Background
Packard Press, Inc. — a commercial printer specializing in legal, municipal, and financial printing for securities firms, financial institutions, and law firms — filed an application to register the service mark PACKARD TECHNOLOGIES for data processing and information processing services. Hewlett-Packard Company, one of the world’s most recognized technology brands, opposed the application on the ground that PACKARD TECHNOLOGIES was likely to be confused with HP’s family of HEWLETT-PACKARD marks. HP argued that consumers encountering PACKARD TECHNOLOGIES would mistakenly believe the services came from or were affiliated with Hewlett-Packard.
The U.S. Patent and Trademark Office Trademark Trial and Appeal Board sustained HP’s opposition, finding a likelihood of confusion. However, the Board’s analysis focused primarily on the PACKARD element of the HEWLETT-PACKARD mark, isolating that single word and comparing it to PACKARD TECHNOLOGIES. Packard Press appealed, arguing that the Board had applied the wrong analytical framework by dissecting HP’s composite mark rather than comparing the marks as a whole.
The Court’s Holding
The Federal Circuit vacated the Board’s decision and remanded for further analysis. Judge Clevenger wrote that the Board committed a fundamental error in trademark likelihood of confusion analysis: it improperly dissected HP’s HEWLETT-PACKARD mark by considering only its PACKARD component in isolation, rather than evaluating the marks in their entireties as required by established trademark law.
The court explained the bedrock principle that trademarks must be compared as wholes, not analyzed piece by piece. The overall commercial impression of HEWLETT-PACKARD — including its first component, HEWLETT, and the combined visual and phonetic impact of the full mark — is not the same as the impression created by the word PACKARD alone. By stripping HP’s mark down to its second component and comparing that component to applicant’s PACKARD TECHNOLOGIES, the Board distorted the proper inquiry. Whether consumers would confuse PACKARD TECHNOLOGIES with HEWLETT-PACKARD must be assessed by comparing both complete marks as consumers would encounter them.
The Federal Circuit also found that the Board’s analysis of the relatedness of the parties’ goods and services was insufficient. The Board had not made adequate findings or applied the correct legal standard to assess whether HP’s trademark-related goods and Packard Press’s data processing services were sufficiently related to increase confusion risk. Without proper analysis of this DuPont factor, the Board’s conclusion on likelihood of confusion lacked an adequate foundation. The case was remanded for the Board to reconsider both the mark similarity analysis and the goods/services relatedness factor under the proper legal standards.
Key Takeaways
- Trademarks must be compared in their entireties — the likelihood of confusion analysis must consider the overall commercial impression of the complete marks as consumers encounter them, not isolated components or fragments of composite marks.
- Focusing exclusively on one component of a composite mark in a likelihood of confusion analysis constitutes legal error; even if marks share a common element, the surrounding elements may substantially reduce consumer confusion risk.
- DuPont factor analysis requires the TTAB to make adequate findings and apply the correct legal standard for each relevant factor — including the relatedness of the parties’ goods and services — before drawing a likelihood of confusion conclusion.
- The shared presence of a common word (here, “Packard”) between two marks does not automatically establish likelihood of confusion, particularly when one mark is a composite with additional distinguishing elements like “Hewlett.”
- Applicants for registration and TTAB practitioners should ensure that the Board’s decision addresses all relevant DuPont factors with adequate findings, as inadequate analysis of any material factor can be grounds for reversal.
Why It Matters
Packard Press v. Hewlett-Packard is an instructive case on the proper methodology for trademark likelihood of confusion analysis before the TTAB and the Federal Circuit. The “anti-dissection” rule — that marks must be compared as wholes, not cherry-picked by favorable components — is one of the most important and frequently litigated principles in trademark law. Courts and the TTAB must resist the temptation to isolate the most distinctive or most similar element of a mark and base the confusion analysis solely on that element.
This case also illustrates the Federal Circuit’s role as the appellate court for TTAB trademark decisions and the court’s willingness to vacate TTAB rulings that misapply the DuPont factors. The Federal Circuit reviews TTAB likelihood of confusion determinations as questions of law, with substantial evidence review for underlying factual findings. When the Board’s factual findings are inadequate or its legal methodology is wrong, the Federal Circuit will remand for proper analysis rather than affirm a conclusion that rests on a flawed foundation. For trademark practitioners, the case reinforces the importance of a thorough, factor-by-factor DuPont analysis in both TTAB proceedings and Federal Circuit appeals.