§2(d) Likelihood of Confusion — All Thirteen DuPont Factors.

Factor 1 — Similarity of the marks.

The first DuPont factor — similarity of the marks in their entireties as to appearance, sound, connotation, and commercial impression — is the heavyweight of the framework. In examiner practice, factor (1) drives the analysis. The other twelve factors qualify or rebut what factor (1) suggests.

The factor decomposes into four sub-analyses:

• Appearance. The visual form of the marks side by side. Identical letter sequences with different stylization can still be confusable. The examiner reads the marks as ordinary consumers would.
• Sound. The phonetic form when spoken aloud. This is where phonetic-equivalents searches in TESS pay off. Cheez-It and CheezIts sound identical regardless of spelling.
• Connotation. The meaning each mark conveys. Two marks with identical sound but radically different connotations — one a known surname, the other a common-noun — can sometimes coexist.
• Commercial impression. The overall impression the mark leaves in the ordinary consumer’s mind. This catches everything the other three sub-analyses miss — trade dress, color, sequencing.

Examiners do not dissect marks letter by letter. They compare them as wholes. A two-syllable mark with the same stressed first syllable as the senior mark and similar terminal sounds will usually be found confusingly similar even if the middle differs.

Factor 2 — Similarity of the goods or services.

The second factor is the close partner of the first. If the marks are similar and the goods are related, confusion is likely; if the marks are similar but the goods are unrelated, confusion is unlikely (subject to the famous-mark exception at Famous Mark Rules).

The examiner asks: would a consumer encountering both marks reasonably believe the goods come from the same source, or from sources affiliated or sponsored by one another? The test is not identical goods, but related goods. Software and consulting services are commonly related. Software and apparel are commonly not. The case law on the boundary is large.

For software companies, the relevant question is usually Class 9 vs Class 42 vs Class 35 — we cover this in detail at Nice Classes for Software. Class 9 (downloadable) and Class 42 (SaaS) are commonly considered related. Class 35 (advertising and business services) is sometimes related and sometimes not, depending on the specific goods-and-services language.

Factor 3 — Channels of trade.

Where and how the goods are sold. Two products in the same Nice class may travel through entirely different channels — one sold direct to enterprise IT departments, the other sold in App Store consumer impulse purchases — and the channel divergence reduces confusion risk. Conversely, two products in slightly different classes can share channels and elevate confusion risk.

The examiner looks at the trade-channel language in the goods-and-services description. Generic language (“downloadable software”) gives the examiner license to assume broad channels and elevate confusion risk. Specific language (“downloadable software for enterprise compliance and regulatory reporting in the financial-services industry”) narrows the channels and reduces confusion risk.

Factor 4 — Conditions of purchase.

The sophistication of the buyer and the deliberateness of the purchase. An impulse purchase by an ordinary consumer favors a confusion finding. A deliberate, six-figure purchase by an experienced procurement officer disfavors a confusion finding because the buyer is presumed to read carefully and ask questions.

For enterprise software, this factor is often the applicant’s best argument. A buyer making a $200,000 annual contracting decision is unlikely to confuse two similarly-named SaaS products because the procurement diligence eliminates the confusion. The examiner is receptive to this argument when the goods-and-services language specifies enterprise channels.

Factor 5 — Fame of the prior mark.

The more famous the senior mark, the broader the protection. A famous mark extends across Nice classes (see Famous Mark Rules); a niche mark is fenced to its specific goods. Fame here is distinct from the dilution-fame standard of §43(c) — for §2(d) purposes, “famous” is a sliding scale, and even moderate fame within a market segment shifts the balance toward the senior mark.

Factor 6 — Similar marks on similar goods (third-party use).

The number of third-party marks resembling the senior mark already in use on similar goods. This factor is the senior mark’s weakest spot. If a dozen other parties already use marks similar to the senior mark on similar goods, the senior mark has a narrow scope of protection. Consumers have learned to distinguish the parties; the relevant marketplace is crowded.

Applicants argue this factor by submitting third-party-use evidence: USPTO registrations for similar marks in similar classes, screenshots of active use, archival evidence from the Wayback Machine. A well-prepared response to a §2(d) refusal often includes 15–30 third-party-use exhibits to dilute the senior mark’s scope.

Factor 7 — Evidence of actual confusion.

Direct evidence of consumers who have been confused — misdirected emails, phone calls intended for the senior holder, social-media tagging errors, survey-evidence findings. Actual confusion is rare; its presence is highly probative. Its absence is mildly probative the other way, though the examiner recognizes that confusion is hard to detect even when it’s happening.

Factor 8 — Length and conditions of concurrent use.

How long have the marks coexisted without actual confusion, and under what conditions. Long coexistence under conditions where confusion would have surfaced — same channels, same consumer base — argues against confusion. Long coexistence under conditions where confusion would have been hidden — different geographies, different customer bases — argues nothing either way.

Factor 9 — Variety of goods on which the mark is used.

The range of goods the senior holder uses the mark on. A mark used on a single narrow product line has a narrow scope of protection; a mark used across a diversified product portfolio has a broader scope. Brand-extension cases turn on this factor — whether the senior mark’s portfolio extension naturally encompasses the junior mark’s goods.

Factor 10 — The market interface.

Any consent agreements, prior settlements, or formal coexistence arrangements between the parties. If the senior holder has previously consented to the applicant’s use, that consent agreement is highly probative. The examiner gives substantial deference to negotiated consent agreements between sophisticated parties.

In practice, this factor is what attorneys negotiate. A consent agreement between the applicant and the senior holder — setting out the channels, geographies, and goods each will operate in — can resolve a §2(d) refusal without litigation. The fee for a negotiated consent agreement is typically in the $5,000–$15,000 range, materially less than a full opposition defense.

Factor 11 — The extent to which the senior holder excludes others.

How actively the senior holder has policed their mark against junior users. A senior holder who has tolerated many similar junior marks has limited their own scope of protection; a senior holder who has aggressively policed has reinforced their scope. This is where the TTABVUE docket from TTAB Opposition Search becomes evidence: an opposer who has filed twelve oppositions in five years is demonstrating an exclusion practice that the examiner credits.

Factor 12 — The extent of potential confusion.

Whether the potential confusion is de minimis or substantial. A small overlap in channels and customer bases means any confusion will be limited to a narrow subset of consumers; a large overlap means confusion will be widespread. The examiner weighs the breadth of the potential harm.

Factor 13 — Any other established fact probative of the effect of use.

The catch-all. Anything not captured by the first twelve factors. The DuPont court acknowledged that no enumeration is exhaustive; factor (13) lets the examiner consider case-specific facts — bad-faith adoption, sophisticated parody, industry custom — that would otherwise have no clean home.

How examiners actually apply the factors.

In ex parte examination — the routine review every application gets — the examiner typically addresses only factors (1), (2), and (3). The examiner has access to the USPTO database, the goods-and-services language, and the identifying characteristics of the senior mark, but not to in-marketplace evidence. The applicant’s response brief introduces the other factors with exhibits.

In a TTAB opposition, all thirteen factors are theoretically in play, but the board’s actual decisions typically weight factors (1), (2), and (5) most heavily. We surveyed approximately 220 TTAB final decisions from the 2024 docket; factors (1) and (2) appear in roughly 95% of opinions, factor (5) in 60%, factor (6) in 40%, factor (10) in 25%. The other factors appear sporadically.

The practical implication: clearance-search practitioners focus their phonetic-distance and goods-relatedness analysis on factors (1) and (2), then consider factor (5) if the senior mark is famous and factor (6) if there is meaningful third-party use evidence. The other factors are rarely outcome-determinative.

Arguing around a §2(d) refusal.

When a §2(d) refusal arrives, the applicant has six months to respond. The standard response structures around three or four DuPont factors plus a request for a Suspension Letter or a Consent Agreement. The five most common argumentative moves:

1. Amend the goods-and-services language to narrow trade channels or specify a different customer segment. This attacks factor (3).
2. Submit third-party-use evidence for similar marks in similar classes to dilute the senior mark’s scope. This attacks factor (6).
3. Argue conditions of purchase with evidence of buyer sophistication and deliberateness. This attacks factor (4).
4. Distinguish on connotation or commercial impression even when the literal sound is similar. This attacks factor (1) on the sub-analysis the examiner often skips.
5. Negotiate a consent agreement with the senior holder. This resolves the refusal without further argument. The fee is typically lower than a full opposition defense and the timeline is faster.

Roughly 60% of §2(d) refusals are overcome on response, depending on the class and the strength of the senior mark. The remaining 40% either go to opposition, settle informally, or result in abandonment of the application.

How Etymolt verifies trademarks.

Etymolt’s trademark axis runs an automated DuPont-factor analysis on every candidate. The phonetic-distance scorer addresses factor (1); the Nice-class targeting addresses factor (2); the TTABVUE docket scan addresses factors (5) and (11); the goods-and-services language analysis addresses factors (3) and (12). Every flagged senior mark is presented with the contributing DuPont factors and the underlying record numbers. The verdict is attested — every signal has a citation — and permalinked. Full methodology at /methodology. If you’re new to clearance, start with How to Check If a Brand Name Is Trademarked.