$350 Buys A Trademark Application In The US—But Nontraditional Marks Present Unique Challenges

Anyone with US $350 and a dream can file an application to register their trademark with the United States Patent and Trademark Office (USPTO). But if that dream involves nontraditional trademarks (NTMs)—such as colors, sounds, scents, textures, product designs, motion, or store layouts—searching for similar marks in the USPTO's Trademark Status & Document Retrieval (TSDR) database can be difficult.

Why is searching TSDR for NTMs challenging? Members of the Non-Traditional Trademarks Committee—U.S. Subcommittee, as well as in-house trademark lawyers and outside counsel INTA members, recently had the opportunity to discuss this question with USPTO leadership and examining attorneys. The USPTO reported that it is actively working on a design code path to make NTM searching more structured and ultimately more reliable, but the final solution will take time, resources, and input from practitioners.

Why Searching for NTMs Is Difficult Using the Current "Trademark Search" System

By definition, NTMs extend beyond words and simple figurative elements. A scent described as a "sweet, slightly musky, vanilla fragrance, with slight overtones of cherry, combined with the smell of a salted, wheat-based dough" (U.S. Registration No. 5467089, Hasbro, Inc.), or a motion that unfolds across frames, can be captured in prose but not easily normalized for database retrieval.

Historically, the USPTO has relied on a combination of mark drawing codes and free-text mark descriptions. For nonvisual marks, that meant the broad bucket of mark drawing code 6, assigned to non-visual marks, such as sound marks, scent marks, or other marks that cannot be represented visually, paired with whatever detail an applicant included in the description field. The TM5 Guidelines for Searching Non-Traditional Marks provides helpful tips for searching various kinds of NTMs.

Since the retirement of the Trademark Electronic Search System (TESS) and the launch of the new trademark search system in November 2023, users must now use "Expert" mode and the "Mark description" field to find NTMs. For example, "Mark description: wheat AND cherry AND vanilla" or "Mark description: magenta."

This manner of searching is difficult because there is no required consistency in how applicants describe their NTMs (beyond "The mark consists of ..." preamble that is hard-coded into the USPTO's application form), and there is an indefinite number of ways for applicants to describe their marks.

Is it possible to make mark descriptions uniform for searching but also accurate? The USPTO seeks practitioner feedback on this question but noted that, for at least sound marks, requiring musical notation—while precise—would exclude nonmusical sounds.

Another discussion point was whether the USPTO could standardize sound mark descriptions and rely on soundbites for accuracy and searchability. Unfortunately, this isn't feasible due to:

• Incompatibility with international offices that don't accept audio files;

• Limitations under the Madrid Protocol;

• Difficulty defining legal boundaries; and

• High system and maintenance costs for hosting soundbites on TSDR.

The USPTO's Design-Code Pivot: Promise and Practicalities

Against this backdrop, the USPTO has prioritized a shift back to design codes for classifying and searching NTMs. The goal is not simply to repurpose Vienna-style figurative tags but to build a richer, searchable taxonomy that encodes what makes an NTM meaningful and comparable. For visual NTMs, that may mean more granular, combinable codes for shapes, colors, positions, textures, and repeating elements. For sensory and motion marks, the USPTO is considering category-level coding (for example, sound with words versus without words; speech, tonal, or environmental sound; simple versus complex motion; presence of transitions or resolutions; and potentially standardized text hooks that pair with code selections).

Several realities shape this work. First, retroactive tagging is labor-intensive. The corpus of existing NTMs includes legacy descriptions that do not map neatly to new codes, and backfilling at usable quality takes time. Second, while preexamination automation is under consideration, it requires appropriate safeguards. The USPTO is exploring AI-assisted assignment of design codes at intake, particularly when applicants utilize "uncommon mark" selections at filing. That can accelerate routing and searching, but it still relies on training data and oversight by examining attorneys to mitigate classification errors. Third, harmonization matters. Any domestic coding schema must align with Madrid constraints, international databases, and how partners in the TM5 (a collaborative group of the five largest trademark offices in the world) search and display NTMs.

None of these challenges are insurmountable, but they involve foundational infrastructure that competes for finite budget and engineering resources. The USPTO has been candid about those constraints and emphasized that targeted practitioner input helps them prioritize effectively.

The USPTO Wants Your Help

As a practitioner, if you have ideas for standardizing NTM descriptions or suggestions for design codes that would improve searching for colors, scents, sounds, and other NTMs, the USPTO wants to hear from you. Send your feedback to feedback@TrademarkCenter.gov. Because NTMs are few, relative to wordmarks and logos, practitioner input can help guide and accelerate progress.

Originally published by International Trademark Association (INTA).