The USPTO likelihood of confusion factors, known as the DuPont factors, help determine whether consumers would mistakenly believe that goods or services come from the same source. The USPTO applies this framework during trademark examination and opposition proceedings, while federal courts use jurisdiction-specific tests in infringement lawsuits.
Key Takeaways
- The USPTO's DuPont test contains 13 factors, not the eight factors found in the Polaroid court test.
- Similarity between the marks and relatedness of the goods or services commonly drive the USPTO's analysis.
- The USPTO weighs relevant evidence rather than counting how many factors favor each party.
- Actual confusion can support a claim, but the law does not require proof that confusion has already occurred.
- A trademark search should cover similar wording, spelling, sound, meaning, designs, and related goods or services.
- An applicant may respond to a Section 2(d) refusal with legal arguments, marketplace evidence, an amended identification, or an appropriate consent agreement.
What Is Likelihood of Confusion?
Likelihood of confusion exists when consumers are likely to believe that two parties' goods or services come from the same source, or that the parties are affiliated, connected, or sponsored by one another. The question is probable consumer perception, not whether the marks are identical or whether anyone intended to copy.
The issue arises in several settings. During examination, a USPTO examining attorney may refuse registration under Section 2(d) of the Trademark Act because an applied-for mark conflicts with a registered mark. An earlier-filed pending application may also delay examination or later become a basis for refusal. A trademark owner may raise the same general issue in an opposition before the Trademark Trial and Appeal Board, commonly called the TTAB.
In an infringement lawsuit, a court considers whether the defendant's use is likely to confuse consumers. The precise factors depend on the federal circuit hearing the case. Ownership, priority, protectability, and the nature of the challenged use can also affect the claim.
Similarity alone does not settle the issue. Similar marks can sometimes coexist for unrelated goods, while less similar marks may conflict when the goods, buyers, and commercial impressions closely overlap. Service marks receive the same likelihood-of-confusion analysis even though they identify services rather than physical goods. Understanding a mark's overall commercial impression helps explain why small spelling changes may not eliminate a conflict.
USPTO Likelihood of Confusion DuPont Factors
The 13-factor framework comes from In re E.I. du Pont de Nemours & Co., 476 F.2d 1357 (C.C.P.A. 1973). The USPTO's Trademark Manual of Examining Procedure applies the framework to Section 2(d) determinations. The following table translates the factors into practical questions and evidence.
| Factor | Issue Examined | Practical Question or Evidence |
|---|---|---|
| 1 | Similarity of the marks | Do appearance, sound, meaning, and overall commercial impression overlap? |
| 2 | Relatedness of goods or services | Could buyers expect the listed offerings to come from one source? |
| 3 | Trade channels | Would the offerings reach buyers through overlapping channels? |
| 4 | Purchasing conditions | Are purchases impulsive, careful, expensive, or made by sophisticated buyers? |
| 5 | Fame of the prior mark | What recognition, sales, advertising, or length-of-use evidence exists? |
| 6 | Similar marks in use | Does marketplace evidence show a crowded or weak field? |
| 7 | Actual confusion | Are there credible mistaken inquiries, purchases, reviews, or communications? |
| 8 | Concurrent use without confusion | Was there meaningful market overlap and an opportunity for confusion? |
| 9 | Variety of goods or services | Does the prior owner use its mark across a broad product or service range? |
| 10 | Market interface | Is there consent, an agreement, an assignment, or other interaction between the parties? |
| 11 | Applicant's exclusion rights | What established right does the applicant have to exclude others? |
| 12 | Extent of potential confusion | Would expected confusion be substantial or minimal? |
| 13 | Other relevant facts | What additional established fact shows the likely effect of marketplace use? |
How the USPTO Weighs the DuPont Factors
The DuPont analysis is not a mechanical checklist. The USPTO considers each factor for which relevant evidence appears in the record, but the factors do not receive equal weight in every case. One factor can matter more than several others when the facts make it especially probative.
The first two factors, similarity of the marks and relatedness of the goods or services, are often central. The inquiry works on a sliding scale. When marks are highly similar, the goods or services may need less similarity to support a refusal. When the goods or services are identical or closely related, smaller similarities between the marks can become significant.
Marks must be compared in their entireties. An examining attorney may still give greater weight to a dominant element, particularly when other wording is descriptive or less source-identifying. Differences in fonts, capitalization, or minor wording may carry little weight if the marks create similar sounds, meanings, and commercial impressions.
The USPTO ordinarily evaluates goods and services as identified in the application and cited registration. It does not rely solely on how the parties currently describe their real-world businesses. If an identification contains no limits on buyers or trade channels, the analysis may account for the normal channels and purchasers associated with those goods or services. Selecting accurate USPTO trademark classes and goods or services descriptions can therefore affect both the search and the examination record.
Similarity, Related Goods, and Marketplace Evidence
Similarity covers appearance, pronunciation, meaning, and commercial impression. Marks do not need to match in every respect. For example, phonetic equivalents may sound alike despite different spelling. Word marks can also conflict with composite marks when consumers are likely to use the same wording to request or remember the product.
Logos can be similar enough to create confusion, but sharing a common shape, color, or design idea is not automatically decisive. The decision depends on the complete designs, their distinctive elements, the relevant goods or services, and how buyers encounter them. Adding a logo to confusingly similar wording may not solve the underlying problem.
Relatedness does not mean the goods must compete directly. The question is whether consumers could reasonably believe the goods or services come from one producer. Evidence may include third-party registrations covering both types of offerings, websites showing that businesses commonly provide both, advertisements, catalogs, or other material connecting the markets.
Actual marketplace conditions can also matter. A crowded field of similar marks may show that consumers recognize small differences. Evidence of coexistence may help when both marks reached overlapping buyers under conditions where confusion could realistically have occurred. Conversely, no reported confusion has limited value if one party had minimal sales, operated in another region, or reached different customers. Expensive or specialized purchases may suggest greater care, but purchaser sophistication does not automatically prevent confusion over source or affiliation.
How to Search for Potentially Conflicting Trademarks
Start with the USPTO's official trademark search resources. Search active federal registrations and pending applications, including earlier-filed applications that could become obstacles. Do not stop after finding no exact match. The legal test reaches marks that look, sound, or mean something similar.
A useful clearance process includes several search paths:
- Search the complete wording and close spelling variations.
- Search phonetic equivalents, abbreviations, spacing changes, and plural forms.
- Search synonyms, translations, and wording that creates a similar meaning.
- Review design elements when the proposed mark contains a logo.
- Compare the listed goods or services, not just their international class numbers.
- Check each record's owner, filing basis, filing date, status, and identification.
- Investigate marketplace use beyond the federal database because unregistered users may have rights.
Trademark classes organize applications, but being in different classes does not automatically prevent a conflict. Goods in separate classes can still be related, and goods within one class can be commercially distinct.
Keep records of the searches performed and the closest results. If you proceed, the application must also satisfy use and filing requirements. Reviewing acceptable trademark specimen examples can help you avoid a separate problem when proof of use becomes necessary.
Responding to a Section 2(d) Refusal or Opposition
A Section 2(d) Office action identifies one or more marks that the examining attorney believes create a likelihood of confusion. Read the cited registrations, the refusal's reasoning, and the evidence attached to the action. Then compare the marks and identifications under each relevant DuPont factor.
A response may argue that the marks differ in appearance, sound, meaning, structure, or commercial impression. It may challenge the asserted relationship between the goods or services, using evidence about how those offerings are produced, marketed, or purchased. Applicants may also submit evidence of similar third-party marks to show that a shared element is weak or diluted.
Other possible approaches depend on the record. An applicant may narrow an identification if the amendment remains within the original scope and meaningfully separates the offerings. A carefully drafted consent agreement may help explain how the parties avoid confusion, although the USPTO evaluates its substance rather than treating consent as automatically controlling. Deleting goods or services can sometimes resolve only part of a refusal.
An opposition requires a litigation-focused response. The parties may exchange discovery, introduce testimony and documents, and present evidence bearing on multiple DuPont factors. Because statements made during examination or an opposition may affect later rights, applicants should consider both registration and broader enforcement consequences before choosing an argument.
If a search reveals a similar mark or the USPTO issues a refusal, a trademark lawyer can assess the cited marks and goods or services, identify useful DuPont evidence, and prepare a response or recommend a safer branding strategy. You can post your legal need on UpCounsel's marketplace to receive proposals from attorneys, with responses typically arriving within a day.
DuPont Factors vs. Polaroid, Sleekcraft, and Lapp
The DuPont and Polaroid factors are not interchangeable names for the same test. DuPont governs USPTO registration disputes, including examination and TTAB proceedings. Polaroid is a federal court test developed by the Second Circuit for infringement analysis. Other circuits use their own formulations, although the tests address many overlapping consumer-confusion concerns.
| Test | Primary Decision Maker and Context | Focus |
|---|---|---|
| DuPont | USPTO examining attorneys and the TTAB | Registration, refusals, oppositions, and cancellations under a 13-factor framework |
| Polaroid | Federal courts in the Second Circuit | Eight factors, including mark strength, similarity, product proximity, actual confusion, good faith, quality, and buyer sophistication |
| Sleekcraft | Federal courts in the Ninth Circuit | Eight factors addressing market proximity, marks, evidence of confusion, channels, purchaser care, intent, and expansion |
| Lapp | Federal courts in the Third Circuit | A circuit-specific infringement framework, including considerations relevant to noncompeting goods |
Federal court tests may consider evidence unavailable during USPTO examination, such as live marketplace presentation, sales conditions, testimony, surveys, advertising, and alleged intent. In contrast, ex parte examination focuses heavily on the application, cited registration, and evidence placed in the record.
Intent may support an infringement claim, but bad faith is not required to find confusion. Likewise, a defendant does not avoid liability merely by showing that its products are high quality. The controlling question remains how relevant consumers are likely to understand the marks and commercial relationship.
Reducing Trademark Confusion Risk
Choose a distinctive mark before investing in packaging, domains, signs, or advertising. Fanciful and arbitrary terms generally provide stronger source-identifying potential than generic or merely descriptive wording. Suggestive marks may also be protectable because consumers must use imagination to connect them with the offering.
Perform clearance before filing or launching. Compare more than names. Consider logos, slogans, product relationships, likely buyers, sales channels, and planned expansion. Online use deserves particular attention because similar names may appear next to each other in marketplace results, app listings, social media profiles, or advertising.
Do not assume that adding a generic word, changing one letter, or placing a house mark beside disputed wording will eliminate confusion. Evaluate the resulting mark as a whole and determine which elements consumers are most likely to remember.
If another party raises a concern, preserve relevant records. Useful material may include first-use documents, advertisements, sales records, customer communications, website captures, packaging, agreements, and evidence showing the parties' actual channels and buyers. Do not manufacture survey results or solicit confusion reports. Poorly designed evidence may carry little weight or create credibility problems.
Finally, monitor federal filings and marketplace activity after adoption. Early attention can create more options, including a negotiated boundary, a revised identification, a consent arrangement, an opposition, or a rebrand before costs increase. Registration strengthens important procedural and enforcement tools, but responsible clearance and consistent use remain central to protecting the brand.
Frequently Asked Questions
What Is the Test for Trademark Infringement?
The test asks whether the defendant's use of a mark is likely to confuse relevant consumers about source, sponsorship, affiliation, or approval. A plaintiff must also establish enforceable rights and satisfy other elements applicable to the claim. Federal circuits apply different multifactor formulations, so the court's jurisdiction and the marketplace evidence can shape the analysis.
What Is Likelihood of Confusion?
Likelihood of confusion is a probability that consumers will misunderstand the source or commercial relationship behind goods or services. It covers more than purchasing one product while intending to buy another. Confusion can involve sponsorship, affiliation, approval, or a belief that separate businesses are connected, even when consumers recognize that the products themselves differ.
What Is the Effect of Actual Confusion?
Credible actual-confusion evidence can strongly support an inference that future confusion is likely, but its weight depends on context. Courts and the TTAB may examine the number, quality, and circumstances of reported incidents. Misdirected communications can be useful, while ambiguous comments or incidents unrelated to purchasing or source identification may carry less weight.
What Are the DuPont Factors?
The DuPont factors are 13 considerations used in USPTO likelihood-of-confusion determinations. They address the marks, goods or services, trade channels, buyers, fame, third-party use, actual confusion, concurrent use, product range, interactions between the parties, exclusion rights, potential confusion, and other relevant facts. Only factors supported by the particular record may materially affect the outcome.
Can Logos Be Similar Without Causing Trademark Confusion?
Yes, similar logos may coexist when their total commercial impressions and market contexts remain distinguishable. Common geometric shapes or design conventions may receive a narrower scope of protection than distinctive artwork. The comparison still considers dominant features, wording, colors when claimed, goods or services, purchasers, and how the designs appear in actual or legally presumed trade channels.
Can Similar Trademarks Be Registered for Different Goods?
Yes, registration may be possible when the goods or services are sufficiently unrelated that consumers would not expect a common source. Different international classes alone do not establish that separation. The applicant should evaluate complementary uses, shared buyers, common producers, sales channels, and the prior mark's recognition before assuming that a different product description resolves the risk.
