Likelihood of Confusion in France: How the INPI Assesses Conflicting Trade Marks (Article L.711-3)

Likelihood of confusion is the central concept in France’s system for resolving conflicts between trade marks. Under Article L.711-3 of the Code de la Propriété Intellectuelle (CPI), a trade mark application may be refused — on opposition by a third party — if it is identical or similar to an earlier mark and the goods or services are identical or similar, such that there exists a likelihood of confusion on the part of the public. France follows the EU framework for assessing confusion, including the global assessment methodology and the interdependence principle. Importantly, Nice Classification classes are not strictly determinative of goods or service similarity.

The Legal Framework: Relative Grounds for Refusal

The relative grounds for refusal are set out in Article L.711-3 of the CPI (as restructured by Ordinance No. 2019-1169). The key provisions include:

• Article L.711-3, I, 1°: An application shall be refused where the sign is identical to an earlier trade mark and the goods or services are identical. In this case, likelihood of confusion is presumed.
• Article L.711-3, I, 2°: An application shall be refused where, because the sign is identical or similar to an earlier trade mark and the goods or services are identical or similar, there exists a likelihood of confusion on the part of the public, including the likelihood of association with the earlier mark.
• Article L.711-3, I, 3°: Trade marks with a reputation (renommée) in France enjoy enhanced protection. An application may be refused even for non-similar goods or services if use of the later mark would take unfair advantage of, or be detrimental to, the distinctive character or repute of the earlier mark.

Because the INPI does not examine relative grounds ex officio, these provisions are considered only when an opposition is filed within the 2-month window, or in subsequent invalidation proceedings before the INPI or the courts.

The Three Dimensions of Similarity

The INPI and French courts assess the similarity of marks across three dimensions, following the framework established by the CJEU and consistently applied by the French Court of Cassation:

Visual Similarity

This examines the overall visual impression of the marks, including the length of words, the arrangement and sequence of letters, shared letter groups, the presence or absence of figurative elements, colours, and layout. For word marks, the beginning of the word carries particular weight, as French consumers — like consumers across Europe — tend to focus on the first part of a word mark. For figurative or composite marks, the relative dominance of the word element versus the design element is assessed in context.

Phonetic (Aural) Similarity

This examines how the marks sound when spoken aloud in French. Phonetic similarity carries significant weight in the French market, particularly for goods and services commonly requested orally (food, beverages, fashion, personal services). The assessment considers the number of syllables, the rhythm and cadence, the dominant vowel and consonant sounds, the position of stressed syllables, and the particular phonetic rules of the French language. Foreign-language marks are assessed based on how French consumers would pronounce them.

Conceptual Similarity

This examines whether the marks convey the same or a similar meaning or idea. Two marks that are visually and phonetically different may still be confusingly similar if they evoke the same concept. Translations are captured: a French word and its English equivalent may be found conceptually similar or identical if the average French consumer understands both. Conceptual differences can also neutralise visual or phonetic similarities, particularly when at least one mark conveys a clear, specific, and immediately recognisable meaning.

The Global Assessment and Interdependence Principle

The INPI conducts a global assessment of likelihood of confusion, following the methodology established by the CJEU and applied consistently by the French Court of Cassation (e.g., Cass. com., January 31, 2018, No. 16-10761; May 30, 2018, No. 16-22993). This includes:

• The visual, phonetic, and conceptual similarity of the marks, assessed on the overall impression of each mark as a whole, bearing in mind their distinctive and dominant components.
• The similarity of the goods or services (not limited to Nice Classification — see below).
• The distinctive character of the earlier mark (inherent and acquired).
• The well-known character or degree of recognition of the earlier mark.
• The relevant public and their level of attention.
• The principle of imperfect recollection — the average consumer does not typically compare marks side by side but relies on an imperfect memory of the earlier mark.

The interdependence principle means that these factors are not assessed in isolation. A high degree of similarity between the marks can compensate for a lower degree of similarity between goods, and vice versa. A highly distinctive earlier mark enjoys a broader scope of protection, meaning confusion is more readily found.

Nice Classification Is Not Determinative

An important principle in French practice — consistent with EU law — is that the Nice International Classification of goods and services is an administrative tool, not a legal determinant of similarity. Goods in different classes may be found similar (e.g., clothing in Class 25 and footwear in Class 25, or food products in Class 29 and beverages in Class 32), and goods in the same class may be found dissimilar (e.g., firearms and musical instruments, both historically in certain shared classes).

The INPI and French courts assess goods and services similarity based on their intrinsic nature, intended purpose, method of use, whether they are complementary or in competition, and their distribution channels. The class number is simply the filing category — it does not define the scope of protection or the likelihood of confusion.

The Role of the Relevant Public

The assessment is based on the perception of the average consumer of the goods or services in question. This consumer is deemed to be reasonably well-informed, observant, and circumspect. However, the level of attention varies: for everyday, impulse-purchase goods, the consumer is less attentive and confusion is more readily found; for expensive, specialised, or professional goods, the consumer is more careful and confusion is less likely.

Proof of Use in Opposition Proceedings

Since the 2019 reform, if the earlier mark on which an opposition is based has been registered for more than five years at the date of the contested application, the applicant can require the opponent to prove genuine use (usage sérieux). If the opponent fails, the opposition is dismissed for the goods and services lacking proof of use. This was a significant addition to French opposition practice — previously, only courts could assess proof of use.

Strategic Recommendations

• Apply the interdependence principle in your risk analysis: Consider the similarity of marks and goods together. A moderate similarity on both axes may be sufficient to create confusion.
• Focus on phonetic similarity for the French market: In many product categories, phonetic similarity carries decisive weight, particularly where goods are frequently requested orally.
• Differentiate across all three dimensions: Strength in one dimension does not compensate for weakness in another. Ensure your mark is distinct from prior marks visually, phonetically, and conceptually.
• Do not rely on different Nice classes: Being in a different class does not automatically prevent a finding of confusion. Assess the intrinsic relationship between the goods.
• Use the proof of use request: If the opposing mark has been registered for more than five years, requiring proof of use is a powerful tactical tool that may eliminate or narrow the opposition.

Common Mistakes

• Focusing on only one dimension of similarity: Visual differences do not matter if the marks sound identical. Conceptual similarities can override other differences.
• Ignoring the interdependence principle: Moderate similarity on both the marks axis and the goods axis may suffice for confusion, even where neither factor alone seems particularly strong.
• Assuming Nice Classification determines similarity: It does not. The intrinsic relationship between goods matters, not the class number.
• Overlooking conceptual neutralisation: If the marks have clearly different meanings that consumers would immediately recognise, this conceptual difference may neutralise visual or phonetic similarities.
• Failing to assess the public’s level of attention: The same degree of similarity may create confusion for cheap consumer goods but not for specialised professional products. Tailor the analysis to the relevant consumer.

Key Takeaway

Likelihood of confusion assessment in France follows the EU framework: a global, multi-factor analysis that weighs the similarity of marks and goods in light of the interdependence principle. Nice Classification classes are administrative tools, not legal boundaries — similarity is assessed based on the intrinsic nature and purpose of the goods. Because the INPI does not examine relative grounds ex officio, these principles come into play only when an opposition or invalidation action is filed. For both applicants and opponents, a thorough understanding of the three dimensions of similarity and the interdependence principle is essential to predicting outcomes and building effective arguments before the INPI.