England and Wales Court of Appeal (Civil Division) Decisions

IN THE COURT OF APPEAL (CIVIL DIVISION)
ON APPEAL FROM THE HIGH COURT OF JUSTICE
CHANCERY DIVISION
INTELLECTUAL PROPERTY COMMUNITY TRADE MARK COURT
Mr Justice Arnold
HC08C03440

		Royal Courts of Justice Strand, London, WC2A 2LL
		20/11/2012

Lord Justice Lewison:

Introduction

1. This is a case about alleged trade mark infringement. The dispute is between Interflora and Marks & Spencer. Both are household names. The particular question raised on this appeal from Arnold J is whether the trade mark proprietor should be allowed to call the evidence of witnesses identified by means of a witness gathering exercise, where the proprietor does not intend to rely on the survey or questionnaire by means of which the witnesses were identified. There are three parts to the question:

i) Is there a rule of law which lays down a principle that such evidence is (a) always admissible or (b) always inadmissible?

ii) If it is admissible, should it nevertheless generally be excluded in the exercise of the court's powers under CPR Part 32.1 (2)?

iii) Should the particular means by which the witnesses were identified in this particular case lead the court to exclude the evidence of those witnesses?

2. First it is necessary to outline the background to the case and the issues. I take much of this verbatim from an earlier judgment of Arnold J at a previous stage in these proceedings: [2009] EWHC 1095 (Ch).

Background and issues

3. Interflora is a well known brand. The Interflora corporation is the owner of the trademark INTERFLORA, which is both a United Kingdom Trade Mark and a Community Trade Mark. Interflora operates the largest flower delivery network in the world. The Interflora network is a network of florists, each of which trades under and by reference to its own trade mark, but which also trades under and by reference to the trade marks in issue in this case. Orders may be placed either in person or by telephone with one member of the network (typically a member close to where the person placing the order lives) for fulfilment by another member of the network (usually the member closest to the address to which the flowers are to be delivered). The order is then transmitted from the first member to the second member. It follows that members of the network both have their own individual reputations under their own trade marks and benefit from the reputation of the trade marks built up by Interflora and the members of the network. Interflora also operates an online flower delivery service which enables orders to be placed via the internet, again for fulfilment by the member closest to the address to which the flowers are to be delivered. This is operated via a website located at www.interflora.com which resolves to country-specific websites such as www.interflora.co.uk.
4. Marks & Spencer plc ("M & S") is also very well known as a national retailer. It operates a flower delivery service via its website which enables orders to be placed online. M & S is not a member of the Interflora network.
5. M & S advertises its flower delivery service on the internet. One means by which it does this is by the use of AdWords bought from Google. AdWords operates as follows. When the Google search engine carries out a search, it shows the user two types of links to third party webpages in the search results. The first type, referred to as "natural" or "organic" links, are links to websites assessed to be relevant to the search query by the search engine's algorithm sorted in order of relevance. Although there are various ways in which website operators can and do seek to influence their position in the "natural" search results, in principle the ranking is an objective one based solely on relevance. By contrast, the second type of link, referred to as "sponsored links", are a form of advertising which are displayed because the website operator has paid for them to appear. Sponsored links are generally displayed in a separate section of the search results page, at the top and/or at the side. A sponsored link is displayed when the user enters one or more particular words into the search engine. These words, which are referred to as keywords, are selected by the advertiser in return for the payment of a fee. This is often referred to as "purchasing" the keywords.
6. A sponsored link consists of three elements. The first is an underlined heading which functions as a hyperlink. That is to say, when the user clicks on the link, the user's browser is directed to the advertiser's website. The hyperlink may consist of or include the keyword or it may not. The second element consists of some promotional text, which may or may not include the keyword. The third element consists of the URL of the advertiser's website. The URL does not function as a hyperlink (although the user could type it or cut-and-paste it into his or her browser and access the website in that way).
7. At the relevant time an advertiser like M & S could purchase a keyword such as INTERFLORA which is a registered trade mark of a competitor such as Interflora. That is what M & S did. It also purchased a number of variants of INTERFLORA. This had the result that, when a user entered the word INTERFLORA (or its variants) into the search engine, in addition to displaying the natural or organic results, the computer would also display the sponsored links including an advertisement for M & S for goods and services covered by that trade mark. One such advertisement read:

"M&S Flowers Online

   www.marksandspencer.com/flowers Gorgeous fresh flowers & plants. Order by 5pm for next day delivery"

8. It is advertising of this kind that is said to amount to infringement of Interflora's trade marks.
9. Interflora's pleaded case is now contained in its Amended Particulars of Claim. Some of the amendments were prompted by the result of a reference to and decision of the Court of Justice at an earlier stage in these proceedings, and to which I will refer later. The first of the allegations to which the contested evidence is said to be relevant is contained in paragraph 13.1.1 of the Amended Particulars of Claim which reads:

"… the manner in which the Defendant's advertising is presented does not enable reasonably well-informed and reasonably observant internet users or enables them only with difficulty to ascertain whether the goods or services referred to by the advertisement originate from the Claimants or an undertaking economically connected to the Claimants or on the contrary originate from a third party."

10. A related allegation is contained in paragraph 13.2.1 which reads:

"… the use complained of is detrimental to the distinctive character of the Trade Marks because there will be a blurring or dilution that will lessen the capability of the Trade Marks to distinguish the Claimants and their goods or services from those of others in that it does not enable reasonably well-informed and reasonably observant internet users or enables them only with difficulty to ascertain whether the goods or services referred to by the advertisement originate from the Claimants or an undertaking economically connected to the Claimants or on the contrary originate from a third party."

The legal framework

11. It is convenient to describe the legal framework by reference to article 9 of Regulation 40/94 (the Community Trade Mark Regulation), which is in similar terms to article 5 of the Trade Marks Directive and section 10 of the Trade Marks Act 1994. Article 9 provides:

"A Community trade mark shall confer on the proprietor exclusive rights therein. The proprietor shall be entitled to prevent all third parties not having his consent from using in the course of trade:

(a) any sign which is identical with the Community trade mark in relation to goods or services which are identical with those for which the Community trade mark is registered;

(b) any sign where, because of its identity with or similarity to the Community trade mark and the identity or similarity of the goods or services covered by the Community trade mark and the sign, there exists a likelihood of confusion on the part of the public; the likelihood of confusion includes the likelihood of association between the sign and the trade mark;

(c) any sign which is identical with or similar to the Community trade mark in relation to goods or services which are not similar to those for which the Community trade mark is registered, where the latter has a reputation in the Community and where use of that sign without due cause takes unfair advantage of, or is detrimental to, the distinctive character or the repute of the Community trade mark.

2. The following, inter alia, may be prohibited under paragraph 1: …

(b) offering the goods, putting them on the market or stocking them for these purposes under that sign, or offering or supplying services thereunder; …

(d) using the sign on business papers and in advertising."

12. The Court of Justice has now considered the question of AdWords on a number of occasions, including a previous stage of this litigation. In Google France SARL v Louis Vuitton Malletier SA Joined cases C-236/08, C-237/08 and C-238/08 [2010] ECR I-2417 the Court held that the sign selected by an advertiser as a keyword in the context of an internet referencing service is the means used by the advertiser to trigger the display of its advertisement and is thus "[used] in the course of trade" within the meaning of article 9. That use is use in relation to the advertiser's goods or services, even where the sign selected as a keyword does not appear in the advertisement itself: BergSpechte Outdoor Reisen und Alpinschule Edi Koblmuller GmbH v Guni Case C-278/08 [2010] ECR I-2517. However, it does not follow that infringement is established, because of an important qualification laid down by the Court. That qualification is that use of the sign will only amount to infringement if it is liable to have an adverse effect on one of the functions of the mark; including its essential function of guaranteeing to consumers the origin of the goods. In this very case the Court of Justice explained the ambit of the inquiry before the national court (Case C-323/09) [2012] Bus LR 1440:

"44. The question whether a trade mark's function of indicating origin is adversely affected when internet users are shown, on the basis of a keyword identical with the mark, a third party's advertisement, such as that of a competitor of the trade mark proprietor, depends in particular on the manner in which that advertisement is presented. That function is adversely affected if the advertisement does not enable reasonably well-informed and reasonably observant internet users, 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…. In such a situation, which is, moreover, characterised by the fact that the advertisement appears immediately after the trade mark has been entered as a search term and is displayed at a point when the trade mark is, in its capacity as a search term, also displayed on the screen, the internet user may be mistaken as to the origin of the goods or services in question….

45. Where a third party's advertisement suggests that there is an economic link between that third party and the proprietor of the trade mark, the conclusion must be that there is an adverse effect on that mark's function of indicating origin. Similarly, where the advertisement, while not suggesting the existence of an economic link, is vague to such an extent on the origin of the goods or services at issue that reasonably well-informed and reasonably observant internet users are unable to determine, on the basis of the advertising link and the commercial message attached thereto, whether the advertiser is a third party vis-à-vis the proprietor of the trade mark or whether, on the contrary, it is economically linked to that proprietor, the conclusion must be that there is an adverse effect on that function of the trade mark….

46. It is for the referring court to assess whether, on the facts of the dispute before it, the trade mark's function of indicating origin as described in the preceding paragraphs is, or is liable to be, adversely affected….

…

49. Indeed, if the referring court's assessments of the facts were to show that M & S's advertising, displayed in response to searches performed by internet users using the word 'Interflora', may lead those users to believe, incorrectly, that the flower-delivery service offered by M & S is part of Interflora's commercial network, it would have to be concluded that that advertising does not allow it to be determined whether M & S is a third party in relation to the proprietor of the trade mark or whether, on the contrary, it is economically linked to that proprietor. In those circumstances, the function of the INTERFLORA trade mark of indicating origin would be adversely affected.

50. In that context, as has been observed at paragraph 44 of this judgment, the relevant public comprises reasonably well-informed and reasonably observant internet users. Therefore, the fact that some internet users may have had difficulty grasping that the service provided by M & S is independent from that of Interflora is not a sufficient basis for a finding that the function of indicating origin has been adversely affected.

51. In carrying out its examination of the facts, the referring court may choose to assess, first, whether the reasonably well-informed and reasonably observant internet user is deemed to be aware, on the basis of general knowledge of the market, that M & S's flower-delivery service is not part of the Interflora network but is, on the contrary, in competition with it and, second, should it become apparent that that is not generally known, whether M & S's advertisement enabled that internet user to tell that the service concerned does not belong to the Interflora network.

52. In particular, the referring court may take into account that, in the present case, the commercial network of the trade mark proprietor is composed of a large number of retailers which vary greatly in terms of size and commercial profile. The Court considers that, in such circumstances, it may be particularly difficult for the reasonably well-informed and reasonably observant internet user to determine, in the absence of any indication from the advertiser, whether or not the advertiser – whose advertisement is displayed in response to a search using that trade mark as a search term – is part of that network.

53. Having regard to that situation and to the other matters that it may consider relevant, the referring court will, in the absence of any general knowledge such as that referred to at paragraph 51 of this judgment, have to determine whether or not the use of words such as 'M & S Flowers' in an advertisement such as the one set out at paragraph 19 of this judgment is sufficient to enable a reasonably well-informed and reasonably observant internet user who has entered search terms including the word 'Interflora' to tell that the flower-delivery service offered does not originate from Interflora." (Emphasis added)

13. Thus "the reasonably well-informed and reasonably observant internet user" enters the pantheon of characters who inhabit the world of intellectual property alongside his fellows: the average consumer (trade marks), the informed user (registered designs) and the uninventive person skilled in the art (patents).
14. The evidence with which we are concerned is, as I understand it, designed to place before the court evidence from real people who will advance Interflora's case on infringement. Mr Hobbs QC objected to the admission of this evidence. His objections were both at a macro and a micro level. But before coming to those I should set out some more of the factual material; and in particular the process by which the potential witnesses were or are to be selected.

The witness gathering exercise

15. Interflora conducted two exercises, with the help of Mr Malivoire of GfK NOP, a market research company. Mr Malivoire is very experienced in this field and indeed has featured in some of the previous cases. The first was carried out on 4 and 6 February 2012 in Bromley and Uxbridge respectively. Respondents were recruited on the street and were first asked a number of screening questions. Three of these questions were whether the potential respondents had purchased or looked into the purchase of flowers in the previous two to three years; whether they had undertaken activities on the internet but were not confident about creating a web page; and whether they were in a position to read a computer screen. Some 95 respondents passed the screening test, and were chosen to answer further questions. They were brought into a hall set up with four laptop computers. Each participant was asked to type the word "Interflora" into the search box on the Google homepage. It was not a real Google search, because each computer was set up so as to display one of the search results pages which are the subject of this litigation. The search results page brought up Interflora, both as a natural or organic result and also as a sponsored link. The search results page also brought up an M & S sponsored link, quoted above, together with other sponsored links. Participants were then invited to scroll up and down the page and to tell the interviewer when they were ready to answer questions. Once ready each participant was asked a series of structured questions. These questions were changed during the course of the survey. The questions did not distinguish between natural or organic results on the one hand and sponsored links on the other. The questions were framed generically by reference to "results" or "search results". Mr Hobbs was particularly critical of question 6. In the first survey it was posed as follows:

"Thinking specifically about this search result (POINT to MARKS & SPENCER RESULT). What if anything do you think this tells you about any relationship between Interflora and Marks and Spencer?"

16. The second exercise was conducted at Bexleyheath on 24 March 2012 and at High Wycombe on 26 March 2012 interviews were conducted. The methodology was broadly the same as for the first survey. This time, however, question 6 had been changed. It was now posed as follows:

"Thinking specifically about this search result (POINT to MARKS & SPENCER RESULT). Is there anything that tells you it is not related to Interflora?"

17. Interflora has disclosed the answers given by all participants to the questions they were asked during the course of the exercises. I shall return to some of the answers when considering Mr Hobbs' micro objection.
18. These exercises have been referred to as "pilot surveys". But I do not think that that is what they are. In his witness statement Mr Phillips, M & S's expert, said that by "pilot survey" he meant "a survey to ascertain the best methodology for undertaking a survey in particular circumstances". In his witness statement in reply Mr Malivoire seems to me to have agreed. He said:

"As Mr Phillips acknowledges … the purpose of a pilot survey is to ascertain the best methodology for undertaking a survey. As stated in "The Practice of Market and Social Research" (the recommended textbook for those studying for the market Research Society Advanced Certificate…) "conducting a pilot survey is an invaluable way of testing it [a questionnaire] out – it will show which questions are difficult, which ones give the type of answers you were expecting and so on"."

19. However, in the present case, although calling the exercise a "pilot survey" Interflora has not used the two exercises as a means of testing a questionnaire for use in a full survey. It does not intend to carry out any further survey. Nor does it rely on the so-called pilot surveys themselves, and accepts that they are not statistically reliable. But it does wish to call witnesses identified by means of these exercises. However, it proposes to call as witnesses only those participants whose answers to the questions are the most favourable to Interflora's case. Even labelling these exercises as "pilot surveys" is apt to cause confusion.

The judgment below

20. Arnold J gave permission to Interflora to call witnesses identified as a result of the first exercise; but refused Interflora permission to call witnesses identified as a result of the second: [2012] EWHC 1722 (Ch); [2012] FSR 32. On the macro question whether the evidence should be permitted at all Arnold J acknowledged the force of Mr Hobbs' objections but said (§ 12):

"The problem, as I see it, at this level of the judicial hierarchy is that there is considerable precedent for judges of this Division taking into account the evidence of witnesses obtained by means of a survey in circumstances where the survey itself is either not relied upon or is conceded not to have statistical validity. … Furthermore, in a number of those cases the judges in question not merely admitted, but also relied upon the evidence of witnesses obtained by such surveys in circumstances where they acknowledged that the questions posed in the surveys were flawed.

13 In those circumstances it seems to me that the present state of the law is that evidence obtained from witnesses by virtue of the fact that they have been respondents to a survey is admissible and may be probative, even though the survey itself is not and cannot be relied upon and even though the questions posed in the survey were flawed. It follows that I am unable to accede to the first submission made by counsel for Marks & Spencer."

21. Mr Silverleaf QC, appearing for Interflora, says that on the macro question Arnold J was following a well trodden path, which has been recognised as valid for many years and at the highest levels of authority.
22. On the micro question the judge acknowledged that question 6 in each questionnaire was a leading question. However, he went on to say (§ 17):

"In my view, it is not possible to draw a hard and fast distinction between questions which are objectionable and questions which are not objectionable. There is a spectrum of questions between questions which are completely open-ended on the one hand through questions that prompt speculation, to questions that have a degree of leading quality to them, to questions that are completely misleading on the other end of the spectrum. Where on the spectrum a particular question lies is a matter of judgment in the individual case."

23. On the particular facts he ruled that although question 6 in the first questionnaire was flawed, it was not so flawed that the evidence of witnesses who had answered it should be excluded. But as regards question 6 in the second questionnaire he came to the opposite conclusion. Mr Silverleaf says that the judge was right as regards questionnaire 1 but wrong as regards questionnaire 2.

The macro objection

24. Mr Hobbs' macro objection is that the court should not permit evidence to be adduced from witnesses selected as a result of a survey unless the survey itself is statistically reliable. It does not matter whether such evidence is technically admissible, because the court has power under CPR 32.1 (2) to exclude evidence that is otherwise admissible. The argument runs thus:

i) The question at issue is whether the reasonably well-informed and reasonably observant internet user would or would not understand the M & S advertisement to indicate that M & S was part of the Interflora network.

ii) The reasonably well-informed and reasonably observant internet user is not a real person. He or she is a legal construct. As the Court of Justice made clear, the fact that some internet users may have had difficulty grasping that the service provided by M & S is independent from that of Interflora is not a sufficient basis for a finding that the function of indicating origin has been adversely affected.

iii) Accordingly, simply to call some internet users to give evidence is not probative of the issue in the case.

iv) That evidence can only be probative if those who are called can in some way be seen to stand proxy for the legal construct through whose eyes the essential question must be judged.

v) Unless the survey used for witness collection is itself a reliable survey, the court will have no means of knowing whether the selected witnesses can be treated as reliable proxies for this legal construct. The problem is compounded where, as here, the party calling the witnesses is permitted to select those who give most support to its case.

vi) Even if the evidence of such witnesses is, in principle, admissible, it is likely to be of such marginal utility and so expensive and time consuming to collect, analyse and deal with in court, that the court ought to exclude it in exercise of its powers under CPR Part 1.4 (2) (h) and CPR Part 32.1.

25. It is common ground that the point of principle has never been argued before in our courts.

The substantive legal question to be answered

26. Mr Silverleaf relied on a number of cases in which surveys and evidence called from witnesses identified as a result of surveys. Many of them were cases of passing off. In my judgment passing of cases are of limited utility in a case such as this, because the legal question to be answered is a different one. This can, I think, be illustrated by two of the cases upon which Mr Silverleaf relied: Neutrogena Corporation v Golden Ltd [1996] RPC 473 and Chocosuisse Union des Fabricants Suisse de Chocolat v Cadbury Ltd [1998] RPC 117.
27. In Neutrogena the claimant had substantial reputation and goodwill in haircare and skincare products sold under the name Neutrogena. It was also the owner of three registered marks for that word. The defendant began to market a range of shampoo and other products under the name Neutralia. Neutrogena sued for passing off and infringement of its marks. It was accepted by the defendant (represented as it happens by both Mr Hobbs and Mr Silverleaf) that the case on trade mark infringement stood or fell with the case in passing off: [1996] RPC 473, 488. So the judge (Jacob J) concentrated on passing off. He directed himself ([1996] RPC 473, 481):

"It is not a defence to passing off that many of a defendant's sales do not cause deception or confusion. There is passing off even if most of the people are not fooled most of the time but enough are for enough of the time. By "enough" I mean a substantial number of the plaintiff's customers or potential customers deceived for there to be a real effect on the plaintiff's trade or goodwill. In this case (where most of these are probably not confused) the crucial question is whether or not the plaintiffs have established a sufficient degree of confusion and deception to take the case above a de minimis level. For there are always some people who are confused and even when products and names are well-differentiated, mistakes do occur."

28. Having considered a variety of different kinds of evidence Jacob J concluded on the facts [1996] RPC 473, 483:

"…the plaintiffs have managed to adduce what I find to be convincing evidence that there is deception and confusion and that this is above a trivial level, even though most people are not deceived." (Emphasis added)

29. He therefore concluded that passing off had been established.
30. The judge's self-direction was attacked in the Court of Appeal but largely survived. Morritt LJ thought that the use of the phrases "more than de minimis" and "above a trivial level" were best avoided; but held that the judge had applied the correct test; namely whether a substantial number of members of the relevant public had been deceived. The judge's finding that passing off had been established was upheld. The important point is that passing off had been established even though most people were not deceived.
31. In Chocosuisse an association of Swiss chocolate manufacturers sued Cadbury in passing off. Cadbury had marketed a chocolate bar called "Swiss Chalet" which was not manufactured in Switzerland. Laddie J followed the guidance in Neutrogena and directed himself that he had to be satisfied that there was a likelihood of confusion "among a substantial number of the public": [1998] RPC 117, 137. He too considered a variety of different kinds of evidence. His overall conclusion was that the case in passing off had been established. He reached that conclusion on the basis of the following finding of fact [1998] RPC 117, 143:

"Many people, and particularly those who are more observant, would not be confused. For them the words "Swiss Chalet" will signify nothing but a pretty sounding name for a bar of chocolate. They will convey no other message. However I have come to the conclusion that there are some who will be struck by the largest and most prominent word on the defendant's packaging namely "Swiss" and think that it is a reference to an attribute of the product itself. I think it is likely that some will think that it is an indication that the product is Swiss chocolate. Some, like Mr Crocker, may not see the reference to Cadbury. Others might not believe that all Cadbury chocolate is made in England. In fact it is not all made here. Cadbury like many other manufacturers has set up factories or formed alliances abroad. Further, for some the get up of the packaging with its typical Swiss scene will tend to reinforce the message of the word "Swiss". Finally some may be left in confusion as to whether Swiss Chalet is Swiss chocolate or not.

I have found this the most difficult issue in the case. However, I have come to the conclusion that a substantial number of members of the public who regard Swiss chocolate as the name for a group of products of repute will be confused into thinking that Swiss Chalet is a member of that group by reason of the use of the name Swiss Chalet. It is likely that the number who think that will be smaller than the number for whom there will be no confusion but, in my view, it is still likely to be a substantial number. It follows that on this issue the plaintiffs succeed." (Emphasis added)

32. The judge thought that he himself would have been in the majority. The judge's decision was subsequently upheld by the Court of Appeal, on the basis that this was a question of fact for the judge: [1999] RPC 826.
33. I should also refer to Reed Executive Plc v Reed Business Information Ltd [2004] EWCA Civ 159; [2004] RPC 40. In the course of his judgment in that case Jacob LJ (with whom Auld and Rix LJJ agreed) said (§ 82):

"Next the ordinary consumer test. The ECJ actually uses the phrase "average consumer" (e.g. Lloyd paras [25] and [26]). The notion here is conceptually different from the "substantial proportion of the public" test applied in passing off (see e.g. Neutrogena Corp v Golden Ltd [1996] RPC 473). The "average consumer" is a notional individual whereas the substantial proportion test involves a statistical assessment, necessarily crude. But in the end I think they come to the same thing. For if a "substantial proportion" of the relevant consumers are likely be confused, so will the notional average consumer and vice versa. Whichever approach one uses, one is essentially doing the same thing—forming an overall ("global") assessment as to whether there is likely to be significant consumer confusion. It is essentially a value judgment to be drawn from all the circumstances. Further conceptional over-elaboration is apt to obscure this and is accordingly unhelpful. It may be observed that both approaches guard against too "nanny" a view of protection—to confuse only the careless or stupid is not enough."

34. I agree entirely that the average consumer (in trade mark infringement) is conceptually different from the substantial proportion of the public test (in passing off). What I find difficult to accept is that they come to the same thing. If most consumers are not confused, how can it be said that the average consumer is? I do not think that this particular paragraph of Jacob LJ's judgment is part of the ratio of the case and, with the greatest of respect, despite Jacob LJ's vast experience of such cases I question it. In some cases the result will no doubt be the same however, the question is approached; but I do not think that it is inevitable.
35. There is, of course, no doubt that a valid survey can be an accurate diagnostic or predictive tool. They are used daily to sample public opinion on a variety of different topics. For example, they are used to gauge support for rival candidates in an election and to predict the eventual result. Suppose that a valid survey shows that in an election 49 per cent of the electorate support candidate A and 51 per cent support candidate B. It would be possible to say on the strength of such a survey that B will win the election. It would also be possible to say that a substantial proportion of the electorate will vote for candidate A. But what a survey does not, I think, tell you is: for whom will the average voter vote? In cases where acquired distinctiveness of a mark is in issue a survey may accurately identify that proportion of the relevant public which recognises the mark as a badge of trade origin. It will then be for the fact finding tribunal, with the aid of such a survey, to decide whether a significant proportion of the relevant public identify goods as originating from a particular undertaking because of the mark: see (Case C-108/97) Windsurfing Chiemsee Produktions-und-Vertriebs GmbH v Boots-und-Segelzubehor [2000] Ch 523 §§52, 53.
36. In our case the question is whether M & S's advertisement would enable a reasonably well-informed and reasonably observant internet user to grasp without undue difficulty that Interflora and M & S were independent. This, as the Court of Justice has emphasised is not a question of counting heads, but is a qualitative assessment. The fact that some internet users might have had difficulty in grasping that Interflora and M & S were independent is not sufficient for a finding of infringement. If, by analogy with Neutrogena and Chocosuisse, the court were to conclude that most internet users would have grasped that, but that some would not, I cannot see that the court would be any closer to answering the legal question.

The average consumer

37. In the most recent case of trade mark infringement to reach this court, Specsavers International Healthcare Ltd v Asda Stores Ltd [2012] EWCA Civ 24; [2012] FSR 19, Kitchin LJ said (§ 87):

"In my judgment the general position is now clear. In assessing the likelihood of confusion arising from the use of a sign the court must consider the matter from the perspective of the average consumer of the goods or services in question and must take into account all the circumstances of that use that are likely to operate in that average consumer's mind in considering the sign and the impression it is likely to make on him. The sign is not to be considered stripped of its context."

38. In Bach and Bach Flower Remedies Trade Marks [2000] RPC 513 the main issue was whether the mark had initially possessed or had acquired distinctiveness. Evidence from members of the public had been called at trial. These witnesses had been identified as a result of a survey of 9,500 people; but the judge rejected the survey itself as being unreliable. The trial took place before the ECJ had laid down the test of the "average consumer". So the wisdom of calling the evidence was not challenged, although the trial judge did not find it particularly helpful. On appeal it was recognised that the judge had not applied the "average consumer" test: § 37. Morritt and Chadwick LJJ took slightly different approaches to the evidence of members of the public. Morritt LJ said (§ 35):

"The usage in question must be by those engaged in the relevant trade or activity. Normally that will be the usage of the average consumer of the goods in question as described in Lloyd Schuhfabrik. Obviously the evidence on that question is not limited to those who are consumers or end-users but may extend to others concerned in the trade such a manufacturers, wholesalers and retailers. But I do not think that the court is assisted by repetitious evidence from individuals put forward by the parties, whether expressly or not, as archetypal average consumers or end-users for, by definition, no one individual is such a consumer or end-user and the issue cannot be resolved by counting heads. We were told that the judges before whom cases of this sort are heard have increasingly imposed restrictions on the quantity of such evidence they are prepared to admit. In my view that practice is to be encouraged."

39. The "usage" to which Morritt LJ referred was usage of the word "Bach" to indicate the origin of flower remedies. Nevertheless Morritt LJ came to his own conclusion on the perception of the average consumer. He said (§ 38):

"On that basis the evidence of a capability to distinguish to which I have referred should be rejected as not applicable to the average consumer as defined by the European Court of Justice. Such a person is deemed to be reasonably informed, reasonably observant and circumspect. But the actions and beliefs of the persons who prompted the comments of the judge I have referred to in paragraph 36 should not, in my view, be attributed to the average consumer. In the case of the first category the relevant action appears to have been prompted by carelessness and ignorance, which are not to be attributed to the average consumer. In the case of the second the judge rejected the evidence as sufficient to undermine his clear findings at page 38 that in 1979 and thereafter the expression Bach Flower Remedies and the word Bach would have been used and understood in a generic sense. In those circumstances I cannot see how any different understanding can be attributed to the average consumer. If the average consumer is to be regarded as reasonably informed, observant and circumspect then he must know the sense in which the word Bach was used and understood and that sense was generic." (Emphasis added)

40. In other words the evidence of members of the public could not stand proxy for the "average consumer". He repeated this later in his judgment when dealing with a different point (§ 49). So the evidence called from the public was not probative.
41. Chadwick LJ, as I read him, went a little further. He said (p. 535):

"I agree with Morritt L.J. that, in seeking to apply the test, the court is unlikely to be assisted by repetitious evidence from individual consumers, put forward by each party as the embodiment of the average consumer. The task for the court is to inform itself, by evidence, of the matters of which a reasonably well informed and reasonably observant and circumspect consumer of the products would know; and then, treating itself as competent to evaluate the effect which those matters would have on the mind of such a person with that knowledge, ask the question: would he say that the words or word identify, for him, the goods as originating from a particular undertaking?" (Emphasis added)

42. Thus Chadwick LJ considered that the evidence that might be helpful was evidence of what consumers knew: not whether they would have identified the words as a badge of trade origin. Thus (for example) evidence of shopping habits might well be of value, especially in a case where the judge is likely to be unfamiliar with the relevant goods or services. But absent special circumstances it does not generally help the court to call members of the public simply to say that they would (or would not) be confused.
43. Thorpe LJ agreed with both judgments. If and in so far as there is any difference in approach between Morritt LJ and Chadwick LJ I prefer the approach of Chadwick LJ, which seems to me to apply more faithfully the requirement that the risk of confusion must be evaluated from the perspective of the average consumer.
44. In the particular context of AdWords, the average consumer has been replaced by the reasonably well-informed and reasonably observant internet user. But the underlying concept of a legal construct is the same.

Ordinary consumer goods and services

45. If Interflora cannot prove its case without calling evidence from internet users that would be a powerful reason for allowing such evidence to be adduced.
46. Dalgety Spillers Foods Ltd v Food Brokers Ltd [1994] FSR 504 concerned an application for an interlocutory injunction rather than a trial. But even in that context Blackburne J made some valuable observations about the evidence. The underlying dispute pitted against each other Golden Wonder Pot Noodles and Nissin Cup Noodles. I must emphasise that the cause of action was solely in passing off, so that the "average consumer" test did not apply. The claimant proposed to rely on evidence from experts and also on evidence from members of the public. So far as the expert evidence was concerned Blackburne J said:

"Whether in any case such evidence will be admissible will depend upon a variety of factors which it is neither possible nor desirable to define. Broadly, the test in such cases must be whether the experience which a judge must be taken to possess as an ordinary shopper of consumer will enable him, just as well as any other, to assess the likelihood of confusion. If it will, then the evidence will not be admissible. If, for whatever reason, it will not, then such evidence will be admissible."

47. On the facts he concluded:

"In the present case I am concerned with a straightforward food product which, according to the plaintiff's evidence, is widely distributed throughout the United Kingdom and is purchased in huge numbers by ordinary members of the general public. How in these circumstances it assists the court to have, on the one side, a body of persons, experienced in retailing, to testify to the likelihood of deception and to have, on the other hand, a collection of similar persons testifying to the converse proposition, the more particularly when the expressions of opinion are no more than bare assertions of the likelihood of confusion, I cannot see. In my judgment therefore I rule that such evidence is, in this case, inadmissible."

48. Turning to the evidence from members of the public he said:

"My overall feeling was that although I did not consider them to be inadmissible nevertheless the circumstances in which those affidavits had been obtained were such that, even assuming that no leading questions had been put to the deponents and that, notwithstanding the use of pre-prepared affidavits tailored to the particular evidence of each deponent, the plaintiff's solicitors had not suggested to the deponents how they should frame their responses, I am not able to attach very much weight to them. On the other hand I do not feel able to disregard them altogether. I cannot, on an interlocutory motion, simply brush them aside. They do provide some basis for the suggestion that members of the public may indeed be deceived when seeing a Nissin Cup Noodles container into thinking that it comes from the plaintiff's Pot Noodle range.

What then is the state of the evidence on the question whether the get-up of the Nissin Cup Noodles container is likely to confuse or deceive? There is some evidence although, as I have stated, unsatisfactory in a number of respects from members of the public of the likelihood of confusion. If, contrary to my view on the admissibility of evidence from those with experience in the trade, such evidence is admissible then there is a body of evidence which, though contradicted by the defendants' evidence, supports the likelihood of confusion. Over and above that there is my own reaction as somebody possessing experience of ordinary shopping habits.

In all the circumstances, even if I disregard the evidence of those with experience in the trade, I cannot say that there is no serious question to be tried as to the likelihood of the Nissin Cup Noodles product being confused with the plaintiff's Pot Noodle products."

49. More recently in esure Insurance Ltd v Direct Line Insurance Plc [2008] RPC 34 Arden LJ said (§ 56):

"The services sold by the parties were identical and were of a kind familiar to members of the public. In those circumstances, I see no reason why the hearing officer should not have decided the issue of similarity on his own in the absence of evidence apart from the marks themselves and evidence as to the goods or services to which they were, or, in the case of esure's mark, were to be applied."

50. In such cases it is clear as a matter of domestic law that not only is the ultimate issue one for the judge, rather than the witnesses; but also that the judge can reach a conclusion in the absence of evidence from consumers. He or she is in the position of a notional juror, using his or her own common sense and experience of the world: Re GE Trade Mark [1973] RPC 297, 321; esure Insurance Ltd v Direct Line Insurance plc [2008] RPC 919, 938.
51. Nor does European law compel any different result. In (Case C-210/96) Gut Springenheide GmbH v Oberkreisdirektor des Kreises Steinfurt—AMT für Lebensmittelüberwachung [1998] ECR I-4657 the Court of Justice said:

"[30] There have been several cases in which the Court of Justice has had to consider whether a description, trade mark or promotional text is misleading under the provisions of the Treaty or of secondary legislation. Whenever the evidence and information before it seemed sufficient and the solution clear, it has settled the issue itself rather than leaving the final decision for the national court.

[31] In those cases, in order to determine whether the description, trade mark or promotional description or statement in question was liable to mislead the purchaser, the Court took into account the presumed expectations of an average consumer who is reasonably well-informed and reasonably observant and circumspect, without ordering an expert's report or commissioning a consumer research poll.

[32] So, national courts ought, in general, to be able to assess, on the same conditions, any misleading effect of a description or statement designed to promote sales.

[33] It should be noted, further, that, in other cases in which it did not have the necessary information at its disposal or where the solution was not clear from the information before it, the Court has left it for the national court to decide whether the description, trade mark or promotional description or statement in question was misleading or not."

52. The court continued:

"[35] The Court has not therefore ruled out the possibility that, in certain circumstances at least, a national court might decide, in accordance with its own national law, to order an expert's opinion or commission a consumer research poll for the purpose of clarifying whether a promotional description or statement is misleading or not.

[36] In the absence of any Community provision on this point, it is for the national court, which may find it necessary to order such a survey, to determine, in accordance with its own national law, the percentage of consumers misled by a promotional description or statement that, in its view, would be sufficiently significant in order to justify, where appropriate, banning its use."

53. The court's final conclusion was:

"… in order to determine whether a statement or description designed to promote sales of eggs is liable to mislead the purchaser, in breach of Article 10(2)(e) of Regulation 1907/90, the national court must take into account the presumed expectations which it evokes in an average consumer who is reasonably well-informed and reasonably observant and circumspect. However, Community law does not preclude the possibility that, where the national court has particular difficulty in assessing the misleading nature of the statement or description in question, it may have recourse, under the conditions laid down by its own national law, to a consumer research poll or an expert's report as guidance for its judgment."

54. There are two points that I wish to make. First, the court said that a consumer research poll is not precluded "where the national court has particular difficulty in assessing the misleading nature of the statement or description in question" (emphasis added). It is plainly not intended that such evidence should be routine. Second, the court went no further than to say that a consumer research poll or expert's report might be of assistance. By "consumer research poll" I understand the court to be referring to a statistically reliable and well conducted poll. That is a long way from blessing the practice of calling a few hand-picked members of the public.
55. Mr Silverleaf referred to a number of cases either in the Court of Justice or the General Court in which survey evidence has been relied on. However, as I think he accepted, the surveys in question were statistically significant surveys. There was no instance of reliance on witness evidence deployed as a result of a statistically unreliable survey. OHIM also recognises the value of surveys if they are "impeccably conducted": see Niedermann: Surveys as evidence in proceedings before ?OHIM (2006) IIC 260.

Are AdWords ordinary consumer services?

56. AdWords and internet searches have already made their way into reported cases. In Reed Executive Plc v Reed Business Information Ltd Jacob LJ said (§§ 137-142):

"…As anyone who uses internet searches knows, in addition to the results of a search under a particular name or phrase, one often gets unasked for "banner" advertisements. Most of the time they are nothing but an irritation and are ignored. But you can, if you wish, "click-through", i.e. click on the banner and be taken to the advertiser's site. … The web-using member of the public knows that all sorts of banners appear when he or she does a search and they are or may be triggered by something in the search. He or she also knows that searches produce fuzzy results—results with much rubbish thrown in. The idea that a search under the name Reed would make anyone think there was a trade connection between a totaljobs banner making no reference to the word "Reed" and Reed Employment is fanciful. No likelihood of confusion was established."

57. In 32Red plc v WHG (International) Ltd [2011] EWHC 62 (Ch); [2011] RPC 26 Henderson J said (§ 145):

"Adwords are a familiar feature of life on the internet, and the reaction of the average consumer, when the sponsored link to 32red.com came up on his screen, would I think be one of indifference or irritation, but not of confusion. If the consumer then clicked on the 32Red site, its clear branding could have left no room for reasonable doubt about the identity of the casino whose services were on offer. If the consumer did not click on the site, he will presumably have continued with the search which he originally intended, and is most unlikely to have supposed that there was any business connection between 32Red and the site which he originally sought to access. In a very few cases, the customer may have been grateful for the reminder that there was another online casino called 32Red, but in such cases confirmation or clarification would have been the result, not confusion. I am therefore wholly unpersuaded that the effect of 32Red's adword campaign was to mislead the public."

58. This finding was not disturbed on appeal: [2012] EWCA Civ 19; [2012] RPC 19 (§ 72).
59. In my judgment the results thrown up by search engines on the internet fall within the general description of ordinary consumer services in relation to which the judge can make up his or her own mind without the need either for expert evidence or the evidence of consumers. In my judgment, therefore evidence from real internet users is not essential to establish Interflora's case.

Surveys and witness collection exercises: a brief overview

60. As Mr Silverleaf points out survey evidence in one form or another has been placed before the courts for many years. But as knowledge about the methodology of the conduct of surveys has grown judges have become more sceptical about their value unless conducted rigorously.
61. Imperial Group plc v Philip Morris Ltd [1984] RPC 293 dealt a body blow to the reception of survey evidence. Whitford J was trenchant in his criticisms of the survey evidence adduced in that case, and laid down guidelines for the conduct of future surveys. These ("the Whitford guidelines") can be summarised as follows:

i) if a survey is to have any validity at all, the way in which the interviewees are selected must be established as being done by a method such that a relevant cross-section of the public is interviewed;

ii) any survey must be of a size which is sufficient to produce some relevant result viewed on a statistical basis;

iii) the party relying on the survey must give the fullest possible disclosure of exactly how many surveys they have carried out, exactly how those surveys were conducted and the totality of the number of persons involved, because otherwise it is impossible to draw any reliable inference from answers given by a few respondents;

iv) the questions asked must not be leading; and must not direct the person answering the question into a field of speculation upon which that person would never have embarked had the question not been put;

v) exact answers and not some sort of abbreviation or digest of the exact answer must be recorded;

vi) the totality of all answers given to all surveys should be disclosed; and

vii) the instructions given to interviewers must also be disclosed.

62. Following this case, surveys came in for more and more judicial criticism. Moreover as Kerly on Trade Marks (15^th edition) points out (§ 21-026) "it is very difficult to design a survey which will pass muster in court". As I understand it the United Kingdom Intellectual Property Office will only accept survey evidence which complies with the Whitford guidelines: See Tribunal Practice Note (2/2012).
63. As Mr Silverleaf correctly submitted, the extensive criticisms of surveys as evidence led in the 1990s to a marked shift in the way that surveys were used in litigation, with the party relying on such material frequently choosing not to rely on the survey as evidence, but rather to use the survey to find members of the public whose reactions to the materials in the case would assist the court in evaluating the likelihood of confusion. The practice has become widely known as "a witness collection programme". The witnesses are selected by the person seeking to call them. The selection is never intended to be a representative sample of all interviewees; it is a selection of those witnesses most likely to be most favourable to that party's case. In practice the survey is carried out by the claimant seeking to establish passing off or trade mark infringement. The defendant rarely does this because it is difficult to prove a negative.
64. A cynic might think that the phrase "witness collection programme" is simply a euphemism for adducing evidence from a skewed selection of witnesses identified by means of a statistically invalid and unreliable survey.
65. Even the more limited exercise envisaged by a "witness collection programme" is expensive and time consuming. That is why the practice has grown up of requiring directions to be sought before evidence generated by a witness collection exercise is admitted. The first recorded example is, so far as I know, the direction given by Pumfrey J in O2 Holdings Ltd v Hutchison 3G Ltd. Although not reported, it is referred to in a ruling I gave on the first day of the subsequent trial for trade mark infringement: [2006] EWHC 601(Ch); [2006] RPC 30. I ruled that without expert evidence to explain the survey sought to be tendered, the evidence of eight individual members of the public was not probative of anything I was required to decide. I therefore excluded that evidence. The point arose again in UK Channel Management Limited v E! Entertainment Television Inc. The admission of such evidence was considered in two stages. At the first stage the question was whether the evidence should be submitted to the court for the purpose of seeking permission to call it at all. That question came before Rimer J: [2008] FSR 5. He described the justification for the practice as follows:

"The reason is … because such evidence does not simply put before the court the spontaneous reaction of members of the public who have been exposed to … [a party's] advertising material in actual use or the defendant's allegedly offending advertisement in actual use; it is evidence obtained as a result of the artificial application to members of the public of stimuli directed at provoking reactions and responses. The reason for Pumfrey J's earlier direction was to ensure that their reactions and responses were provoked by appropriate questions."

66. The second stage was whether permission to call the evidence should be given. That question came before me: [2007] EWHC 2339 (Ch). There were three surveys in issue. The first was an omnibus survey; the second was called Mailvoire 1 and the third was called Malivoire 2. Both the latter two had been conducted under the guidance of Mr Malivoire, who also features in this case. I said of the omnibus survey that "the omnibus survey is of such little value that it is not worth the time and effort that would be expended in considering it" and therefore declined to admit it. I should point out that I did not decide that it was technically inadmissible. So far as Malivoire 1 was concerned I decided that the questions it posed were misleading and that answers (and subsequent evidence) stimulated by those questions were themselves likely to be misleading and tainted. So far as Malivoire 2 was concerned I concluded:

"Whether the survey proves anything at trial is, I think, a matter for the trial judge. Although I am sceptical about its value I do not feel so confident that the answer is that it proves nothing that I think it right to rule it out at this stage."

67. I therefore gave permission for Malivoire 2 to be used as the basis for a survey, together with permission for Mr Malivoire to give expert evidence to explain the methodology and results. I did not give permission to call a selection of witnesses identified by means of a statistically invalid survey. Thus far the practice had evolved through a series of extempore rulings either on interim applications or rulings in the course of a trial.
68. It was not until A & E Television Networks LLC v Discovery Communications Europe Ltd [2011] EWHC 1038 (Ch); [2011] FSR 31 that Mann J attempted to find the underlying rationale for this practice. He said (§ 8):

"In the case before me, as will appear, it is necessary to bear in mind the juridical basis of what it is that the court is doing when exercising its control. In my view it is doing (at least) the following:

i) So far as a party is going to seek to put expert evidence before the court, the court is exercising its power to control the amount and nature of expert evidence in order to make sure the expert evidence is proper evidence, admissible, and proportionate.

ii) So far as a party seeks to put in the actual answers to questions, the court is ensuring the evidence is admissible and probative.

iii) So far as the court is controlling the calling of live witnesses obtained as a result of some form of survey evidence (so-called witness collection exercises) it is again ensuring that the evidence is admissible and probative. In particular, it is acting to prevent a party seeking to call a witness whose evidence is going to be tainted to an unacceptable degree by the mechanism under which it is collected (an inappropriate question).

iv) In so doing, the court is ensuring that costs are not wasted and are proportionate. It is wrong for costs to be wasted in conducting hopeless surveys, for the other party to have to waste costs dealing with that evidence, and for court time to be wasted in dealing with it at trial.

v) When a court is acting in this capacity it must bear in mind that it is acting at some remove from the trial. If it disallows a survey it is concluding, short of a trial, that evidence which one party wishes to adduce should not be allowed in because it will be of no or insufficient value. In embarking on that exercise it must acknowledge that there will be cases in which it is not wholly clear that the evidence in question will be valueless. In those circumstances the right course may be not to bar the evidence or survey at the interim stage, but to allow it and to have more informed argument at the trial (or conceivably at another interim stage, provided that that is a cost-effective way of going about the matter)."

69. Most recently in Specsavers Kitchin LJ described the purpose of the practice as being:

"to avoid the spending of time and money on what is clearly irrelevant and unsatisfactory evidence."

70. Thus the purpose is not merely to avoid irrelevant (i.e. inadmissible) evidence; it is also to avoid evidence which is unlikely to be of real value.
71. In his skeleton argument Mr Silverleaf criticises Mann J's approach as too cautious. He said:

"In Interflora's submission, Mann J's judgment in this respect is over-cautious and should be rejected. Whenever a court gives permission to carry out a survey plus witness statements of respondents, it will not know what the respondents will say or what their evidence will say or what the CEA notices will say. Mann J's approach would turn all survey permission applications into an even more costly, two-stage process. First, permission would have to be sought to carry out a survey and secondly, witness statements would have to be prepared and adduced and a separate permission be sought in relation to those. There is no precedent for such an unwieldy approach and it seems contrary to the cost-saving objectives of the practice for seeking permission that was originally devised by Pumfrey J in the O2 case."

72. Instead, he commends the following practice:

"… there should be only a single application in relation to survey evidence at which the court considers the merits of the questions and methodology and rules on whether the survey should be permitted to be adduced in evidence once it has been carried out. That permission will necessarily carry with it permission to adduce evidence from selected respondents to the survey to give direct oral evidence of their experiences. Normally the application for permission to adduce survey evidence should be made, as envisaged by Pumfrey J in O2, following the conduct of a pilot survey so that the court is furnished not only with the proposed survey materials but the initial results of putting those materials into practice from a limited pool of respondents. It will, with that material, be in a much better position to assess the value of the survey which is proposed to be carried out."

73. I do not agree. In my judgment Mann J's approach, understandable because of the way that the practice has developed, is not cautious enough. I stress again that what is in issue is the effect on a hypothetical legal construct: the reasonably well-informed and reasonably observant internet user. In the absence of special circumstances, how can calling ten, twenty or thirty witnesses selected from a statistically invalid survey be extrapolated into the effect on that legal construct of the advertisement or sign in issue?
74. In my judgment the question is well and persuasively discussed by the Federal Court of Australia in Arnotts Ltd v Trade Practices Commission [1990] FCA 473 (1990) 97 ALR 555. The case was about competition. What was in issue was the extent to which other foods were substitutable for biscuits. One side commissioned a market survey to answer that question. The survey was based on interviews with 1200 people. The interviewees were selected at random, in the sense that the person interviewed was generally the person who answered the interviewer's door knock; but the residences to be visited were selected by reference to demographic factors, so as to be representative of the whole Australian population. In addition to putting in the survey itself, three of the interviewers (not the interviewees) were called to give evidence. Objection was taken to this evidence. The argument was that all 1200 interviewees should have been called. One basis for the objection was that the views of the interviewees were no more than hearsay evidence. I need say no more about that. But the court discussed the acceptability of survey evidence generally. They rejected two methods of proving consumers' habits and attitudes. They said:

"One theoretical possibility, in a case like the present, would be for a party to call such of the 1,200 respondents to the Roy Morgan survey as were contactable. This course would have the advantage of providing a fairly selected group of witnesses; subject to any distortion which might be caused by difficulties in locating respondents. But it would add enormously to the cost and duration of a trial.

The second possibility would be for a party to call evidence from a lesser number of selected witnesses. This course was taken in Ritz. The plaintiff there called 152 members of the public. The majority of these witnesses were stopped in a public place by a representative of the plaintiff and questioned as to the significance to them of the word "Ritz". It seems that those who gave answers favourable to the plaintiff's case were asked to give evidence. Those who did not, were not. As a result, the evidence of these persons was of negligible value. All that it established was that, with the expenditure of sufficient effort and money, 152 people could be found somewhere in Australia who claimed to associate the word "Ritz" with the plaintiff. The 152 witnesses were not a fair sample of the general public; so that, as McLelland J noted at 215, there was "no ground in the evidence for any extrapolation on a statistical basis, or on the basis of any mathematical or logical probability, of the views of the 'public' witnesses (or any selection from them) as representing the views of the relevant class of the Australian public or a significant section of that class". The plaintiff was not even willing to reveal the total number of persons interviewed; for all the judge knew, the persons who associated the word "Ritz" with the plaintiff may have been a tiny minority. The tender of such partisanly selected evidence was an absurdity.

A further possibility is that there will be no evidence about matters such as public recognition of names or attitudes to products. Perhaps this would not matter greatly. As the authorities emphasise, where the question is whether use of a particular name or trade mark would constitute misleading conduct or cause confusion, the court must make its own assessment of the situation. The position is similar where the question is whether one product is substitutable for another. But information is preferable to intuition. Where the state of public knowledge of, or attitudes to, some subject is a relevant factor in the court's adjudication of an issue, it is better to admit than to preclude evidence on those matters."

75. They concluded that survey evidence should only be admitted if the survey complied with criteria for reliability which closely resemble the Whitford guidelines.
76. The proposal in the present case is not quite as dire as the second possibility discussed in Arnotts because Interflora have disclosed all the answers to the questionnaires. But the remaining objections still hold good: viz. that the selected witnesses are not (or at least cannot be shown to be) a fair sample of the class of reasonably well-informed and reasonably observant internet users, with the consequence that there is no ground for any extrapolation on a statistical basis, or on the basis of any mathematical or logical probability, of the views of the selected witnesses as representing the effect of the M & S advertisement on the hypothetical reasonably well-informed and reasonably observant internet user. If evidence of this kind cannot form the basis for extrapolation on the basis of any mathematical or logical probability leading to a conclusion about the effect of M & S's advertisement on the hypothetical reasonably well-informed and reasonably observant internet user, then in the absence of special circumstances it cannot be useful. And if it cannot be useful, it should not be allowed to distract the focus of the trial even if it is technically admissible.

Previous cases

77. Mr Silverleaf referred us to a number of cases in tracing the history of the admission of survey evidence. They enable one to see whether, in the general run of case, evidence such as that which Interflora wishes to call has proved to be of value. I preface my observations by saying that the watershed case on the admission of survey evidence is Imperial Group plc v Philip Morris Ltd [1984] RPC 293 which laid down the guidelines for valid surveys which had not previously been considered in detail. Thus, in my judgment, cases decided before Imperial, even at the highest level, must be regarded with caution.
78. The first in time was the decision of the House of Lords in A Baily & Co v Clark Son & Morland [1938] AC 557. Both parties had for many years carried on business in the town of Glastonbury as sheepskin slipper manufacturers. One of them had obtained registration of the mark "Glastonburys" in relation to the sheepskin slippers on the basis of alleged distinctiveness. The other applied to expunge the mark. The first issue was whether the mark had acquired distinctiveness. An issue arose concerning a type of "survey" that was carried out designed to establish distinctiveness. Lord Russell described it thus:

"The appellants circularized 91 members of the public asking them to answer the following question: "If you were offered in the ordinary way of trade, slippers described as 'Glastonbury slippers,' what would the word 'Glastonbury' mean to you?" They received 83 replies. To save expense they only filed affidavits by 11 of those who had replied: but in order to prove that the 11 were a proper and fair sample of the bulk, the remaining 72 replies were exhibited to an affidavit. This procedure was the subject of severe comment in the Court of Appeal. No doubt it was technically wrong; but the appellants not unnaturally shrank from the great expense of filing some 70 affidavits. It will be wiser perhaps in a similar case in the future to file an affidavit merely stating the number of the other answers which have been received, and that they are open to inspection by the other side's advisers. The Court will then be protected from the embarrassment of being in possession of documents which are in no sense evidence of the facts stated therein."

79. The Court of Appeal had been critical of this way of proceeding. The principal objection seems to have been that all the respondents should have been called. This strict insistence on live evidence has been overtaken by the general use of hearsay evidence in civil proceedings. But what, I think, is of more interest is the value that the judges attributed to the evidence. Viscount Maugham LC presided in the House of Lords. He said:

"Moreover, I am of opinion that a great deal, if not the bulk, of the evidence relied on by the respondents was evidence of little weight. If a manufacturer in a particular town chooses for some years to use the name of the town as a trade mark in respect of his goods, and then (whilst still manufacturing there) proceeds to ask all his trade customers what the name of the town indicates to them, it is, I think, inevitable that many of them will reply that the name indicates to them the manufacture of the person from whom they have been in the habit of procuring the goods. The association of the goods with the place is not, in my opinion, thereby negatived. Accordingly, for my part, I attach small weight to the affidavits of the respondents' customers - all of them, or nearly all of them, framed in the same words - stating that the name of "Glastonburys," when used in connection with slipper-shoes and over-shoes, indicates the respondents' goods to the exclusion of those of all other manufacturers. I should add that it is plain that such a customer of the respondents can only give evidence as to what the name means to him and that he is not entitled to give evidence as to what the name would mean to any one else in the trade or to the public. Evidence as to the meaning which members of the public, actual or probable buyers of sheepskin slippers, would attach to the word is exceedingly scanty."

80. Lord Russell said:

"In the case of household articles in general use such as slippers, the meaning which the word in question conveys to the purchasing public is of vital importance. In the present case the direct evidence given by the public does not support the view of distinctiveness. No such direct evidence was tendered by the respondents. It is true that some of the trade witnesses testified that their customers ordered "Glastonburys" or "Glastonbury Slippers," and that in response to such orders they supplied the respondents' goods; but this evidence (in the absence of the customer's explanation) is subject to the criticism that such expressions may in fact only have been used as indicative of the place of manufacture."

81. Lords Thankerton and Macmillan agreed with both judgments. It seems to me, therefore, that although the evidence was admissible, it was regarded as being of little weight.
82. In the GE Trade Mark case Graham J at first instance ([1969] RPC 418) admitted survey evidence, but thought it of no weight. He did not criticise the way in which respondents to the survey had been selected; but he thought that the questions they had been asked were leading questions. It should be particularly noted that the evidence was adduced on the basis that it was (or at least purported to be) a statistically valid survey. In the Court of Appeal ([1970] RPC 339) the judges were more receptive. Salmon LJ began by looking at the question of confusion "from a broad common sense point of view", which included his own subjective impression. He went on to say:

"Allowing for every criticism that was made of the survey evidence by the American company's statistical experts, the fact remained that at least 12% to 35% of the adult population would think that the GE mark on domestic electrical goods signified that the goods were manufactured by the English company."

83. He said that the judge had "wrongly … discounted the survey evidence" but he did not really explain why he took that view. Winn LJ also began by addressing the question as "a matter of first impression, on a common sense approach." He did not refer to or rely on the survey evidence. Cross LJ first formed a view without reference to the evidence, and then considered whether the evidence confirmed or displaced that view. He said that it was strikingly confirmed both by the survey and the trade evidence. The survey evidence showed "a very large number of people would think that an article such as an electric toaster bearing the GE mark was made by G.E.C." On further appeal to the House of Lords ([1973] RPC 297) the question arose whether the judges should have formed their own view at all, or whether they should have relied only on the evidence. Lord Diplock said:

"But where goods are sold to the general public for consumption or domestic use, the question whether such buyers would be likely to be deceived or confused by the use of the trade mark is a "jury question." By that I mean: that if the issue had now, as formerly, to be tried by a jury, who as members of the general public would themselves be potential buyers of the goods, they would be required not only to consider any evidence of other members of the public which had been adduced but also to use their own common sense and to consider whether they would themselves be likely to be deceived or confused.

The question does not cease to be a "jury question" when the issue is tried by a judge alone or on appeal by a plurality of judges. The judge's approach to the question should be the same as that of a jury. He, too, would be a potential buyer of the goods. He should, of course, be alert to the danger of allowing his own idiosyncratic knowledge or temperament to influence his decision, but the whole of his training in the practice of the law should have accustomed him to this, and this should provide the safety which in the case of a jury is provided by their number. That in issues of this kind judges are entitled to give effect to their own opinions as to the likelihood of deception or confusion and, in doing so, are not confined to the evidence of witnesses called at the trial is well established by decisions of this House itself."

84. On the particular facts he said:

"In the instant case all three members of the Court of Appeal were alert to the danger of relying upon their own idiosyncratic knowledge and temperaments. All gave due weight to the evidence adduced upon the issue at the hearing and to the criticisms of it by the trial judge. All were of opinion that they themselves would have been likely to be confused by the use of the rondel mark simpliciter upon electrical consumer goods if, as had been the fact before the hearing of the appeal, they had known of the English company but had not known of the American company.

My Lords, I myself, as a member of the public in the United Kingdom, am idiosyncratic in that I have for many years known of both companies. I should not have been likely to be confused by the use of the rondel mark simpliciter upon electrical consumer goods. Nevertheless, my common sense convinces me that there would be a likelihood of confusion among a substantial number of members of the public who did not share my idiosyncratic knowledge."

85. It is notable that in coming to his own conclusion Lord Diplock did not rely on the survey evidence.
86. Lego System Aktieselskab v Lego M Lemelstrich Ltd [1983] FSR 155 was a passing off case. Trade mark infringement was not involved. The claimant was the famous toy brick company, and the defendant was an Israeli company specialising in garden equipment. The claimant relied on survey evidence to show that their brand was a household name. But it also relied on the survey, and witnesses identified as a result of the survey, to establish that a substantial proportion of the public would be deceived. Falconer J said of the survey that it was common ground that the sample was broadly in line with the age distribution, sex ratio and social grading of the adult population in general and, therefore, representative of the adult British population within the normal standards of market research categories. Of the individual witness called, the claimant selected from among those who were classified as "confused" for the purposes of the survey. It did not call any of the interviewees who were classified as "not confused" or "equivocal." The judge said of those witnesses:

"I accept that they did not in themselves constitute a representative sample of the whole adult population—but clearly they did constitute in themselves a substantial part of the whole representative sample of the survey and were, therefore, as Mr Aldous submitted, representative of a very substantial number of persons, members of the public."

87. As I read this, the reasoning proceeds in two stages. Stage 1 is the survey. That is representative of the population as a whole. Thus the picture revealed by the survey stands proxy for the whole population. However, since this was a passing off case, what needed to be proved was that a substantial proportion of the population would be deceived. Hence at stage 2, the selection of witnesses, it was sufficient to call witnesses who were representative of a substantial proportion of those surveyed. That was equivalent of a substantial proportion of the whole population for whom the survey stood proxy. In my judgment this case, properly read, does not support the cherry-picking of witnesses from a statistically invalid survey.
88. White Horse Distillers Ltd v Gregson Associates Ltd [1984] RPC 61 was another passing off case. The issue was whether the defendant whisky exporters had sold Scotch whisky in bulk to Uruguay for admixture with local spirits in circumstances that rendered them liable for passing off. The claimants put in evidence a street survey carried out by J Walter Thompson, Argentina, in Montevideo at 16 "strategic points" in the city. 327 people were interviewed. Respondents were shown a colour photograph of an advertising hoarding showing a bottle of the product with its name in large letters and the slogan "Ecoscés por crianza" (meaning "bred in Scotland".) The respondents were asked a series of questions about what they thought the product was, culminating in questions about whether they drank whisky and in particular knew of Gregson's whisky. Of the respondents, 110 said that they were willing to give evidence. Subsequently, there were 45 who agreed to give notarially attested statements. There was a survey report which expressed views about the percentages of responses that went one way or another. All this material was adduced in evidence under the Civil Evidence Act then in force. In addition, five of the deponents came to the trial and gave oral evidence on which they were cross-examined. In looking at the labels, Nourse J said:

"In many, perhaps the majority, of passing off actions which depend on the get-up of an allegedly deceptive product, it is inevitable that judges will in the end make their own assessment of the likelihood of deception, even in cases where there is extensive evidence of confusion."

89. As regards the survey evidence, he said that he accepted the submission that:

"…it would be very dangerous for me not to give full weight to the evidence as a whole in a case where what is in issue is a question off passing off in a distant country with whose people and culture the court is unfamiliar."

90. Whether he would have taken the same view in a case concerned with familiar consumer goods sold in England and Wales is highly debatable. This may be one of those (rare) cases in which the national court has particular difficulty in assessing the misleading nature of the statement or description in question.
91. As mentioned Imperial Group plc v Philip Morris Ltd [1984] RPC 293 dealt a body blow to the reception of survey evidence. Whitford J was trenchant in his criticisms of the survey evidence adduced in that case, and laid down guidelines for the conduct of future surveys. I have already referred to them.
92. Mr Silverleaf's next case was McDonald's Hamburgers Ltd v Burgerking (U.K.) Ltd [1986] FSR 45. This was another case of passing off. Burgerking advertised Whopper hamburgers on the London Underground with the words 'It's Not Just Big, Mac." McDonalds relied on three opinion polls in support of its case. 28 pollees were also called to give evidence. Whitford J quoted extensively from their evidence. He concluded that the evidence was "confirmatory of my own immediate impression on first being shown the Burgerking tube card, which was that it was an advertisement for a hamburger called "Big Mac." I, like all the witnesses, missed what was referred to as the pun, and it had to be explained". His ultimate conclusion on passing off was:

"My view, unaided by evidence, would be that this would be likely to be taken by people as an advertisement for a "Big Mac" hamburger and, in the light of the evidence of reputation, so far as a very large number of people are concerned, this tube card would be taken as advertising McDonald's "Big Mac" hamburger. Because of the prominence of the Burgerking logo it is, I think, quite plain that a by no means negligible number of people might be led into thinking that there is an association between "Big Mac" and Burgerking, and that they can get a "Big Mac" at the Burgerking establishments. Of course there are bound to be a number of people, it was a relatively small number of people in this case, who are going to go so far as to identify this advertisement as advertising a Burgerking "Whopper," but the evidence in this case quite clearly shows, as I would expect, that practically nobody is going to pay any attention to the rest of the small print at all. This advertisement is not going to be read as a knocking advertisement, and I myself never realised that it was intended to be an advertisement of this character."

93. It can hardly be said that the evidence, even in a passing off case, was decisive. It did no more than confirm the conclusion that the judge would have reached "unaided by evidence."
94. Scott v Nice-Pak [1988] FSR 125 is another case of passing off. It concerned the blue colour of baby wipe tubs. The claimant relied on survey evidence but both Walton J and the Court of Appeal held that the poll was so flawed that it had to be rejected in its entirety. Fox LJ said:

"In my view the poll was so flawed, for the reasons I have indicated, that it cannot be relied upon at all. I agree with the judge on that, though I do not think that he was justified in referring, as he did in certain passages of his judgment, to deceit or fraud. Nor, it seems to me in view of the incorrect premises upon which the poll was conducted as I have indicated, can one safely place any reliance upon the affidavits subsequently obtained by the plaintiff from the respondents. They are respondents who were subjected to an unsatisfactory poll and asked questions at the poll upon false premises, and those are matters which plainly affected their minds by the time they came to swear their affidavits. In the circumstances I would not myself be prepared to rely upon those affidavits at all."

95. Thus not only was the survey itself rejected; but even the evidence of witnesses "collected" by the poll was rejected. In other words the evidence was useless.
96. Unilever v Johnson Wax [1989] FSR 145 was a trade mark infringement case about lavatory cleaners branded, respectively LIFEGUARD and LIFEBUOY. The claimant relied on a survey and called a number of witnesses who had participated in the survey. Whitford J rejected both the survey and the evidence of the participants in the survey. Having cited from the speech of Lord Diplock in GE Trade Marks, Whitford J said:

"I am dealing with commonplace goods sold for domestic use to the public at large. Setting aside, I hope, any relevant idiosyncracies of knowledge or temperament, if I ever had any, I remain of the same opinion as that which I reached at the beginning. My first impression is unshaken. These marks do not so nearly resemble each other as to be likely to deceive or cause confusion when used on the same goods."

97. In other words both the survey and the witness evidence were useless.
98. More recently in esure Insurance Ltd v Direct Line Insurance Plc Arden LJ drew attention to a number of concerns. It should be particularly noted that Arden LJ was dealing with "traditional consumer surveys" and not with cherry-picked witnesses identified by means of a "witness collection programme". She said (§ 63):

"…consumer surveys are costly to produce. They can, moreover, sometimes be based on the wrong questions and thus produce irrelevant or unhelpful responses or for some other reason, as in this case, be of no evidential value."

99. Jacob LJ agreed and pointed out that survey evidence often failed to prove anything.
100. The first of the so-called "witness collection surveys" that we were referred to was Reckitt & Colman Products Ltd v Borden Inc (No 3) [1990] RPC 341 (the well-known Jif lemon case). The claimant sold lemon juice packaged in a yellow squeezy container in the shape of a lemon. It was sold with a label attached to its neck. The label bore the word "Jif" in prominent lettering. The defendant proposed to market lemon juice also packaged in a yellow squeezy container in the shape of a lemon. Three versions of the defendant's packaging were before the court. Each was proposed to be sold with a label attached to the neck of the plastic container. The prominent wording on each of these labels was "ReaLemon" and "Lemon Juice". The action was one for passing off; not trade mark infringement. Walton J at first instance heard evidence from a marketing expert whose evidence was (in effect) that some brands are instantly recognisable from the shape of the packaging, even if all wording or surface graphics are blanked out. These included the Perrier bottle, a Coca Cola can, the shape of the label on a tin of Heinz baked beans, the Jif lemon (without its label) and a packet of Marlboro cigarettes (which Walton J rejoiced that he could not recognise). Walton J described this as "entirely independent evidence to the effect that what the housewife looks for is the lemon shape and nothing else to tell her the product is Jif". In addition to this evidence he also heard evidence from consumers gathered as a result of what he called "a series of experiments". He described them as follows:

"These basically took three forms. As far as the Mark I lemons were concerned, shoppers were stopped in the street and asked what the Mark I lemon, which they were shown, was. A very considerable proportion said it was a Jif lemon, either spontaneously, or after having initially replied that it was lemon juice, when they were asked whether it was any particular brand. As far as the Mark II and III lemons were concerned, there were rather longer interviews in a more relaxed atmosphere in a room in a hall or public house or similar, with again roughly the same results. There was a third experiment, in which on pancake day (which is a day which naturally accounts for a large number of the sales of Jif lemon) the Mark III was prominently displayed in some co-operating supermarkets, in the trays which were proposed for its packaging, and once again the defendants' proposed lemon was picked up unhesitatingly by a large number of shoppers under the impression that they were buying Jif."

101. He concluded that the evidence established "beyond the slightest peradventure that the effect of the introduction of any of the defendants' lemons onto the market would be bound to result in many housewives purchasing them in the belief that they were obtaining the well known and liked Jif brand."
102. Walton J acknowledged that such evidence can be unreliable. However, on the facts he said:

"But in the present case firstly, both sides have presented such evidence. Second, the plaintiff tendered for examination by the defendants the whole segment of witnesses, starting with the professional organisers of their surveys, through the persons who conducted their interviews down to a fair sample, so far as it was possible for the court to judge, of the persons who had been interviewed. These persons ranged from fairly humble housewives, doubtless noted for their kind hearts and passion for Jif lemon juice with pancakes, but with no other great claim to fame, to some extremely intelligent shoppers, who, immediately they were put on notice that there might be some doubt as to the product they had purchased, did not require very long to appreciate the exact position of things.

Third, the plaintiff supplied the defendants with the names of their, or at any rate, some of their, potential witnesses, and the defendants did indeed, in the interlocutory stages of the first action, interview them. They did not interview any of such potential witnesses in relation to the second action. As a result of the interviews that were carried out, there was no complaint whatsoever as to the views of those witnesses having been misrepresented, or exaggerated, in the slightest."

103. In addition to the fact that this was a passing off action, four points should be particularly noted. First both sides called survey evidence from consumers, so the question of principle cannot have been in issue. Second, the claimant called the professional organisers of the survey, so this was more than a mere witness collection exercise. Third, the judge was satisfied that those witnesses who had been called were "a fair sample" of the interviewees; not those most favourable to the claimant's case. Fourth, the really important point that the evidence established was that shoppers did not look at the label before buying the product. Thus the consumer evidence corroborated the independent expert evidence about the way in which consumers selected lemon juice. This is, I think, borne out by later passages in the judgment; viz:

"But, at the least, I feel that I am now fully instructed in the way in which shoppers who shop, as the vast majority of us willy nilly have to shop, in self-service stores of some description, do so shop, and of what their apprehension of the Jif lemon really is."

104. He then set out the facts that, in his view, the evidence had established. They were:

i) Jif was and had for a long time been the only lemon-shaped lemon-sized squeezy pack of lemon juice on the market;

ii) Jif as a specific make of lemon juice produced by one particular proprietor was well known among shoppers generally;

iii) shoppers generally were well aware of the existence of various other brands of lemon juice;

iv) the crucial point of reference for a shopper who wished to purchase a Jif squeezy lemon was the lemon shape itself. Virtually no, if any, attention was paid to the label which that lemon bore;

v) lemons in squeezy containers were bought by consumers who used only a small quantity of lemon juice. Anybody with a requirement for a larger quantity bought a bottle, which was better value. The result was that purchases, though made steadily, were made at some little interval. During the whole of this interval, the product which had been used consisted of an unadorned squeeze pack lemon. That reinforced the position that when the consumer went to buy another such lemon, the starting point of the reference was the unadorned lemon, and not the lemon plus label.

105. What is striking is that Walton J did not include in his list of critical findings any finding of confusion, likelihood of confusion or deception. Instead, he continued:

"Now, paying proper attention to all these matters, and placing myself in the position of a shopper in relation to whom all these matters apply as part of his or her shopping knowledge and habits, I have really no hesitation in coming to the conclusion that there is bound to be confusion in the shopper's mind in relation to all three marks of the defendants' lemons. None of them is really sufficiently distinctive, nor are the labels such as to impinge sufficiently forcefully upon the shopper's attention, as to call immediately to mind that the item is not a Jif lemon: it would be supposed by a very large number of shoppers—probably, on any attempted arithmetical calculation running into millions—that each of the defendants' lemons was no more than an immaterial variant of the Jif lemon."

106. In other words the evidence went to the shopping knowledge and habits of the shopper; not to whether there was a likelihood of confusion. The latter question was one for the court, having informed itself of what consumers knew. That process is the same as that identified by Chadwick LJ in the Bach Flower Remedies case ("The task for the court is to inform itself, by evidence, of the matters of which a reasonably well informed and reasonably observant and circumspect consumer of the products would know; and then, treating itself as competent to evaluate the effect which those matters would have on the mind of such a person with that knowledge, ask the question").
107. In the Court of Appeal Slade LJ treated as the only relevant finding of fact the finding that "a substantial proportion of potential purchasers who were not indifferent to the manufacturing source of small plastic squeeze lemons would pay little or no attention to the label, because they would assume that they emanated from the manufacturer of JIF lemons, which had up to that time been the only product of that type on the market". He also noted that there had been no objection to the introduction of the evidence at trial and noted further that the claimant:

"… did not in this court rely on it as constituting more than corroboration of Colmans' other evidence as to the likelihood of confusion. I do not think that in order to satisfy this court as to the likelihood of confusion Colmans need to rely on the market survey evidence at all."

108. On further appeal to the House of Lords Walton J's findings of fact were described as at first sight surprising but unassailable.
109. I have already mentioned Mr Silverleaf's next case, Dalgety Spillers Foods Ltd v Food Brokers Ltd [1994] FSR 504. Blackburne J was very sceptical about the utility of consumer evidence but said that he could not rule it out at an interlocutory stage. Wagamama Ltd v City Centre Restaurants Plc [1995] FSR 713 was a case of both trade mark infringement and passing off. The defendant had opened a restaurant called "Rajamama" or "Raja Mama's", which was said to amount to infringement of the claimant's mark WAGAMAMA and also to amount to passing off. Laddie J heard evidence from a number of people who had eaten at Wagamama's restaurant. He said:

"A judge brings to the assessment of marks his own, perhaps idiosyncratic, pronunciation and view or understanding of them. Although the issue of infringement is one eventually for the judge alone, in assessing the marks he must bear in mind the impact the marks make or are likely to make on the minds of those persons who are likely to be customers for goods or services under the marks. Not all customers are the same. It is therefore sometimes of assistance for the court to hear evidence from witnesses who will help him to assess the variety of ways in which members of the target market will pronounce the marks and what, to them, will be the visual or phonetic impact of the marks. When considering infringement it is also necessary to bear in mind the possible impact of imperfect recollection on the part of members of the target market."

110. Having said that whether there was a trade mark infringement is "more a matter of feel than science" Laddie J concluded:

"I have come to the conclusion that the defendant's mark, in either form, is so similar to the plaintiff's registered mark that in use there exists a substantial likelihood of confusion on the part of the relevant public. That confusion is likely to take the form that some members of the public as a result of imperfect recollection will think the marks are the same while others will think that they are associated in the sense that one is an extension of the other …or otherwise derived from the same source."

111. He held that the claim succeeded. He reached his conclusion without any reference to evidence from consumers. Laddie J then turned to the claim in passing off. It was in that context, and in that context only, that he considered the evidence produced as a result of the witness collection exercise. He began by saying:

"As is usual in passing off cases, the plaintiff produced a number of members of the public to prove that confusion was likely. No attempt was made to choose a group of people who would represent a reasonable cross-section of the population of England or London. The plaintiff did not have the financial resources to engage in that sort of exercise. Instead its solicitors attempted to contact all those customers who at one time or another had left their name and telephone number at the WAGAMAMA restaurant. On the evidence, no attempt was made to skew the results." (Emphasis added)

112. The evidence was not seriously challenged, and Laddie J said that it was consistent with what he would have expected. I do not consider that this case supports Mr Silverleaf's submission that evidence from consumers is necessary or desirable in cases of trade mark infringement.
113. I have already dealt with Neutrogena, which was Mr Silverleaf's next case. As I have said it was common ground in that case that the passing off claim and the claim of trade mark infringement stood or fell together. I have explained why, in my judgment, what is probative evidence in a passing off case does not necessarily determine what is probative evidence in a case of trade mark infringement. It is also pertinent to note what evidence of confusion was in fact called. First there was unprompted and unsolicited evidence of confusion caused by an advertisement for Neutralia which many members of the public had thought was advertising Neutrogena. Second there was evidence from consumers identified by means of an experiment. But the nature of the experiment is also relevant. What happened was that a "buy one get one free" board was set up in 22 chemists shops. What it said was that anyone who bought a Neutrogena product would get another one free. The experimenters waited to see which customers picked up which products. As Jacob J pointed out this kind of experiment was "rather different from an ordinary market survey" in which people are interviewed according to a set pattern of questions. To put it slightly differently the stimulus that prompted the reaction was not a directed one. Accordingly, I do not accept Mr Silverleaf's submission that Neutrogena "is clear authority for, and endorsement of, the practice that Interflora contends for in this appeal and is wholly supportive of Arnold's J's approach in admitting the evidence of witnesses from the first Pilot survey." For the same reason I do not accept his submission that Chocosuisse supports his argument.
114. United Biscuits (UK) Ltd v Asda Stores Ltd [1997] RPC 513 was a claim both for passing off and for trade mark infringement. The claimant produced a chocolate coated biscuit called "Penguin" and the defendant launched a rival own label product called "Puffin". The claim in passing off succeeded, but the claim for trade mark infringement failed. Robert Walker J heard evidence from witnesses identified as a result of a "witness collection exercise". He concluded that although this evidence was selective and unrepresentative, it confirmed his own first impressions. He recorded his conclusions on passing off as follows:

"Judicial first impression is of some importance, especially where the goods in question are familiar and inexpensive (as opposed to the sort of specialised designer label goods with which Sir Nicolas Browne-Wilkinson V-C was concerned in Guccio Gucci v Paolo Gucci [1991] FSR 89… ). My impression of the puffin packaging (in all four colours or varieties) is that it would cause a substantial number of members of the public to suppose that there was a connection between the Puffin biscuit and the Penguin biscuit. Despite the evidence of isolated mistakes, I do not consider that a substantial number would believe that the Asda Puffin is the McVities Penguin. But many would believe that the two must be made by the same manufacturer.

That judicial first impression was largely confirmed by the oral evidence. Witnesses repeatedly referred to the dark seabird, the black lettering and the similarity of names as the features which together induced that belief. One or two came close to saying that the Puffin biscuit must be produced by the manufacturer of the Penguin, because otherwise it would not be allowed (there is here some hint of a self-fulfilling prophecy, and this point must not be given any more weight than it merits)."

115. The claim that the trade mark had been infringed failed; and the judge decided that part of the case without reference to the evidence from consumers. The evidence from consumers was thus only of any utility in relation to the passing off claim, and even then it did no more than confirm his own first impression.
116. Daimlerchrysler AG v Alavi [2001] RPC 42 was another case of both passing off and trade mark infringement. The claimant was the proprietor of a number of registered trade marks including MERC. The defendant sold clothes and shoes in Carnaby Street, some of which bore the legend "Merc". Pumfrey J heard evidence from witnesses identified by means of a survey. He said:

"During the course of the proceedings, a survey was conducted on behalf of DaimlerChrysler. The results of this survey were not relied on to show confusion and deception in the minds of the public caused by Mr Alavi's use of MERC, but to identify particular members of the public whose evidence would, in the claimant's view, be of assistance in deciding the action. This employment of the survey results accords with what Jacob J. thought was the right course in the Neutrogena/Neutralia case … It is of much more value to hear evidence of the public than to see imperfect records of unsupervised interviews. In this case there was a preliminary survey by White & Case, DaimlerChrysler's solicitors, and then a more complete survey by NOP. The White & Case survey covered about 240 members of the public, who were shown three of the defendant's T-shirts. It should be remembered that the public were shown the T-shirts, not in a commercial context, but in the possession of an interviewer in the street. Such an approach will of course excite trains of thought or speculation, which is why such surveys are most use in identifying potential witnesses who are prepared to be cross-examined. In response to "What are your first thoughts on this?" and "anything else?" if an uninteresting answer was given, about 20 or so interviewees referred to the claimant, and one referred to the defendant."

117. Six of the respondents gave witness statements and four were cross-examined. Their evidence went to the claim in passing off. Pumfrey J concluded that "The evidence from members of the public failed to give any convincing evidence of confusion". Pumfrey J returned to the evidence when considering trade mark infringement, and in particular infringement under section 10 (2). He said:

"Absent the evidence to which I have referred which has satisfied me that some people use the word Merc as an abbreviation or nick-name for Mercedes Benz, I should have regarded the allegation of infringement of the U.K. Clothing Marks under sub-section 10(2) as hopeless. What I have to consider is whether there is any likelihood of what I can call "spillover" of this use of MERC in respect of the clothing marks, having regard to the way in which the clothing marks are otherwise used. So far as Mr Alavi's shop is concerned, there obviously is no spillage. So that use is unaffected. This leaves the use on the clothing itself. In my judgment, Mr Anderson's thought processes, and those of Mrs Beaumont, were not those which are relevant to a customer for the claimant's goods. They are not shown to be in any way characteristic of the public, and are plainly, it seems to me, deductive. In my view, the possibilities of such spillover are very slight. I do not think the evidence can justify a finding that there is a likelihood of confusion between the defendant's mark and the U.K. Clothing Marks."

118. In other words, the evidence could not stand proxy for the average consumer; and was of no assistance. It is also to be noted that the question Pumfrey J posed was whether the witnesses called could be said to have thought processes "characteristic of the public". It is, I think, clear that he considered that the thought processes of one or two individuals were not probative of anything, unless it was possible to extrapolate from them.
119. Julius Sämaan Ltd v Tetrosyl Ltd [2006] EWHC 529 (Ch); [2006] FSR 42 was another case of both passing off and trade mark infringement. The products in question were car air fresheners in the shape of a conifer tree. Kitchin J first formed his own view about whether the defendant's sign was confusingly similar to the claimant's mark before turning to the evidence from consumers. As Mr Silverleaf rightly points out the evidence was called from witnesses identified as part of a witness gathering exercise and not from a statistically reliable survey. However, no objection seems to have been taken to the admission of this evidence. Kitchin J said that the claimants:

"…were not attempting to rely upon the survey as such but rather as a witness gathering exercise. They relied not upon the survey but upon the evidence that the witnesses so identified actually gave in their witness statements and under cross examination. This is certainly a permissible approach and it is one that was followed in United Biscuits (UK) Ltd v Asda Stores Ltd [1997] R.P.C. 513. Further, as was said in Neutrogena Corp v Golden Ltd [1996] R.P.C. 473 at 485 to 486, questionnaire evidence alone is seldom helpful and frequently unnecessarily elaborate and, I would add, very expensive. Unless the court has some real evidence, tested in cross examination, it is very difficult to determine what the respondents were actually thinking. However, it is also important that the court is given the full picture of any witness gathering exercise so that the court can assess whether the witnesses actually called truly reflect the views of the average consumer. In the present case I have not been given a complete picture because I have no evidence as to how the respondents or locations were selected. However, it is fair to say that all the questionnaires have been disclosed and I have been able to make an assessment of the respondents upon whom the claimants have chosen to rely from the cross examination."

120. Both United Biscuits and Neutrogena were, of course both passing off cases. As I have said, in United Biscuits the consumer evidence played no part in the trade mark claim and in Neutrogena it was agreed that the two claims stood or fell together. Having considered the consumer evidence Kitchin J concluded (§ 75):

"I have considered all of the evidence of the respondents who came to court. I must be cautious in drawing conclusions from them for all the reasons I have given. They were, however, all witnesses who gave their evidence carefully and fairly and do, in my judgment, give me some idea of the way that the average consumer would perceive the products in issue. I believe they support my impression that there is a real likelihood that the average consumer would, on seeing the Christmas Tree product, think that it is another product in the Magic Tree range or a Christmas version of the Tree products with which he or she is familiar."

121. This, then, is a case in which although the evidence was considered in the case of trade mark infringement, without objection, again it did little more than confirm the judge's independent conclusion.
122. D Jacobson & Sons Ltd v Globe UK Ltd [2008] EWHC 88 (Ch); [2008] FSR 21 was a case about trainers. In his skeleton argument Mr Silverleaf said that this was another case of a "witness collection exercise". I do not think that it was. The claimants presented in evidence a survey conducted by a market research company called MDL, in which over 1,000 people had taken part, and vouched for by an expert witness (Mr Blackett) who confirmed its validity. It survived attack by the defendants. Etherton J said (§ 93):

"In all the circumstances, I consider that Mr Blackett's evidence on the survey is admissible, notwithstanding that much of it is based on what he was told by MDL. In so concluding, I take into account all the matters I have mentioned, as well as Mr Blackett's unchallenged professional endorsement of MDL's expertise, Mr Blackett's personal collaboration with MDL in connection with the survey, his own marketing experience and expertise, and his knowledge and understanding of the RTR and the appendix to it."

123. In addition to the survey the claimants called a number of witnesses who had participated in the survey. The survey itself was of considerable help to Etherton J on the question whether the mark was distinctive: see §§ 142, 144. He also found it helpful on the question of confusion: see § 181. This, then, was a case in which the "live" evidence was called in order to amplify the findings of a statistically reliable survey.
124. Whirlpool Corp v Kenwood Ltd [2008] EWHC 1930 (Ch), [2009] RPC 2 is a decision of Mr Hobbs QC himself, sitting as a judge of the Chancery Division. The claim was a claim for infringement of the claimant's registered mark. The claimants carried out a number of public surveys with a view to establishing both the distinctiveness (in the trade mark sense) of the Whirlpool KitchenAid Artisan mixer and that the shape and appearance of the defendant's kMix mixer was confusingly similar. Some 600 questionnaires were obtained with 164 providing sufficient contact details. Witness statements were obtained from 23 interviewees. Others of the 164 interviewees were contacted but no witness statements were obtained. The screening of the interviewees was carried out by the claimants' legal advisers and litigation privilege was claimed in respect of that aspect of their work. The process was referred to as a 'witness gathering exercise'. No attempt was made to claim statistical significance for the survey as a whole or to the 23 witness statements. Mr Hobbs QC said (§ 35):

"The need for circumspection in the assessment of questionnaire survey evidence is well-understood. The concern is that to a greater or lesser degree: "Interviews and questionnaires intrude as a foreign element into the social setting they would describe, they create as well as measure attitudes, they elicit atypical roles and responses, they are limited to those who are accessible and will co-operate and the responses obtained are produced in part by dimensions of individual differences irrelevant to the topic at hand". There is, accordingly, a practical requirement for information relating to the structure, method and results of questionnaire surveys to be full enough to enable the strengths and weaknesses of the research work to be evaluated. It is now the practice for the utility of any proposed survey work to be considered as a matter of case management during the interim stages of actions for infringement of registered trade mark and passing off.

36 If the research work provides no sufficient or proper basis for extrapolation, the responses of individual interviewees can really only be taken into account for what they may individually be thought to be worth. That may be little or nothing."

125. He then referred to Arnotts Ltd v Trade Practices Commission, which I have already quoted. Although Mr Hobbs said that empirical evidence could not be ignored (§ 70) he nevertheless came to the conclusion that the risk of confusion had not been established (§ 71). It is quite clear that he regarded that evidence as being of very little value.
126. Dismissing an appeal against Mr Hobbs' decision ([2009] EWCA Civ 753; [2010] RPC 2) Lloyd LJ revisited the evidence and said that it was inadequate to show that "the hypothetical (or paradigm) reasonably well-informed and reasonably observant and circumspect consumer" would have been confused. As Mr Silverleaf rightly observes there was no suggestion that the evidence was inadmissible. On the contrary both sides invited the Court of Appeal to review it. But Lloyd LJ does make clear that the court's concern is with a paradigm or hypothetical person; not with the selection of witnesses actually produced in court.
127. Numatic International Ltd v Qualtex Ltd [2010] EWHC 1237 (Ch); [2010] RPC 25 was a quia timet action in passing off based on the get-up of the 'Henry' vacuum cleaner. Floyd J heard evidence from witnesses selected from a witness gathering exercise. He said:

"A number of authorities make it clear that great care needs to be taken with survey evidence obtained from members of the public in passing off actions: see e.g. Scott Ltd v NicePak Products Ltd [1989] FSR 100. In particular, if the object of the exercise is to establish consumer reaction to seeing a product for the first time, it can be misleading to suggest to the interviewee, by the form of the question, that the product is one which they could already have seen on the market: see per Lewison J in UK Channel Management Ltd v E! Entertainment Television Inc [2007] EWHC 2339 at [9] to [12] and per Geoffrey Hobbs Q.C. (sitting as a deputy judge of the Chancery Division) in Whirlpool Corp v Kenwood Ltd [2009] RPC 2 at [39]."

128. He also added that "one must beware of witness gathering exercises which are founded on surveys of this kind, because evidence gathered in this way is likely to be tainted".
129. The claimant tendered in evidence a survey that had been conducted by Mr Malivoire, involving 535 respondents, of whom 25 gave witness statements. Ten of them gave oral evidence. A number of criticisms were levelled at the survey that had been used to identify the witnesses. Floyd J accepted some of those criticisms but nevertheless said that the survey provided him with enough evidence to be confident that the shape of the defendant's prototype prompted a strong recognition of the Henry. He said however that whether that recognition was sufficiently strong to amount to deception was a different question. As to that he preferred to rely on the evidence of those witnesses who had actually been called. He concluded (§ 81):

"In the end I felt that the evidence of the members of the public provided support for the view that there would be origin confusion if the Qualtex replica was sold in the way in which Qualtex threatened, even taking into account the way in which their evidence had first been obtained. The overall impression given by the evidence was that the replica prototype did convey quite a strong message that it was a genuine Henry, and that there was a real likelihood that at least some members of the public would buy it thinking it was a Henry. I have to say that this evidence confirms my own impression, for reasons which I explain below."

130. Once again, therefore, this was a case in which the evidence simply confirmed the judge's independent conclusion. Moreover, as noted, it was a case of passing off.
131. Specsavers International Healthcare Ltd v Asda Ltd [2010] EWHC 1497 (Pat); [2010] FSR 28 was a case of both trade mark infringement and passing off by (among other things) logos. In the course of the proceedings Mann J had to rule on the admissibility of evidence. Asda tendered witness statements from 11 persons who were shoppers in its stores. It had selected the 11 witnesses by approaching 200 people in its stores and asking them very limited questions, the principal one of which was to ascertain whether they would be willing to attend court to give evidence on a matter not then closely identified. Only about 20 indicated they were prepared to give evidence and, of those, only 11 were finally prepared to give evidence and sign witness statements. In the last few paragraphs of each witness statement, the witness gave evidence as to whether they were confused by the logos in question. Mann J said that although this was not survey evidence in the sense that it was being tendered to support some form of statistical analysis, it was nevertheless the extraction of evidence whether directly from a witness in the witness statement or indirectly through the results of a survey by the prompted reaction to what is described as stimuli; and that the witnesses were tendered as being reasonably representative of the public. He held that he had a discretion whether to admit the evidence. Although Mann J thought that the evidence was of "little real value" he was nevertheless prepared to allow it to be called. In the judgment he delivered after trial he said ([2010] EWHC 2035 (Ch); [2011] FSR 1 § 107):

"I ruled that it was survey evidence, but let it in in my discretion. I observed at the time that it did not seem strong evidence. Having heard all the evidence and submissions in this case I maintain that view. It proves, of course, that 11 members of the public were not confused, but Mr Bloch accepted that not all members of the public would be confused. We know that 11 people out of the 600 interviewed expressed that view. They were not selected because they held that view; they were "selected" as being the only people who, at the end of the day, went so far as to agree to give evidence in court. But they are only a small part of the number of people originally approached, and there is no indication of the views of the others. We do not know how representative they really are. All in all, therefore, this is evidence which does not help much as a piece of evidence by itself."

132. Two further witnesses were called to give evidence but Mann J said that their evidence could not stand proxy for the reasonably well informed, reasonably observant and circumspect consumer (§ 113). On appeal this court upheld the judge's conclusion on that evidence. Kitchin LJ commented on the unsatisfactory nature of it ([2012] EWCA Civ 24; [2012] FSR 19 §§ 105-113).
133. In 32Red plc v WHG (International) Ltd evidence from four members of the public was called. Henderson J held that the impugned sign was an infringement of the registered mark. He said (§ 102):

"I reach this conclusion without needing to rely on any evidence of actual confusion, although a certain amount of such evidence was tendered by 32Red and I will consider it in the next section of this judgment. The test laid down by Art.9(1)(b) does not require evidence of actual confusion, and if such evidence is available it is likely to be of a confirmatory nature rather than an essential ingredient of the value judgment that, in the words of Jacob LJ in the Reed case, the court has to draw from all the circumstances."

134. The final case in this review is the recent decision of Arnold J in Starbucks (HK) Ltd v British Sky Broadcasting Group plc [2012] EWHC 3074 (Ch). That was a case of both passing off and trade mark infringement. The claimants called evidence from four customers. Although they were called principally to show that the relevant claimant had protectable goodwill, each of them expressed a view on the likelihood of confusion. Arnold J pointed out a number of defects in their evidence (§§ 13-17) and concluded that he could give the evidence no weight (§ 140). In other words, it was useless.
135. The upshot of this review is that courts have allowed the calling of evidence of the kind that Interflora wishes to call and have considered it, either in conjunction with or in the absence of a statistically valid and reliable survey. But it is generally of little or no value. Sometimes it does no more than confirm the conclusion that the judge would have reached without the evidence. In passing off cases it sometimes has greater effect, but as I have said more than once, passing off raises a different legal question. Unless the court can be confident that the evidence of the selected witnesses can stand proxy for the persons or construct through whose perception the legal question is to be answered it simply represents the evidence of those individuals. In a case in which the witnesses are called in order to amplify the results of a statistically reliable survey their evidence may be probative. But unless the court can extrapolate from their evidence, it is not probative.
136. Mr Silverleaf argued that if we acceded to Mr Hobbs' submissions then evidence from consumers would never be admitted in a case of trade mark infringement in the absence of a statistically valid and reliable survey. I do not think that follows. One of the objections to the witness collection exercise, as Rimer J pointed out in UK Channel Management is that the evidence thus collected is not the spontaneous reaction of members of the public who have been exposed to the allegedly infringing sign or advertisement, but is evidence obtained under artificial conditions by applying artificial stimuli. If there is evidence of consumers who have been confused in the real world, there can be no objection to calling it. He also submitted that the calling of evidence from witnesses identified by means of a witness collection exercise was quite independent of the process by which they had been identified. That may be so in some cases; but in the general run of cases where witnesses have been identified by a tailored series of questions, they will have been led towards a particular mindset which no longer represents the unstimulated evidence of people in the real world.
137. That is not to say that there can never be evidence called in a case of trade mark infringement. The court may need to be informed of shopping habits; of the market in which certain goods or services are supplied; the means by which goods or services are marketed and so on. In addition I must make it clear, however, that different considerations may come into play where:

i) Evidence is called consisting of the spontaneous reactions of members of the relevant public to the allegedly infringing sign or advertisement;

ii) Evidence from consumers is called in order to amplify the results of a reliable survey;

iii) The goods or services in question are not goods or services supplied to ordinary consumers and are unlikely to be within the judge's experience;

iv) The issue is whether a registered mark has acquired distinctiveness; or

v) Where the cause of action is in passing off, which requires a different legal question to be answered.

138. Outside these kinds of cases there may be others where a judge might think that it would be useful to hear from consumers. I would not wish to rule out the possibility. So I would not accept the proposition that evidence from respondents to a questionnaire can never be called in the absence of a statistically valid and reliable survey. But (apart from those I have mentioned) the cases in which that kind of evidence might be of real use are difficult to imagine. I would not therefore hold that such evidence is inadmissible as a matter of law.
139. However, it does not follow that if evidence is technically admissible, the court in civil proceedings must admit it. CPR 32.1 provides:

"(1) The court may control the evidence by giving directions as to –

(a) the issues on which it requires evidence;

(b) the nature of the evidence which it requires to decide those issues; and

(c) the way in which the evidence is to be placed before the court.

(2) The court may use its power under this rule to exclude evidence that would otherwise be admissible."

140. CPR Part 1.4 provides:

"(1) The court must further the overriding objective by actively managing cases."

141. This is a positive duty placed on the court. CPR Part 1.4 (2) goes on to say that active management of cases includes:

"(h) considering whether the likely benefits of taking a particular step justify the cost of taking it"

142. This, I think, was the step that Arnold J felt obliged to omit. Once he had decided to reject the submission that the evidence proposed to be tendered was inadmissible and would not be probative, he allowed it to be admitted without further ado. However, I would accept Mr Hobbs' submission that even at an interim stage a judge who is asked for permission to adduce such evidence should evaluate it carefully in order to see (a) whether it would be of real utility and (b) whether the likely utility of the evidence justifies the costs involved.
143. In the general run of cases it seems to me to place an undue and unfair burden on the other party for one party to tender in evidence witness statements from selected respondents to a questionnaire without even undertaking to produce a selection that demonstrates the full range of answers to the questions. As Mr Hobbs said, that places the burden on the defendant to disprove the validity of the selection, rather than on the claimant to validate it. Such a burden could in my judgment only be justified if the party tendering the evidence can show that it is likely to have a real impact on the outcome of the trial. In addition the cost of undertaking even a witness collection exercise and any subsequent investigation or follow up is extremely expensive. Mr Hobbs told us (without contradiction from Mr Silverleaf) that the cost of the exercise in the present case would exceed £250,000.
144. The current practice, which Arnold J understandably followed, is to allow the evidence in unless the judge can be satisfied that it will be valueless. In my judgment that is the wrong way round. I consider that, even if the evidence is technically admissible, the judge should not let it in unless (a) satisfied that it would be valuable and (b) that the likely utility of the evidence justifies the costs involved.
145. It follows, in my judgment, that the approach that I took in UK Channel Management Limited v E! Entertainment Television Inc (and followed by Mann J in A & E Television Networks LLC v Discovery Communications Europe Ltd §8 (v)) should no longer be followed.
146. In the present case I do not consider that Interflora has demonstrated that the evidence it wishes to call would be of real value. To put it bluntly, Interflora starts with an unreliable dataset from which it proposes to select the witnesses most favourable to itself. I would hold, therefore that Mr Hobbs' macro objection is well founded. I would therefore allow the appeal on that basis.
147. There was some debate before us about the procedure that should be followed. First, it is clear that the court cannot make any order without some material on which to base its decision. Thus there can be no objection to the carrying out of a true pilot survey, at the risk as to costs of the party carrying it out, before applying for permission to adduce the results of a survey. But that pilot survey will be no more than a basis for a further survey. Second, an application to admit survey evidence or evidence from respondents to a survey (or pilot survey) should be made as early as possible in the course of case management. It would not be right to leave it to the time when witness statements are exchanged. The objective of such an application is to have a definitive ruling one way or the other. It is a natural temptation for a judge who is not immersed in the case to leave questions of admissibility to trial. It is the temptation to which I succumbed in UK Channel Management Limited v E! Entertainment Television Inc. But balancing the cost of a survey (or witness collection exercise) against its likely utility, this temptation should be resisted. Third, the form of order that has evolved provides that:

"…neither party has permission to adduce survey evidence without first having obtained the leave of the Court. Any application for such leave is to include details of any questions proposed to be used in any such survey and details of the method and procedures proposed to be adopted in relation to the conduct thereof."

148. It is, in my judgment, doubtful whether this form of order catches a witness collection exercise. After all, Interflora does not wish to rely on the survey evidence of questionnaires. It wants to rely only on the evidence of the selected witnesses. In Specsavers International Healthcare Ltd v Asda Ltd [2010] EWHC 1497 (Pat); [2010] FSR 28 Mann J held that although a witness collection exercise might not fall within the literal effect of the order, it nevertheless fell within the vices which such an order was designed to eliminate. His decision in this respect was upheld by this court: [2012] EWCA Civ 24; [2012] ETMR 17. Whether this was right as a matter of interpretation of the order does not matter. What matters is that the purpose of the order is, as Kitchin LJ put it in Specsavers:

"to avoid the spending of time and money on what is clearly irrelevant and unsatisfactory evidence."

149. For the future, the standard form of order should be redrafted so as to make it clear that:

i) A party may conduct a true pilot survey without permission, but at his own risk as to costs;

ii) No further survey may be conducted or adduced in evidence without the court's permission; and

iii) No party may adduce evidence from respondents to any survey without the court's permission.

150. In deciding whether to give permission, the court must evaluate the results of whatever material is placed before it. Only if the court is satisfied that the evidence is likely to be of real value should permission be given. The reliability of the survey is likely to play an important part in that evaluation. Even then the court must be satisfied that the value justifies the cost. As Mr Hobbs said, this requires the court to conduct a cost/benefit analysis. In a case of trade mark infringement in which the issue is one of deception in relation to the provision of ordinary consumer goods or services, these criteria are likely to be satisfied only in a special or unusual case.
151. If what is sought is permission to carry out a survey, the applicant should provide the court with:

i) The results of any pilot survey;

ii) Evidence that any further survey will comply with the Whitford guidelines; and

iii) The cost of carrying out the pilot survey and the estimated cost of carrying out the further survey.

152. If what is sought is permission to call witnesses who have responded to a survey or other experiment, the applicant should:

i) Provide the court with witness statements from the witnesses proposed to be called;

ii) Demonstrate that their evidence will be of real value in deciding the issues the court has to decide;

iii) Identify the survey or other experiment and, in the case of the administration of a questionnaire disclose how many surveys have been carried out, exactly how those surveys were conducted and the totality of the number of persons involved and their answers to all questions posed;

iv) Disclose how the proposed witnesses were selected from among the respondents to the survey; and

v) Provide the court with the cost of carrying out the pilot survey and the estimated cost of carrying out any further work in relation to those witnesses.

153. It may have a material effect on the court's decision whether (and if so to what extent) the applicant is prepared to waive privilege in so far as it attaches to the selection, interviewing and preparation of witness statements for the witnesses proposed to be called. In the absence of a waiver of privilege in this respect, a party who wishes to challenge the evidence is likely not to be able to do so effectively; and in particular would not know what questions were asked of the witness in order to prepare the witness statement. If the evidence proposed to be called cannot be effectively challenged, that may in itself reduce its potential probative value.

The micro objection

154. Strictly speaking that makes it unnecessary to consider the micro objection; but I will do so briefly. It will be recalled that one of the concerns expressed in the Whitford guidelines is that questions might lead a witness into a field of speculation upon which that person would never have embarked had the question not been put. That can be seen to have happened in this case. I give some examples, all taken from the answers to question 6 in the first questionnaire ("Thinking specifically about this search result (POINT to MARKS & SPENCER RESULT). What if anything do you think this tells you about any relationship between Interflora and Marks and Spencer?"):

i) No 33: "Any relationship between them? I don't think of it that way. Nothing";

ii) No 34: "Well, it could be that Interflora is using M & S as a portal and as a launch pretty much to reach customers of M & S";

iii) No 39: "Well I've just never associated the two but now you've mentioned it, I suppose they might do the deliveries";

iv) No: 56: "None. You wouldn't think they are connected but I'm sure they are in some way";

v) No 60: "I don't know anything about it. You picked a subject that I am not into";

vi) No 71: "I didn't know they had one";

vii) No 73: "Never thought about that";

viii) No 79: "I suppose if I thought about it there may be a relationship between them as the qualities are equally good";

ix) No 86: "My guess would be that Interflora is a franchise and M & S and other individual suppliers would be linked to it but I don't know how exactly".

155. Mr Hobbs also pointed to a number of answers that either showed that the respondent drew nothing from the results, or appreciated that there was no link between Interflora and M & S. Again I give examples of responses to the first questionnaire:

i) No 3: "It doesn't tell anything – other than possibility that M & S use Interflora as a go-between";

ii) No 6: "I don't see any relationship between the two of them, there's nothing to link them";

iii) No 7: "It doesn't give me any clue that there'd be any connection at all";

iv) No 10: "It tells that there isn't a relationship between M & S and Interflora";

v) No 12: "Strange thing is I don't think of a relationship between Interflora and M & S";

vi) No 13: "Don't think it does";

vii) No 16: "Normally I wouldn't expect there is a relationship because it's on the right side so a Google selected ad";

viii) No 19: "Doesn't suggest any particular relationship between them, but you can get the same service from both";

ix) No 21: "It doesn't";

x) No 34: "Nothing";

xi) No 39: "I don't think it does, it's just that they deliver it, whether they do the work for M & S I just don't know."

xii) No 41: "I'm probably guessing that Interflora deliver the flowers for Tesco's M & S and other places at cheap prices";

xiii) No 40: "Without opening the website? Nothing".

156. It would be tedious to multiply examples. I simply fail to see how against that background calling a selection of witnesses would advance the case that Interflora wants to make.
157. Interflora has served a Respondent's Notice in relation to the second questionnaire. I can deal with this even more shortly. The judge decided that the critical question posed to the interviewees overstepped the mark. It carried the innuendo that there was a relationship between Interflora and M & S where that was the very question in issue; and that evidence from such interviewees would be tainted by a flawed question. This was a value judgment to which the judge was entitled to come.
158. In my judgment Mr Hobbs' micro objection is also well made.
159. I would allow the appeal.

Lord Justice Etherton:

160. I agree.

Lord Justice Hughes:

161. I also agree.