You look at your mark. You look at the one the examining attorney cited. Different words, different logo, different customers, maybe a different country. Then you read “likelihood of confusion” and wonder whether anyone at the USPTO actually looked.
We hear that reaction on almost every discovery call. A robotics software founder whose mark was cited against a fitness brand. A clothing boutique owner who had used her name since 2017. A cosmetics founder whose brand comes from her own first name. Each of them said some version of the same sentence: “But they’re obviously different.”
Here is the part most applicants never hear. The first refusal usually rests on only two of the thirteen questions the law asks. The other eleven are still unanswered. By the end of this guide you’ll know which two the examiner used, why “different” to you can read as “similar” to them, and the one move that turns a first refusal from a verdict into an opening. If you are new to 2(d) refusals, our plain-English guide to likelihood of confusion is a good primer to keep open alongside this one.
A first refusal usually balances two considerations, the marks and the goods. The rest of the scale is empty until you add to it.
Executive summary
1. The sentence every applicant says
“They’re obviously different” is not a naive reaction. It is usually true from where you sit. You know your customers, your packaging, your price point and your story. You know that the cited company sells to gym owners while you sell to aerospace engineers, or that your name comes from a founder and not from a dictionary.
The trouble is that almost none of that knowledge is in front of the examining attorney. What they have is your application, the registration or earlier application they found in their search, and a short list of legal questions. The first Office Action reflects what they could see, not what you know.
That gap explains most of the frustration we hear. Clients tell us they feel the examiner is “trying to find something to deny,” or that the refusal is “obviously wrong.” It is more useful to read the refusal as a list of the questions your application did not yet answer.
2. The two factors behind most first refusals
Likelihood of confusion is decided under the thirteen factors from In re E. I. du Pont de Nemours & Co., 476 F.2d 1357 (C.C.P.A. 1973). We break each one down in How the USPTO Decides Likelihood of Confusion: The DuPont Factors.
Not every factor carries the same weight in every case. The court that followed du Pont said that “the fundamental inquiry mandated by § 2(d) goes to the cumulative effect of differences in the essential characteristics of the goods and differences in the marks.” Federated Foods, Inc. v. Fort Howard Paper Co., 544 F.2d 1098 (C.C.P.A. 1976). In practice, many first Office Actions rest on exactly those two:
- Factor 1, similarity of the marks. Appearance, sound, meaning and overall commercial impression.
- Factor 2, relatedness of the goods or services. Usually supported by a handful of web pages or third-party registrations suggesting one company might offer both.
That is a reasonable starting point for an examiner reviewing a large docket. It is not the whole analysis.
Most first refusals rely on factors 1 and 2. Factors 3 to 13 enter the analysis when the record contains evidence about them.
If you would like to see how your own refusal scores across these factors before writing a word of the response, that is the exact job the TRACE™ Report was built to do.
↑ Back to guide3. Why “different” to you reads “similar” to them
Three rules explain most of the gap between how an owner sees a mark and how an examining attorney sees it.
Rule 1: Marks are compared through four lenses, as a whole
The USPTO compares marks in their entireties for appearance, sound, connotation and commercial impression. Palm Bay Imports, Inc. v. Veuve Clicquot Ponsardin, 396 F.3d 1369 (Fed. Cir. 2005). Similarity in one lens can be enough to start a refusal. Two marks can look nothing alike in a logo and still sound alike when spoken, which is how many refusals involving creative spellings begin.
Rule 2: The dominant part carries extra weight
Examining attorneys may give more weight to the dominant portion of a mark, as long as the final comparison still considers the whole. In re National Data Corp., 753 F.2d 1056 (Fed. Cir. 1985). Leading words often draw that weight, because buyers tend to read and remember the first term. Words that describe the goods, like “Boutique,” “Beauty” or “Collection,” are often discounted or disclaimed. So a mark you see as a four-word phrase may be treated, at first, as its one distinctive word.
This rule cuts both ways. If the cited mark leads with a different term than yours, the same logic can support you.
Rule 3: Goods are compared as written, not as sold
The comparison uses the identification of goods and services in your application and in the cited registration. Octocom Systems, Inc. v. Houston Computer Services, Inc., 918 F.2d 937 (Fed. Cir. 1990). If you sell custom-engraved crystal but your application lists every glassware item in the class, the examiner compares the whole list. If the registration says “software” without limits, it is presumed to cover all kinds of software sold to all usual buyers.
- Your logo and colors
- Your actual customers
- Your price and channel
- Where the name came from
- The mark as filed, weighted to its dominant part
- The goods as written in both filings
- A few web pages linking the goods
- Doubt resolved for the registrant
That last point matters. In a close case, doubt is resolved in favor of the prior registrant. In re Hyper Shoppes (Ohio), Inc., 837 F.2d 463 (Fed. Cir. 1988). So your job is not to create doubt. Your job is to remove it with evidence.
4. The other eleven factors are still open
Here is the answer to the question from the top of this guide. The move that changes everything is to build the record. The first Office Action is not the end of the analysis. It opens a written exchange, and whatever you put into it becomes part of the case.
The Federal Circuit has held that the Board must consider every du Pont factor for which there is evidence and argument in the record. In re Guild Mortgage Co., 912 F.3d 1376 (Fed. Cir. 2019). In that case, the court sent back a decision that had skipped a factor the applicant had raised. The lesson for applicants: a factor you document has to be weighed. A factor you leave out does not exist.
Your response is the one point in the process where you decide what evidence the examining attorney must weigh.
The factors that most often change the outcome
- A crowded field (factor 6). If many businesses already use similar terms for similar goods, buyers learn to tell them apart by small differences. Evidence of third-party use and registrations can show the shared term is weak. Juice Generation, Inc. v. GS Enterprises LLC, 794 F.3d 1334 (Fed. Cir. 2015); Jack Wolfskin v. New Millennium Sports, S.L.U., 797 F.3d 1363 (Fed. Cir. 2015). Words like “shade” for tanning, “logic” for test prep, or a Bible verse on apparel are common examples.
- Careful buyers (factor 4). A company buying robotics simulation software, or a student choosing an LSAT course, does not buy on impulse. Evidence of price, research and purchase process supports this factor.
- Different trade channels (factor 3). This works best when the identifications themselves limit the channel or the buyer. That is why narrowing the description matters.
- A narrower identification. You can amend your description to clarify or limit the goods, for example “none of the foregoing in the nature of physical fitness services.” You cannot broaden it. A well-drafted limit can remove the overlap the refusal depends on.
- Different meaning and impression. A name that comes from a founder, a coined word, or a phrase that reads as a company name can create a different commercial impression from a cited mark that shares one word.
One recent example: FIT FROM HOME (Serial No. 99461715) was refused over FIT AT HOME. The response put evidence of the weak, commonly used shared terms on the record, and the mark was approved for publication shortly after.
One caution: arguments alone are not evidence. We cover this in Who Has the Burden of Proof in a 2(d) Refusal? Why Argument Is Not Evidence. Every factor you raise needs something attached: registrations, web pages, pricing, a declaration from you as the person who knows the market.
5. What won’t move the examiner
Some of the most natural reactions to a refusal do little or nothing. Knowing them saves time and goodwill.
- “The cited company doesn’t even use its mark anymore.” In an ex parte application, you cannot attack the validity of the cited registration. In re Dixie Restaurants, Inc., 105 F.3d 1405 (Fed. Cir. 1997). Non-use is handled in a separate proceeding, not in your response.
- “We’ve never had a single complaint.” On its own, an applicant’s statement that it knows of no confusion carries little weight in an ex parte case. In re Majestic Distilling Co., 315 F.3d 1311 (Fed. Cir. 2003). It can help as part of a larger record, but it will not carry the response.
- Evidence of how the other company actually sells. The comparison is based on the goods as identified in the registration, not on the owner’s real-world business, and not on foreign registrations.
- An angry tone. The same examining attorney reviews your response and any later request for reconsideration. You cannot ask for a different examiner. A firm, cooperative response that helps them reach a better answer tends to work better than one that tells them they made a mistake.
- Asking the examiner what to do. Examining attorneys cannot give you legal advice about how to overcome their refusal. Save the effort for the evidence.
6. Possible or probable? Measuring your odds
The legal standard is likelihood. The Federal Circuit has said the law is not concerned with “mere theoretical possibilities of confusion.” Electronic Design & Sales, Inc. v. Electronic Data Systems Corp., 954 F.2d 713 (Fed. Cir. 1992). Our guide Likelihood of Confusion Means Probable, Not Possible walks through that line in detail.
The practical question for you is simple: does your refusal fall on the possible side or the probable side? The answer decides whether to respond with evidence, narrow your goods, or think about a different path. Guessing is expensive. A response takes real hours to prepare, and a rebrand costs far more.
Illustration only. The TRACE™ Confusion Index scores a refusal from 0 to 10. Below 7.00 the refusal is treated as contestable; at or above it, confusion is treated as probable. It is a proprietary metric, not a USPTO or TTAB standard.
The TRACE™ Report scores a 2(d) refusal across nine factors and 180 benchmarks calibrated to more than 600 decided TTAB cases. It also tells you where the examiner’s case is weakest and which arguments to lead with. It costs $25, arrives within 24 hours, and is credited in full if you go on to respond. For most applicants, that is a small price to know which side of the line they are on.
↑ Back to guideHow to: 10 steps after an “obviously different” refusal
- Find your deadline. Check the issue date in TSDR. You have three months to respond, with one three-month extension available for a fee.
- Read the refusal for its factors. Mark every sentence about the marks and every sentence about the goods. Note which factors are never mentioned.
- Pull the cited registration. Read its exact identification of goods, its filing date and its status in TSDR.
- Compare the marks through all four lenses. Write down how they differ in appearance, sound, meaning and commercial impression, as a whole.
- Identify the dominant part of each mark. Ask whether the cited mark leads with a different term than yours.
- Search for the crowded field. Look for live registrations that share the common term for similar goods. Quantity and relevance both matter.
- Draft a narrower identification. Limit your goods to what you actually offer, and consider exclusionary wording that removes the overlap.
- Document your buyers and channels. Gather pricing, purchase process and customer type as evidence, not just statements.
- Score the refusal before you write. Decide whether the record points to possible or probable confusion. A TRACE™ Report gives you that answer in a day.
- File early and keep the tone cooperative. Organize the response factor by factor, attach every exhibit, and file well before the deadline.
What if: 10 scenarios
Then relatedness is your strongest ground. Check whether the examiner’s evidence shows the same company offering both, or only that both exist online.
Collect third-party registrations and uses. A weak shared term gets a narrow scope of protection, and small differences can be enough.
Your response must address each one. Start with the most problematic; if there is no confusion with that one, the others are usually easier.
You cannot argue non-use in your response. That is a separate cancellation matter, so build your likelihood-of-confusion case in the meantime.
That origin can support a different meaning and commercial impression, but buyers who don’t know the story see only the word. Pair it with evidence.
Narrow it. You can delete or limit goods in a response. Over-broad lists are a common cause of avoidable refusals.
You can file a request for reconsideration, a notice of appeal to the TTAB, or both. The arguments you made first still matter, so make them complete the first time.
A new application goes to a new examination, and the same conflict may be cited again. Abandoning also gives up your filing date.
That can draw attention to your brand and invite a challenge. Weigh it carefully before reaching out.
The application is abandoned. A petition to revive may be possible if you act quickly, and it carries a government fee.
Most of these scenarios turn on one question: how strong the refusal really is. That is worth knowing before choosing a path, and it is what a TRACE™ Report answers.
↑ Back to guideFAQs: 10 questions
Why did the USPTO refuse my mark when the logos look nothing alike?
Marks are compared by sight, sound, meaning and commercial impression. If the wording sounds alike or shares a dominant term, a different logo may not be enough on its own.
Does the examining attorney consider all thirteen DuPont factors?
Only the factors with evidence in the record. A first refusal usually relies on the similarity of the marks and the relatedness of the goods. Your response can add evidence on the others.
Is a first Office Action a final decision?
No. A non-final Office Action gives you three months to respond with arguments, evidence and amendments. A final action comes only after that.
Do I have to prove there will be no confusion?
The USPTO must show confusion is likely. But in close cases doubt goes to the registrant, so the practical task is to give evidence that removes that doubt.
Can I change my goods or services to overcome the refusal?
Yes, you can clarify or narrow them. You cannot broaden them or add new goods to the application.
Does it matter that I’ve used my mark for years with no problems?
It can help as part of a larger record, but a bare statement that you know of no confusion carries little weight on its own in an application.
Should I call or email the examining attorney?
You can contact them for procedural questions, but they cannot advise you on how to overcome the refusal. The response is where you make your case.
Is filing a response without a lawyer a disadvantage?
Responses are decided on their merits and the evidence in the record. What matters most is a complete, well-organized response filed on time.
How long does the USPTO take after I respond?
It varies. Some responses are reviewed within days, others take weeks or months. Filing early gives you more room if a second round is needed.
How can I tell if my refusal is worth fighting?
Measure whether confusion is probable or only possible across all the factors. The TRACE™ Report scores your refusal against more than 600 TTAB decisions and returns a Confusion Index within 24 hours.
Your next move
“Obviously different” is often true. It just hasn’t been shown yet. The first refusal tells you which two questions the examiner answered. Your response answers the rest. Start by reading the refusal for what it leaves out, and gather evidence for exactly those factors.
Find out which side of the line your refusal falls on.
Your TRACE™ Report shows whether the refusal can be beaten, where the examiner’s case is weakest, and which arguments to lead with. $25, within 24 hours, credited in full if you respond.
Get your TRACE Report →Trademark Scrivener is an independent trademark administrator, not a law firm. This article is general information about USPTO procedure, not legal advice, and reading it does not create any client relationship. For advice about your specific situation, consult a licensed trademark attorney. The Confusion Index is a proprietary metric, not a USPTO or TTAB standard. Examples are drawn from anonymized client situations and are for illustration only.