be of some value or more than minimal value in deciding the issues in a case.  Judges must exercise their power to limit or exclude technically admissible evidence which is not of real value.
  • Applications to adduce survey evidence or to rely upon witness evidence identified through surveys should be made well in advance of trial. Interflora's application was made in January 2013 and heard in February 2013, a mere 2 months before trial and this was criticised by the Court of Appeal.
  • The use of a leading question in the survey was found to have undermined the reliability of the survey and resultant witness evidence, with the Court of Appeal finding that it was clear that many witnesses had been led by a question into speculating about something which they had not thought about before.  Surveys should be carried out in compliance with the Whitford Guidelines (for more detail on these guidelines, see our Inteflora I newsflash) and should not contain leading questions. This is even more important in surveys involving spontaneous reactions to allegedly infringing signs as by almost very definition, leading questions do not produce spontaneous reactions.
  • The Court of Appeal found that Mr Justice Arnold had been misled by the impression created by the witness statements when compared to the raw data on which they were based.  The consumer witness statements were not a fair reflection of the raw data collected in the survey and the witness statements painted a picture more favourable to Interflora's case than the actual answers provided by these witnesses in the survey. 

Business impact:

  • The decision appears to make it even harder to adduce survey evidence in a trade mark infringement than was the case after Interflora I.  In light of the high hurdle which has been set, parties should give serious consideration to the evidential value of surveys in trade mark infringement cases before embarking on such surveys, especially in cases involving consumer confusion for ordinary consumer goods or services.  However, Jacob LJ does expressly state "that the decision does not mean that the days of survey evidence are over".  It is still conceivable to deploy a survey in a trade mark infringement case and to rely upon witnesses identified through such a survey, provided that such evidence is of real value and the party seeking to rely upon the survey can demonstrate that it has complied with the Whiftord Guidelines. However, if the survey amounts to "no more than scratching around for something" in the words of Jacob LJ, and produces little of value, common sense and a cost/benefit analysis dictates that the survey and the resultant witnesses should be excluded.
  • The decision is a further reminder of the importance of not "over-lawyering" witness statements.  The Court of Appeal draws attention to the clear direction given in paragraph 18.1 of Practice Direction 32 which states that a witness statement "must, if practicable, be in the intended witness's own words".  This is even more important in survey cases where the underlying answers as provided in the survey are available to Court so any discrepancies between the raw data and witness statements will be readily apparent.

  It remains to be seen whether Interflora will be able to convince the trial Judge of its infringement case in the absence of evidence from consumer witnesses identified through its surveys.