Phonetic similarity of less importance when goods are sold in self-service stores

In Esge AG v Office for Harmonisation in the Internal Market (OHIM) (Case T-444/10, February 21 2013), the General Court has upheld a decision of the Second Board of Appeal of OHIM in which the latter had found that there was no likelihood of confusion between the marks KMIX and BAMIX.

De'Longhi, formerly Kenwood, had applied for the registration of the word mark KMIX with OHIM for domestic and electrical appliances used for cooking in Classes 7 and 11 of the Nice Classification. An opposition was made on the basis of the word mark BAMIX, registered in Class 7 for the same type of products and in Class 40.

Before the General Court, the usual case law on how to compare goods and assess the likelihood of confusion was quoted. With regard to the relevant public, the court found that it was composed of both the general public and professional cooks. As a consequence, the relevant public would have an above average or a reasonable level of attention and circumspection. Referring to the decision in Kido (Case T-152/08), the court stated that the section of the public having the lower level of attention must be taken into consideration (ie, the general public). The court then confirmed that the goods at issue were identical or similar.

However, the real issue was the comparison of the marks. With regard to the visual comparison, the court recalled that both signs ended in the same three letters, but started with different letters, with one sign having four letters and the other five. According to established case law, consumers pay greater attention to the beginning of a mark than to the end (Wessang v OHIM - Greinwald (star foods) (Case T-492/08)). Furthermore, the brevity of the marks enabled consumers to better grasp the variations in their spelling (Ontex v OHIM - Curon Medical (CURON) (Case T-353/04)). As a consequence, the visual differences outweighed the common characteristics of the marks.

With regard to the phonetic similarity, both parties agreed that the previous mark would be pronounced as two syllables, namely 'ba' and 'mix'. The issue was whether KMIX would be pronounced as 'ka' and 'mix' or 'kmix'. The court stated that 'kmix' does not correspond to any existing word in the European Union; it may be pronounced by part of the relevant public as it is written, as a single syllable. However, the pronunciation of the mark as a two-syllable word, namely ‘ka’ and ‘mix’, was also possible. In certain languages of the European Union (in particular French and German), the letter ‘K’ is pronounced as ‘ka’ and the pronunciation ‘km’ is not usual. Therefore, the Board of Appeal had erred in not taking that fact into account, and the court concluded that both words had an average phonetic similarity for the part of the public which pronounces the mark as 'kamix'.

As a conceptual comparison was not possible, the issue was thus whether the visual similarity between the marks had more weight than the phonetic similarity in the assessment of the overall similarity.

The court recalled that it is appropriate to examine the objective conditions under which the marks may be present on the market. If the goods covered by the mark in question are usually sold in self-service stores where consumers choose the goods themselves and must therefore rely primarily on the image of the trademark applied to the goods, the visual similarity between the signs will, as a general rule, be more important. If, on the other hand, the goods covered are primarily sold orally, greater weight will usually be attributed to any phonetic similarity between the signs (see NLSPORT, NLJEANS, NLACTIVE and NLCollection (Joined Cases T-117/03 to T-119/03 and T-171/03), Paragraph 49, and BAUHOW (Case T-106/06), Paragraph 44).

In the present case, the goods covered by the marks were intended, in particular, for the general public and could be bought in department stores and specialty shops. These shops often operate on a self-service basis, even though they employ personnel to assist customers in their choices. Therefore, even though the purchase of the goods at issue may be made on the basis of an oral recommendation, the marks at issue will generally be perceived visually prior to purchase. The court added that oral communication will take place, where necessary, with qualified sales personnel who are capable of informing customers about the various brands.

As a consequence, the court decided that the visual similarity between the marks had more weight than the phonetic similarity in the assessment of the overall similarity. Therefore, the Board of Appeal's finding that the marks were not similar was upheld and the action was dismissed.

What is interesting in this case is the fact that the court did not compare the marks as filed - it took into consideration not only the distribution channels, but also the way in which the products were commercialised. Arguably, the General Court went too far since no one can predict how the registrant of the opposed mark is going to use its mark. Would the conclusion have been the same if a selective distribution system was intended to be used?

Richard Milchior, Granrut, Paris