Trade mark infringement assessment: the questions a court asks.

The scale of it

The registers keep growing and so do the fakes. The UK Intellectual Property Office received 173,180 trade mark applications in 2024, granted 156,596 registrations and recorded 6,695 oppositions. Global trade in counterfeit goods reached an estimated $467 billion in 2021, 2.3% of world imports, and 4.7% of everything imported into the EU.

The test

In the United Kingdom the claim is section 10(2) of the Trade Marks Act 1994; in the European Union it is article 9(2)(b) of the EU Trade Mark Regulation. Both ask the same thing: is the sign identical or similar to the registered mark, used for identical or similar goods or services, such that there exists a likelihood of confusion on the part of the public, including a likelihood of association.

The Court of Justice set the method in Sabel v Puma (C-251/95): a global appreciation of all the relevant factors, through the eyes of the average consumer of the goods, who is reasonably well informed, observant and circumspect, sees the marks as wholes and rarely compares them side by side. Canon v MGM (C-39/97) added that the factors interact: a lesser degree of similarity between the goods can be offset by a greater degree of similarity between the marks, and the more distinctive the earlier mark, the greater the likelihood of confusion.

The five questions

• Are the marks identical or similar? Compared visually, aurally and conceptually, giving more weight to the distinctive and dominant elements.
• Are the goods or services identical or similar? Nature, purpose, method of use, whether they compete or complement each other, the channels they sell through.
• How distinctive is the earlier mark? Inherently, and through use. A well-known mark is protected more widely.
• Who is the average consumer, and how much attention do they pay? A pharmacist and a teenager buying a T-shirt look at a mark differently.
• Taking all of that together, is confusion likely? Including the belief that the goods come from economically linked undertakings.

Where the earlier mark has a reputation, section 10(3) and article 9(2)(c) add a sixth question: whether the sign takes unfair advantage of, or is detrimental to, the mark's distinctive character or repute, with or without confusion.

Building an assessment a partner can sign

The questions are settled. The work is the evidence beneath each answer, and that is where assessments go wrong: a registration number copied from memory, a search run in one office but not the others, a "similar" with no record of what was compared. A file that survives the other side's letter has, for each element of the accused product, the register entry it was compared against, the office record it was verified in, and the reasoning on each of the five questions in writing.

How we do it

Signum breaks the accused product into its elements, searches each against the trade mark registers in play, links every match to its office record with its verification status, and scores the questions above with the evidence beneath each score. What comes out is an assessment, not a verdict: the lawyer challenges it line by line, corrects it and signs it, in the firm's own form. It is in pilot with a law firm.