Bulletins | July 19, 2017

Colour trade marks and branding – The struggle continues

It is often said that a picture conveys more than a thousand words. In the world of brands, the same can be said of colours.

Colours and colour schemes provide an instant means of enabling consumers to identify their favourite products and suppliers, whether via walking along a supermarket aisle, surfing a website or driving along the road searching for a petrol station.

But when it comes to registering and protecting colours as trade marks, and so monopolising their use in relation to particular products or services, there are many legal and technical obstacles in the way.

This article briefly examines the current state of the law in this area and looks at whether or not it is likely to change when amended trade mark legislation comes into force later in the year.

Colours per se as trade marks

It is a fundamental requirement of any registered trade mark that it must be sufficiently distinctive to indicate the trade origin of the particular goods or services for which it is registered.[1]

In most cases, a new brand will cross that threshold with ease. It may be a logo, a word, a combination of the two or even in some cases a shape. But if a brand owner wishes to monopolise a particular colour alone – without more – for its products, services or corporate image, then it will have no hope of achieving registration where that colour has yet to see any use.

A pure colour, per se, is inherently non-distinctive. It can only become distinctive by being used in trade to such an extent that the relevant public becomes educated to associate the particular colour with a particular undertaking.

Thus, to register a pure colour mark, in order to overcome the distinctiveness requirement, the proprietor will have to be able to demonstrate that the mark has become distinctive through use. In the case of a UK national trade mark, that will mean showing that it has acquired the necessary distinctive character in the UK. But for an EU trade mark, the proprietor will need to provide evidence to support that contention in every one of the EU’s Member States.[2]

This is a formidable hurdle and helps explain why so few colour marks are presently on the register.

However, even where a colour mark does achieve registration, such marks create all kinds of difficulties when it comes to registering and enforcing them. For example, what if a brand owner has a shade of purple registered as a colour mark for confectionary and another trader happens to use that same or a similar colour on the packaging of its own products? Will there be infringement if only a speck of the offending colour appears on the rival’s packaging or does it have to be sufficiently prominent in use to suggest it is being used in a trade mark sense rather than simply as decoration?

The latest “colour” case to come before the Courts in England & Wales is Glaxo Welcome v Sandoz[3] in which judgment was handed down by the Court of Appeal on 10 May 2017.

The case provides a helpful summary of the law concerning colour marks as that law has evolved since the original introduction of the European trade mark regime more than 25 years ago. It also underlines again the perils for trade mark owners of getting it wrong when filing these types of marks.

Evolution of the case law on colour marks

In the early noughties, the Court of Justice of the European Union (CJEU) had to grapple with the question of whether an application to register a pure colour mark was acceptable.

In the Libertel[4] case, the Court held that in principle such a trade mark was registerable but that merely filing a graphical image of the particular colour was not sufficient. There had to be an accompanying verbal description and/or the use of an internationally recognised colour code. Above all, the application to register the mark needed to meet the criteria identified by the case of Sieckmann[5] – i.e. there needed to be an unequivocal, clear, precise, self-contained easily accessible, intelligible, durable and stable form of graphic representation.

It thus became fashionable to apply to register colours as trade marks by reference to, for example, a specific pantone colour reference.

However, some proprietors went further and sought to include verbal descriptions to help define their colour marks. For example, in Heidelberger[6], the CJEU had to pronounce on whether the following description of a blue and yellow colour mark was acceptable:

“The trade mark applied for consists of the applicant’s corporate colours which are used in every conceivable form, in particular on packaging and labels….”

This was found to be unacceptable as it was not possible for anyone consulting the register to see a “sign” and compare that to what they were or were not entitled to use themselves.

The need to ensure that a trade mark could not obtain an unfair advantage by monopolising something unclear in scope was further emphasised by the Court in Dyson.[7] In that case, Dyson had attempted to register as a trade mark the transparent bin which formed part of their bagless vacuum cleaners. Since this would have covered any kind of transparent bin, not just that depicted in the application for the mark, this was unacceptable. It was not a genuine “sign”.

Cadbury’s purple trade mark

Cadbury had applied to register a trade mark for the colour purple for various goods in class 30. The trade mark was described by reference to a pantone colour reference. But it was also accompanied by a verbal description as follows:

“The colour purple (Pantone 2685C), as shown on the form of application, applied to the whole visible surface, or being the predominant colour applied to the whole visible surface, of the packaging of the goods.”

The Court of Appeal held that this form of words including the phrase “or being the predominant colour….” were fatal to the validity of the trade mark. This was because they opened up the mark potentially to taking the form of other colours or visual material. In other words, the scope of the trade mark could not be ascertained with the necessary precision and clarity. The term “predominant” was not precise enough.

However, attempts at proving precision does not always come hand in hand with clarity. Sometimes, it does not help to provide information that is too detailed!

For example, in the Seven Towns[8] case, the trade mark applicant (owner of the IP rights in the famous Rubik’s Cube) had sought to register the following colour mark comprising seven colours:

In each case, as shown above, a pantone reference was provided for the colour blocks representing the trade mark. But there was an additional verbal description as follows:

“Six surfaces being geometrically arranged in three pairs of parallel surfaces, with each pair being arranged perpendicularly to the other two pairs characterised by: (i) any two adjacent surfaces having different colours and (ii) each such surface having a grid structure formed by black borders dividing the surface into nine equal segments.”

This is a description of a Rubik’s Cube. However, the EU’s General Court upheld the rejection of this trade mark and ruled that the European Union Intellectual Property Office (EUIPO) had been correct in deciding that “a reasonably observant person with normal levels of perception and intelligence would, upon consulting the…register…be unable to understand precisely what the mark consists of, without expending a huge amount of intellectual energy and imagination.” The mark was therefore invalid.

In the Red Bull case[9], the EUIPO Board of Appeal dismissed an application by Red Bull to register a colour mark comprising a sign which depicted two strips of colour, identified by pantone references where the ratio of their used was described as being “approximately 50%-50%.”

This too was too uncertain. The dual colours could be used in any number of different combinations to form different signs. This was too imprecise.

The Glaxo v Sandoz case

In Sandoz the registration applied for was of a drawing of an aspect of an inhaler with purple covering as follows:

This was accompanied by the verbal description:

“The trade mark consists of the colour dark purple (Pantone Code 2587C) applied to a significant proportion of an inhaler, and the colour light purple (Pantone code 2567C) applied to the remainder of the inhaler….” [emphasis added].

The mark was designated with the INID[10] code 558 to denote that it was a mark consisting exclusively of one or more colours.

The judge held that the registration set a puzzle. It wasn’t clear what the mark was supposed to be.

For example, did it refer to just the colours in the patterns depicted in the photo of the particular inhaler shown in the registration (i.e. was it a case of “what you see is what you get”)? This was unlikely because the mark was an INID code 558 – colour mark and the description referred to “an” inhaler rather than “the” inhaler suggesting it was not specific to that depicted in the picture.

Or, did it mean that there could be any number of different proportions of dark and lighter purple colours compared to that shown in the picture depicted in the registration?

In the conclusion of the Court, the trade mark lacked clarity, intelligibility, precision and specificity. It was registered in breach of Article 4 of the Regulation and had to be declared invalid. A reference to the CJEU was refused.


Although the Red Bull case is currently under appeal to the CJEU, it would be surprising if the CJEU were to overrule the General Court. It seems to the writer that the principles upon which the General Court dismissed Red Bull’s trade colour trade mark application were squarely in keeping with previous case law on the subject and common sense.

It is evident that colour trade marks remain fraught with difficulty because of the issues that arise when attempts are made to describe what precisely is being registered. Some verbal descriptions are too detailed, some are not detailed enough. When it comes to infringement, the problems multiply.

It remains to be seen whether forthcoming changes to EU trade mark law will open the way to further, successful colour mark registrations. In particular, from October of 2017, the need for a trade mark to be capable of being represented “graphically” will disappear. This is due to the new Trade Marks Directive. From 1 October 2017 onwards, it will be enough if the trade mark is capable of:

“being represented on the Register of…trade marks, in a manner which enables the competent authorities and the public to determine the clear and precise subject matter of the protection afforded to its proprietor.”

This provides for more flexibility in the way that a trade mark is represented and will open the way for new forms of expression that would not have been possible under the old “graphical representation” requirement.

For further information, please contact Michael Gardner at mgardner@wedlakebell.com.

[1] Article 3(1)(b) Directive 2008/95/EC

[2] Mondelez UK Holdings & Anor v EUIPO Case T-112/13

[3] Glaxo Wellcome UK Ltd & Anor v Sandoz Ltd [2017] EWCA Civ 335

[4] Libertel Groep BV v Benelux –Merkenbureau Case C-104/01

[5] Sieckmann v Deutches Patent und Markenamt Case C-273/00

[6] Heidelberger Bauchemie Case C-49/02

[7] Dyson v Registrar of Trade Marks Case C-321/03

[8] Seven Towns Ltd v OHIM Case T-293/10

[9] Red Bull GmbH v Optimum Mark Sp z.o.o Case R 2037/2013-1

[10] INID is short for “International agreed Numbers for the Identification of (bibliographic) Data” developed by WIPO and adopted by EUIPO which enables proprietors to specify characteristics of trade marks and third parties to understand them more easily.