The use of survey evidence in trade mark disputes is often desirable but it is not without its challenges.  One of which is that admissibility of the evidence is entirely up to the court. Sometimes the judge will find survey evidence useful but he will rarely rely heavily on it.  There are many examples where a judge has criticised the evidence and found it almost useless in reaching his decision. 

In M&S v Interflora Inc, M&S bought the keyword "Interflora" from Google as part of a marketing campaign for an online flower delivery service.  This meant that the M&S webpage was displayed as one of Google’s sponsored links when a user searched for "Interflora".  Interflora Inc, a flower delivery company that owned the trade mark Interflora, objected to the use of their mark arguing that there was a likelihood of confusion.  (We have examined this aspect of the case here, here and here.

Survey Evidence

The Court originally ordered that neither party had permission to adduce survey evidence without first obtaining leave of the Court.  Interflora applied for permission to adduce evidence at trial from witnesses gathered from two pilot surveys.  Interflora did not intend to rely on the evidence gained from the surveys, just use the witnesses that the survey highlighted as favouring Interflora’s case. 

The High Court allowed the evidence accepting that it may have probative value but the Court of Appeal overturned this decision. 

The impression of the average consumer is crucial in trade mark disputes.  The average consumer is a legal construct and "evidence of members of the public cannot stand proxy for [it]”.  The Court of Appeal said that surveys "can be an accurate diagnostic or predictive tool”.  They can be used to inform the judge of what consumer knew or shopping habits but the Court said generally "it does not help the court… to call members of the public simply to say that they would (or would not) be confused".

The general position, the court highlighted, should be that even if the evidence is technically admissible it should not be allowed unless it is valuable.  Interflora had failed to demonstrate the value of the evidence.

The court set out the following guidelines on admissibility:

  1. A party may conduct a true pilot survey without permission, but at his own risk as to costs; 
  2. No further survey may be conducted or adduced in evidence without the court’s permission; 
  3. No party may adduce evidence from respondents to any survey without the court’s permission.

The court should consider whether there is any real value in the evidence.  Reliability of the survey is key to this and even if the court is satisfied there is value, this value must not be outweighed by the costs.  Survey evidence will only be allowed in special or unusual cases. 

If witnesses identified through the survey are to be called the court must be provided with witness statements, the responses to the survey, evidence of how the survey was conducted, how the witnesses were selected from the respondents and the cost of carrying out the survey.  The court must also be satisfied that there is value in the evidence of the witnesses. 


This judgement will have a significant impact on the use of survey evidence in trade mark disputes. Obtaining survey evidence can be expensive and so it is important that a party carefully considers the guidance of the Court of Appeal before incurring the expense.  A party should be clear on what aspect of their case the evidence is designed to support and that the survey consists of open questions allowing for reliable responses.  This should go part of the way to ensuring that there is some value in the evidence.