Refused Because the Words Outweigh Your Logo: What Carried Section 2(d) Cases When the Design Alone Would Not

You are in this situation if your Section 2(d) refusal compares a mark that has words and a design (yours, the registrant's, or both) and says the words count for more than the picture, so the logo does not separate the marks.

Before you argue

If the owner of the cited registration will sign a consent agreement, that is often the shortest path, and a consent is itself one of the du Pont factors the examiner must weigh (BRECKENRIDGE filed one and was approved). If you can live without the overlapping goods or services, you can delete them: JANUS removed batteries from its list and was approved. Narrowing only works if the examiner's evidence no longer reaches what is left: IMERSE. narrowed its list and still received a final refusal that cited new accounting-firm websites.

If you decide to argue: three arguments that carried cases on their own

Examiners weigh the du Pont factors, a list of considerations from the 1973 du Pont decision (factor 1 compares the marks, factor 2 the goods, and so on). Their starting point is the manual's rule: "If a mark comprises both wording and a design, greater weight is often given to the wording, because it is the wording that purchasers would use to refer to or request the goods or services" (TMEP 1207.01(c)(ii)). Each approved case below won with argument and evidence alone: no consent, no change of ownership and no change to the goods, unless noted.

Whichever you choose, attach copies of the registrations or pages you rely on (USPTO status printouts or certificates). A list typed into the response does not count as evidence.

1. Crowded field: the shared wording is weak, so the design and the other words carry the impression (du Pont factor 6)

The factor: a crowded field is the sixth du Pont factor at work. The manual: "Evidence of third-party use falls under the sixth DuPont factor – the 'number and nature of similar marks in use on similar goods.'" And on a weak shared word: "If the common element of two marks is 'weak' in that it is generic, descriptive, or highly suggestive of the named goods or services, it is unlikely that consumers will be confused unless the overall combinations have other commonality" (TMEP 1207.01(b)(viii)).

When it fits: the wording both marks share is descriptive of the goods or common in the field (MIDWAY for restaurants, POOCH for dog care), and your mark has something else, a second word or a real design, to put in its place. Once the shared word stops pointing to one source, the rest of each mark is what gets compared.

What to file: other owners' registrations of the shared wording for goods like the cited ones, screenshots of those marks in use, a dictionary entry, and, if the examiner required you to disclaim the shared wording, that requirement itself. The manual's test: "a large number of active third-party registrations including the same or similar term or mark component for the same or similar goods or services may be given some weight to show" that the term is weak, and "The third-party marks must generally be as similar to the registered mark as the applied-for mark" (TMEP 1207.01(d)(iii)).

In their words. POOCH'S response said that, with the shared words disclaimed, what was left was "a distinctive design of a dog silhouette within a dog paw print as the dominant feature of Applicant’s mark". The final refusal that answered ASCEND WELLNESS: "evidence comprising only a small number of active use-based third-party registrations including the same or similar component is generally entitled to little weight in determining the strength of a mark." The one that answered DELIRIO: "Applicant contends that the shared term is a common Spanish word. However, there is no evidence that the term has any significance with respect to applicant's goods."

The pattern: the approved responses gave the examiner something countable (25 websites in X MIDWAY CAFE, thousands of disclaimers in NY INSTITUTE OF AROMATHERAPY) or the examiner's own disclaimer requirement (POOCH'S), for the same kind of goods or services. The refused ones filed three registrations or a statement about the word's meaning.

2. The goods side: take the examiner's evidence apart and file Thor Tech pairs (du Pont factor 2)

The factor: when the wording is the same in both marks and the design cannot help, the relatedness of the goods (the second factor) is where several cases turned. The burden is the examiner's: "The examining attorney must provide evidence showing that the goods and services are related to support a finding of likelihood of confusion" (TMEP 1207.01(a)(vi)). The model for pairs is In re Thor Tech (2015), where the Trademark Trial and Appeal Board held: "Applicant’s evidence of dozens of third-party registrations for the same or very similar marks owned by different entities for vehicles and recreational vehicle trailers rebuts the relevant, two third-party registrations made of record by the Trademark Examining Attorney."

When it fits: the examiner's relatedness proof is a few websites or registrations, and your goods differ in kind from the registrant's (roofing services against shingles, pilates classes against clothing).

What to file: go through the examiner's evidence one item at a time, and add pairs: the same mark owned by one company for your kind of goods and by another for the registrant's kind, with the registration certificates.

In their words. THUNDERGUARD's response: "five of the seven third-party registrations submitted in support of relatedness do not cover the registrants’ ‘shingles’ at all". The final refusal that answered NEFER AESTHETIC CLINIC: "This argument is unpersuasive because the application also includes skincare products in Class 3 which are encompassed by the goods identified in the application." The one that answered IMERSE.: "prior decisions and actions of other trademark examining attorneys in applications for other marks have little evidentiary value and are not binding upon the USPTO or the Trademark Trial and Appeal Board."

The pattern: the approved files had a real gap between the goods (services against a product) and answered the examiner's own evidence. In the refusals, the application still listed identical goods (NEFER) or the examiner's evidence reached the services (IMERSE.).

3. The marks differ as a whole, and the drawings show it (du Pont factor 1)

The factor: the manual limits its own word-over-design rule: "[t]here is no general rule as to whether letters or designs will dominate in composite marks; nor is the dominance of letters or design dispositive of the issue" (TMEP 1207.01(c)(ii), quoting the Federal Circuit). The next section cuts the other way: "an applicant cannot, by presenting its mark in special form, avoid likelihood of confusion with a mark that is registered in standard characters because the registered mark presumably could be used in the same manner of display" (TMEP 1207.01(c)(iii)).

When it fits: the marks differ in more than the picture. The other mark adds words, or its literal element is not what the examiner says it is. If the wording is identical and the cited mark is in standard characters, see the losses below before relying on this.

What to file: both drawings side by side, the cited registration's own description of its mark, and Board decisions where a prominent design carried the result.

In their words. SENSE BY IPCOM's response: "Applicant’s mark has 3 words, 5 syllables, and 12 letters. The cited mark has 1 word, 1 syllable, and 5 letters." CX BRANDS's: "Importantly, the mark is identified in the Trademark Office records as solely consisting of the letter ‘C’." The final refusal that answered WE R ONE: "Applicant's reliance on the design elements incorporated into applicant's mark to differentiate the marks is not compelling because the registered mark is in standard characters."

The pattern: in each approved file the design came with a difference in the words themselves (PREMIUM and GERMANY, BY IPCOM) or, in CX BRANDS, evidence of what the cited mark is. A design alone, against identical wording in a standard-character registration, did not carry a case here. The record does not show how much weight the examiner gave the drawing in AAA or SENSE BY IPCOM.

Arguments that rarely carried a case alone

A logo set against the same wording in a standard-character registration did not carry a case on its own (WE R ONE, and the design paragraph of the NEFER final refusal). Descriptions of how the businesses really sell: DELIRIO compared its bottle label, and the final refusal answered, "In this case, applicant's identification includes all alcoholic beverages except beer." The manual states the limit: "An applicant may not restrict the scope of its goods and/or the scope of the goods covered in the registration by extrinsic argument or evidence, for example, as to the quality or price of the goods" (TMEP 1207.01(a)(iii)). Approved files made points like these too (X MIDWAY CAFE noted three years without known confusion), but alongside one of the three arguments above.

Other routes that apply to any Section 2(d) refusal, such as owning or canceling the cited registration, are covered in Fighting Back, Part 2.

Finding cases like yours

GleanMark's Office Action research (Professional and Max plans) lets you select a Section 2(d) refusal and the argument hook "Word Over Design" (it always comes with a mark-comparison hook such as "Shared Dominant Wording"), then filter to cases won by argument alone. Its Thor Tech pairs and crowded-field tools search the register for the evidence described above.

This article describes public USPTO records. It is not legal advice; each refusal turns on its own marks, goods and evidence.