Understanding Trademark Similarity
Whether two marks conflict depends on how similar they are and how close the goods are. This article explains the test and how its elements are weighed.
- By
- By Bart Lieben, Attorney-at-Law
- Published
The legal framework
Three questions in trademark practice come down to the same test: can a new mark be registered, can it be used, and should you oppose someone else's. In EU law the test is likelihood of confusion, including the likelihood of association. It appears as a relative ground for refusal in Article 8(1)(b) of the EU Trade Mark Regulation (EUTMR) and as an infringement criterion in Article 9(2)(b). The Benelux Convention on Intellectual Property (BCIP) applies the same standard, which derives from Directive (EU) 2015/2436.
The Court of Justice has set out the method. The assessment is global (Sabel, C-251/95; Canon, C-39/97; Lloyd Schuhfabrik, C-342/97). It takes account of all relevant factors and is made from the viewpoint of the average consumer of the goods concerned, who is reasonably well informed, observant and circumspect but rarely sees two marks side by side and relies on an imperfect recollection.
Four elements recur:
- The similarity of the signs: visually, phonetically and conceptually, judged by overall impression, with attention to distinctive and dominant elements.
- The similarity of the goods and services: their nature, purpose, method of use, and whether they compete or complement each other.
- The distinctiveness of the earlier mark, which can be inherent or acquired through use. A strong mark receives a wider scope of protection.
- The relevant public and the territory: for an EU mark a conflict in part of the Union, even one language area, can be enough.
The elements are interdependent. A high degree of similarity between the goods can offset a lower degree of similarity between the signs, and the other way round. A conflict also requires some similarity of both: identical signs for clearly dissimilar goods do not give rise to a likelihood of confusion under Article 8(1)(b), although a mark with a reputation may still be protected under Article 8(5).
Classification does not decide similarity. Goods are not similar merely because they sit in the same Nice class, nor dissimilar because they sit in different classes (Article 33(7) EUTMR). See the Nice classification.
In practice
Visual. Marks are compared by length, structure, shared letters and the position of the differing parts. Consumers usually pay more attention to the beginning of a word, so "Alvento" and "Alventa" are closer than "Alvento" and "Valento". Figurative elements count, but a word element often dominates in the overall impression because consumers name a product by its words.
Phonetic. Marks are compared as spoken, in the languages of the relevant public. A mark that looks different on paper may sound the same in Dutch, French or English, which is why a search looking only at spelling misses real conflicts.
Conceptual. If both marks convey the same idea, they may be confused even when they look and sound different. If one has a clear meaning for the public and the other does not, or they have different meanings, the conceptual difference can offset the visual and phonetic similarity. This works only where the meaning is immediately grasped by the public.
Weak elements. Descriptive or laudatory components are given less weight. If the shared part of two marks is descriptive, such as "Bio" for organic food, the common element does not bring the marks much closer together.
Consider a Mechelen company that makes accounting software under the name "Ledgro" and a competitor that applies for "Ledgrow" for payroll software. The signs differ by one letter and sound identical, and the goods are close. Confusion is likely even though the second mark is not identical. If the competitor applies for "Ledger Pro", the shared element "Ledger" is arguably descriptive for accounting software, and the picture is less clear.
The same test runs through our work. In a clearance it decides how serious a hit is. In a trademark watch it decides which new filings reach you, and in an opposition it is the argument itself. Our technology scores signs visually, phonetically and conceptually to organise the work; an attorney makes the legal assessment.
Frequently asked questions
Do two marks have to be identical to conflict?
No. Identity is one ground, but a likelihood of confusion between similar marks for identical or similar goods is the more common basis for refusals, oppositions and infringement claims.
Is a different class enough to be safe?
Not necessarily. Similarity of goods is assessed in substance, not by class. Goods in different classes can be similar, and goods in the same class dissimilar.
Does a figurative element avoid confusion?
Sometimes, but not reliably. If the word element is distinctive and dominant, adding a logo often does not remove the likelihood of confusion.
Can a famous mark block a very different product?
Yes, if it has a reputation and your use would take unfair advantage of it or harm its distinctiveness or repute (Article 8(5) EUTMR), even without confusion.
How this connects
Trademark clearance and trademark watch and opposition both apply this test. It is built into our clearance engine and our trademark and design watch.
In short
- Likelihood of confusion is assessed globally
- Visual, phonetic and conceptual comparison
- Similar goods and similar signs compensate each other
- The same test applies in clearance, watch and opposition