Trademark Likelihood of Confusion: How the USPTO Decides

The Legal Standard — What "Confusion" Actually Means

The Lanham Act (the federal trademark statute) prohibits registration of a mark that, when used with specific goods or services, is likely to cause confusion with an already-registered mark. The standard is not certainty of confusion — it's likelihood. And the relevant consumer isn't an expert carefully comparing two brand names side by side. It's an ordinary purchaser with imperfect recollection, encountering one mark at one time and one mark at another, deciding whether they might come from the same source.

That "ordinary consumer with imperfect recollection" standard is important. It's why "Froot" and "Fruit" can conflict in the same class. It's why "Citi" and "City" for related financial services are treated as potentially confusingly similar. The question isn't whether you, examining both marks carefully, would confuse them — it's whether a consumer shopping in the category might.

The DuPont Factors: The 13-Part Framework

In 1973, the U.S. Court of Customs and Patent Appeals (now the Federal Circuit) established a multi-factor framework for evaluating likelihood of confusion in In re E.I. DuPont DeNemours & Co. USPTO examiners and courts have applied these 13 factors ever since — though not all factors are relevant in every case, and not all factors carry equal weight.

1. Similarity of the marks — in appearance, sound, and meaning. The most important factor in almost every case. Evaluated holistically, not feature by feature.
2. Relatedness of the goods or services — the second most critical factor. How related are the products or services covered by the two marks? The more similar the commercial space, the more likely confusion.
3. Similarity of trade channels — do the goods or services reach consumers through the same distribution channels? Two brands sold exclusively through different platforms, to different buyer types, in different markets face lower confusion risk.
4. Conditions of purchase — are consumers making careful, deliberate purchasing decisions (expensive items, professional buyers) or impulse, low-price purchases? Sophisticated purchasers who research carefully are less susceptible to confusion than impulse buyers.
5. Strength of the prior mark — how distinctive and commercially well-known is the mark that might be infringed? Famous, strong marks receive broader protection. Weak or descriptive marks receive narrower scope.
6. Number and nature of similar marks — if the relevant market is crowded with similar marks coexisting without confusion, that weakens the argument that one more similar mark will cause confusion.
7. Actual confusion — evidence of actual consumer confusion is highly significant when it exists. Absent evidence, this factor is often neutral.
8. Length of concurrent use without confusion — if both marks have been in use for years in the same market with no documented confusion, that's relevant evidence.
9. Variety of goods and services — the range of goods or services the prior mark covers, including natural expansion zones.
10. Market channels — overlap in target customers.
11. Extent of potential confusion — how significant would confusion be in practice.
12. Concurrent use agreements — existing agreements between the parties about coexistence.
13. Any other facts probative of the effect of use — a catch-all for case-specific evidence.

The Two Factors That Actually Drive Most Decisions

In practice, factors 1 and 2 — similarity of marks and relatedness of goods/services — dominate the vast majority of likelihood of confusion analyses. When an examining attorney at the USPTO issues a Section 2(d) refusal (likelihood of confusion), these are almost always the core issues. The other factors often play supporting roles or are neutral.

Similarity of marks is evaluated in three dimensions:

• Sound — how the marks are pronounced. "Neva" and "Nova" sound similar. "Aqua" and "Aquila" sound similar. The examiner considers how an ordinary consumer would say the mark.
• Appearance — the visual impression of the marks as written. Similar arrangements of letters, similar overall visual format.
• Meaning — the commercial impression or conceptual meaning. "Big Red" and "Large Red" for the same goods might be found confusingly similar in meaning even if they sound and look different.

Any one of these dimensions can create a finding of similarity. A mark that looks different but sounds the same in a category where consumers often hear brand names (radio advertising, word-of-mouth) may still be found confusing.

Relatedness of goods/services is evaluated by asking: would an ordinary consumer, encountering both marks at different times, assume the goods or services come from the same source or are affiliated? This requires looking beyond the Nice Classification class numbers to the actual goods and services descriptions. A Class 25 apparel brand and a Class 35 online retail store "featuring apparel" might be found related even though they're in different classes.

Identical Marks in Unrelated Industries — When They Coexist

One of the most commonly misunderstood aspects of trademark law: identical marks can legally coexist if the goods or services are completely unrelated. "Delta" for airlines and "Delta" for faucets have coexisted as federal registrations for decades. "Polo" for clothing (Ralph Lauren) and "Polo" for vehicles (Volkswagen) coexist internationally.

The test remains the same: would an ordinary consumer assume the airline and the faucet manufacturer are related? No reasonable person believes Delta Air Lines makes faucets. The industries are so distinct, with such different purchase contexts and consumer bases, that confusion is implausible.

This is why the goods and services description in your trademark application is strategically important. A narrow, precise description limits the scope of your mark to what you actually do — but it also limits the scope of conflicts. A broad description ("business services" rather than "consulting for biotechnology companies") may increase your protection but also draws more existing marks into potential conflict.

What Happens When You Get a Section 2(d) Refusal

A Section 2(d) Office Action is not a final rejection. It's the examiner's initial finding that your mark is likely to cause confusion with an existing registration. You have 3 months to respond (extendable to 6 months for a fee).

A well-reasoned response to a Section 2(d) refusal argues through the DuPont factors:

• Distinguish the marks: sound different, appear different, create different commercial impressions
• Distinguish the goods/services: explain why the industries are sufficiently different that consumers wouldn't assume a common source
• Distinguish the trade channels: if the marks operate in entirely separate markets, document that
• Argue the weakness of the cited mark: if the cited mark is descriptive or there are many similar marks coexisting in the market, use that to limit its scope
• Submit evidence of coexistence without confusion if you've been operating alongside the cited mark

Many Section 2(d) refusals are overturned on response, particularly when the industries are adjacent rather than identical, or when the marks differ in meaningful ways beyond the single similarity the examiner focused on.