Who Owns a Brand Colour

Why We Look · 24 October 2026 · Reviewed by Faisal Zaidi

A brand can hold a legal right to a colour, but only to a named shade, on named goods, once buyers read it as a signature. Whether buyers do can be counted, and the two questions are answered in different places.

Answer to the question

Can a brand own a colour?

Partly. In the United States and Europe a brand can hold a legal right to its brand colour, but only to a named shade, for named goods, once buyers read it as pointing to one source. Courts have refused loosely worded claims, as with Cadbury's purple in 2013, and narrowed broad ones, as with Louboutin's red sole in 2012.

Key takeaways

The US Supreme Court held unanimously in Qualitex v. Jacobson Products (1995) that colour alone can be a trademark, provided buyers have come to see it as identifying one source.
In Libertel (2003) the Court of Justice refused orange for telecommunications because words did not represent the shade clearly enough, and said a few registrations could exhaust the range of colours.
Louboutin's red sole was limited in 2012 to red against a contrasting upper, and Cadbury's purple failed in 2013 because the word predominant described no single sign.
A 2014 test with 880 consumers across three categories measured fame and uniqueness, and recommended showing the colour and asking for the brand with no list.

A colour can be a trademark, and in the United States that has been settled law for three decades. In Qualitex Co. v. Jacobson Products Co. a unanimous Supreme Court held that a shade of green-gold, used on the dry-cleaning press pads one company had sold for decades, could identify that company and no other. The Court's reasoning was short, but the conditions it attached have been tested in courtrooms on both sides of the Atlantic ever since.

A brand colour qualifies only once buyers read it as pointing to one source, and the cases since show how much work sits inside that condition. Courts have kept the right narrow and refused claims that were loosely worded, while the buyers' side of it, whether people link a colour to one company, can be put to a survey.

What Courts Allow a Brand to Claim in a Colour

The Qualitex dispute began with a rival. The company registered its green-gold as a trademark after a competitor started using a similar shade, and the Ninth Circuit barred the registration. The Supreme Court reversed that on 28 March 1995. It wrote that there was "no rule absolutely barring the use of color alone", and it attached two conditions. The colour must have "secondary meaning", which here means buyers have come to see it as identifying a single source. It must also not serve a significant non-trademark purpose, such as making a product work better or cost less to make.

Europe took a more cautious line. In Libertel, decided in 2003, a company sought to register orange for telecommunications services. The Court of Justice said a colour on its own has little inherent capacity to communicate information, and that distinctiveness without prior use is conceivable only in "exceptional circumstances". It also named a public interest in "not unduly restricting the availability of colours": the number the public can tell apart is limited, and a small number of registrations could exhaust the range, a worry with a practical form in categories where many brands lean on one hue, which [Why Brands Reach for Blue](https://thewhyweproject.com/articles/why-brands-reach-for-blue) takes up.

Libertel's application failed on a narrower point. Words did not represent the shade clearly enough, because "orange" means different things to different people. The Court noted that a widely recognised code, such as a Pantone reference, could supply the precision.

Does Anyone Link That Colour to the Brand at All?

A court asks whether buyers link a brand colour to a company, and survey researchers were asking the same thing while the early applications were fought. A 2004 master's thesis at Massey University, known here from its repository abstract, took two live cases as its subjects: Cadbury's application to register purple for chocolate, and the unsuccessful application by Effem Foods to register orange for Uncle Ben's rice. It compared four survey methods: direct association questions, a sequence of increasingly specific questions, a task matching shades to brands, and a choice-modelling experiment.

All four methods showed a strong link between Cadbury and purple, although Cadbury was also associated with other colours. For Uncle Ben's, the results suggested orange was not inherently tied to the brand. Where strong links appeared, the methods agreed with one another, and the author recommended choice modelling as the strongest method because it measures how a pairing affects choices instead of relying on stated associations.

That gives the law two facts to work with. A brand colour can be strongly linked to a company in buyers' minds, and the same brand can fail to have that link at all. Neither fact settles whether the link is exclusive.

Louboutin Kept Red, but Only Against a Contrasting Upper

The best-known test of the Qualitex rule on a single brand colour came from fashion. Christian Louboutin's registration, as a law-firm summary of the judgment reports it, covered "a lacquered red sole on footwear". Yves Saint Laurent's accused shoe was red all over, and a district court had ruled that a single colour could not be a trademark in fashion at all. On 5 September 2012 the Second Circuit reversed that blanket rule, because the Supreme Court had forbidden such rules.

It then found secondary meaning only where the red sole contrasts with the upper, and limited the registration to those cases. A monochrome red shoe with no contrasting sole did not infringe, and the denial of a preliminary injunction stood.

The result gave Louboutin a brand colour and withdrew the claim to a red sole in every setting. What the brand holds is a relationship: red against a different upper. Why red means anything to buyers in the first place is the subject of [What Red Actually Does](https://thewhyweproject.com/articles/what-red-actually-does).

What the brand holds is a relationship: red against a different upper.

Cadbury's Purple Fell on the Wording of Its Application

Cadbury's brand colour was Pantone 2685C, a purple, and its application covered milk chocolate. It described the colour as applied to the whole visible surface of the packaging, or as "predominant" on it. On 4 October 2013 the Court of Appeal, in the account WIPO Magazine gives, overturned the trial judge in Société des Produits Nestlé SA v Cadbury UK Ltd and refused the registration. The word "predominant" implied other colours the application never described, so the mark could cover many permutations of packaging, and that was not a single sign of the kind the law requires.

The court distinguished Libertel, which concerned one unchanging colour, and declined to refer the question to the Court of Justice. The UK Supreme Court refused Cadbury permission to appeal on 11 March 2014, on the ground that the application raised no arguable point of law.

The ruling concerned wording and not buyers. The court's stated reason was the application, not any finding that people failed to link purple to Cadbury, and the 2004 survey work had found a strong link in its samples. A High Court judgment in 2022 dealt with three Cadbury applications for the same purple, and its reasoning again turned on how each one was written.

The court's stated reason was the application, not any finding that people failed to link purple to Cadbury.

Counting the Link, and What the Book Left Out

By 2014 there was a method for measuring a brand colour against the question courts treat as evidence. A paper co-written by Jenni Romaniuk of the Ehrenberg-Bass Institute reported a split-sample test with 880 consumers in banking, chocolate and hair care. It defined two outcomes: fame, meaning how many people link the brand to the colour, and uniqueness, meaning how often competitors are linked to that colour. A colour that many people tie to one brand can still score low on uniqueness if rivals share it.

They varied whether respondents were cued with the brand or the colour, and whether they answered from a list or unprompted. They recommended the colour-cued, unprompted brand response. The abstract does not give the category results, so none are reported here.

The same logic runs through chapter 8 of Byron Sharp's How Brands Grow (2010), which Romaniuk co-wrote. It argues that brands should build distinctive assets such as a brand colour because they can be trademarked and meaningful differentiation cannot (from p. 112); why distinctiveness helps memory is argued in [Why the Odd One Out Gets Remembered](https://thewhyweproject.com/articles/why-the-odd-one-out-gets-remembered). The legal half of that argument survives the cases above, with the conditions attached. The measurement half is thinner. The chapter's four tables measure how different brands are perceived to be, not how recognisable their assets are, and no table in the book measures a commercial return to distinctiveness. The book's acknowledgements list the corporations that financially support the institute's work.

Ownership of a brand colour is therefore held in two places. A court holds the written part: one shade named exactly, tied to goods and a setting, and backed by proof that buyers read it as a signature.

Buyers hold the rest, which can be counted as the share who link the colour to the brand and the share who link it to someone else, and the 2014 paper recommends taking that count by showing the colour and asking for a name, with no list of names to prompt the answer.

Common questions

Can a brand own a colour?

What is secondary meaning for a colour trademark?

It is the condition the US Supreme Court set in Qualitex in 1995: buyers have come to see the colour as identifying a single source. A colour is not treated as distinctive on its own, so the right depends on evidence that people read it as a signature.

How do researchers measure whether a brand colour belongs to a brand?

A 2014 paper co-written by Jenni Romaniuk measured fame, how many people link the brand to the colour, and uniqueness, how often competitors are linked to it. Its recommended method shows the colour and asks, with no list, which brand it brings to mind. The test used 880 consumers in banking, chocolate and hair care.

Did Cadbury lose its purple because buyers did not link it to Cadbury?

The Court of Appeal's stated reason in 2013 was the wording of the application, which included the word "predominant" and so did not describe a single sign. It was not a finding about buyers' associations.

Sources

Supreme Court of the United States (1995). 'Qualitex Co. v. Jacobson Products Co., 514 U.S. 159.' Decided 28 March 1995. https://www.law.cornell.edu/supremecourt/text/514/159

Court of Justice of the European Communities (2003). 'Libertel Groep BV v Benelux-Merkenbureau, Case C-104/01.' Judgment of 6 May 2003, paragraphs 40, 47, 54, 55, 65 and 66. https://www.ippt.eu/sites/default/files/2003/IPPT20030506_ECJ_Libertel.pdf

Pfahlert, A. (2004). 'Brand-colour associations: a comparison of survey methods.' Master of Business Studies thesis, Massey University. Repository abstract only. https://mro.massey.ac.nz/items/10d58ced-67e2-4e73-98e6-b2475f40d5e1

United States Court of Appeals for the Second Circuit (2012). 'Christian Louboutin S.A. v. Yves Saint Laurent America Holding, Inc., No. 11-3303-cv.' Decided 5 September 2012; via a law firm summary. https://www.orrick.com/en/Insights/2012/10/Seeing-Red-When-Its-a-Trademark-and-When-Its-Not

Court of Appeal of England and Wales (2013). 'Société des Produits Nestlé SA v Cadbury UK Ltd [2013] EWCA Civ 1174.' Decided 4 October 2013; read through a WIPO Magazine summary. https://www.wipo.int/pt/web/wipo-magazine/articles/registering-trademarks-uk-court-sets-boundaries-38696

United Kingdom Supreme Court (2014). 'Societe des Produits Nestle S.A. v Cadbury UK Limited, UKSC 2013/0233: permission to appeal refused.' Decision of 11 March 2014. Reported at https://www.foodnavigator.com/Article/2014/04/04/Cadbury-purple-trademark-appeal-against-Nestle-denied/

High Court of England and Wales, Chancery Division (2022). 'Société des Produits Nestlé SA v Cadbury UK Ltd [2022] EWHC 1671 (Ch).' Read via a Wolters Kluwer blog summary. https://legalblogs.wolterskluwer.com/trademark-blog/the-purple-reign-sweet-taste-of-relief-for-cadbury-in-the-high-court/

Romaniuk, J., & Nenycz-Thiel, M. (2014). 'Measuring the strength of color brand-name links: the comparative efficacy of measurement approaches.' Journal of Advertising Research 54(3), 313-319. Abstract only. https://doi.org/10.2501/JAR-54-3-313-319

Sharp, B. (2010). How Brands Grow: What Marketers Don't Know. Oxford University Press, with a reliability audit. Ch. 8 (co-written with Jenni Romaniuk), from p. 112.