How to Respond to a Section 2(d) Office Action: A Data-Driven Guide

You just opened an Office Action. The examiner cited Section 2(d). Your client's application is refused based on "likelihood of confusion" with a prior registered mark.

Before you draft a response based on instinct, consider this: we analyzed 1,687 Section 2(d) cases to identify which arguments actually work. The data reveals patterns that most attorneys miss—and strategies that succeed far more often than others.

This guide walks you through a systematic, data-backed approach to responding to Section 2(d) refusals. You'll learn which arguments have the highest success rates, how to structure your response, and how to avoid the mistakes that sink otherwise strong applications.

Understanding What the Examiner Is Actually Saying

A Section 2(d) refusal means the examining attorney believes your mark, when used with your identified goods/services, so resembles a registered mark that it's likely to cause confusion among consumers.

The examiner applied the DuPont factors—a 13-factor test established in In re E.I. DuPont DeNemours & Co., 476 F.2d 1357 (CCPA 1973). Not all factors apply to every case, and no single factor is determinative. The two most important factors:

1. Similarity of the marks (in appearance, sound, connotation, and commercial impression)
2. Relatedness of the goods/services

The examiner's job is to err on the side of protection. Your job is to show why confusion is unlikely despite surface-level similarities.

For a deeper dive into how the DuPont factors work in practice, see our DuPont factors guide.

Step 1: Analyze the Cited Mark—Before You Write a Word

Pull up the cited registration on our trademark database or the USPTO Trademark Search. Answer these questions:

Is the cited mark actually live?

Check the status. If it's abandoned, cancelled, or expired, you may have an immediate win. Our data shows cited mark abandoned arguments succeed 85% of the time when properly documented.

What to check:

• Status on USPTO TSDR
• Last maintenance filing date
• Whether the owner responded to the most recent Section 8/15 deadline

If the mark is dead, your argument is straightforward: there's no likelihood of confusion with a mark that's no longer in use. Cite TMEP § 1207.01(d)(iii) and provide evidence of abandonment.

What goods/services does it actually cover?

Read the ID of goods/services carefully. Don't assume relatedness based on the mark alone. Examiners sometimes cite marks with only tangential overlap.

What to look for:

• Specific limitations in the ID (e.g., "for use with X" or "excluding Y")
• Different classes entirely
• Overlapping language that's actually describing different products

Example: An examiner might cite a mark registered for "software for managing employee schedules" against your application for "software for tracking customer appointments." Both are software, but the end users and use cases are entirely distinct.

How similar are the marks, really?

Strip away the examiner's comparison and look objectively:

• Appearance: Are they visually similar, or does font/design create a different impression?
• Sound: Do they sound alike when spoken? (This matters more in some industries than others)
• Connotation: Do they suggest the same concept, or are they completely different ideas?
• Commercial impression: Would a consumer see them as related, or as competitors?

The examiner has already concluded they're similar. Your job is to find the distinguishing features the examiner underweighted.

Does the cited mark have a consent history?

Search for the cited mark's serial number on TSDR and check its prosecution history. Sometimes marks have already coexisted with similar marks via consent agreements. This is powerful evidence that the mark is not as strong as the examiner assumes.

You can also check for third-party registrations of similar marks in the same class. If the register is crowded with variants, the cited mark is weaker—and less entitled to protection.

Step 2: Choose Your Strategy Based on the Data

We analyzed 1,687 Section 2(d) cases and ranked argument strategies by frequency and effectiveness. Here's what works:

Strategy	Times Used	Effectiveness	When to Use
Consent Agreement	40	0.90	You can get the cited mark owner to agree
Class Deletion	64	0.86	You're willing to drop one or more classes
Cited Mark Abandoned	55	0.85	The cited mark is dead or lapsed
Supplemental Register	99	0.80	Your mark is descriptive or weak anyway
Commercial Impression	73	0.78	Marks create different consumer impressions
Goods/Services Coexistence	143	0.77	Goods/services don't actually compete
Third-Party Registrations	61	0.77	Crowded register = weaker cited mark
Weakness of Cited Mark	66	0.77	The cited mark is descriptive, generic, or diluted
Goods/Services Amendment	133	0.76	You can narrow your ID to avoid overlap
Market/Channel Distinction	76	0.76	Different distribution channels or customer bases
Consumer Sophistication	55	0.75	Your customers are highly discerning (e.g., B2B software)
Mark Distinction	114	0.72	Marks differ in key ways the examiner missed

Key insight: The highest-effectiveness strategies (consent, class deletion, abandonment) require specific circumstances. The next tier—commercial impression, coexistence, and weakness arguments—are your workhorses for most cases.

Let's break down the top strategies.

Consent Agreement (0.90 effectiveness)

If you can get the owner of the cited mark to agree in writing that there's no likelihood of confusion, the examiner will almost always withdraw the refusal.

When it works:

• The cited mark owner is reachable and cooperative
• You can demonstrate coexistence without confusion (e.g., you've been using the mark for years without issues)
• The goods/services are complementary, not competitive

How to do it:

1. Find the owner's contact info via TSDR (look for the correspondence address)
2. Draft a consent agreement (there are templates, but customize to your facts)
3. Explain why coexistence benefits both parties (or why there's no real conflict)
4. File the signed agreement with your Office Action response

Pro tip: Consent agreements work best when you can show the other party that your use doesn't threaten their brand. If you're in different markets or customer segments, emphasize that.

Class Deletion (0.86 effectiveness)

If your application covers multiple classes and the conflict is limited to one class, deleting that class resolves the refusal.

When it works:

• Your client doesn't actually need protection in the conflicting class
• The other classes are more important to the business
• The cost of fighting the refusal exceeds the value of the dropped class

Example: You file for "software" in Class 9 and "consulting services" in Class 35. The examiner cites a mark in Class 9 only. If your client's revenue is 90% consulting, delete Class 9 and move forward.

How to do it: File a request to divide the application (if needed) or simply amend to delete the conflicting class. Cite TMEP § 1403.01 for partial refusals.

Cited Mark Abandoned (0.85 effectiveness)

If the cited mark is dead, your path is clear.

What counts as abandoned:

• Mark is listed as "Abandoned" or "Cancelled" on TSDR
• Owner failed to file Section 8 or Section 15 declaration
• Mark expired after 10 years (or 20 years for older registrations)

How to argue it:

1. Provide evidence from TSDR showing the abandonment date
2. Cite TMEP § 1207.01(d)(iii): "A mark that is no longer used or registered cannot serve as a bar to registration."
3. Request withdrawal of the refusal based on non-existence of the cited mark

Common mistake: Don't just say "the mark is abandoned"—attach proof. Examiners want to see TSDR printouts or declaration dates.

Supplemental Register (0.80 effectiveness)

If your mark is descriptive or weak, you may not qualify for the Principal Register anyway. The Supplemental Register provides many of the same benefits (use of ® symbol, basis for international filings) without the strict confusion analysis.

When it works:

• Your mark is descriptive and you're not claiming acquired distinctiveness
• You're willing to forgo certain Principal Register benefits (e.g., incontestability after 5 years)
• You need the registration quickly and don't want to fight

How to do it: File an amendment to the Supplemental Register under Section 23. This moots the Section 2(d) refusal (likelihood of confusion doesn't apply to the Supplemental Register).

Trade-off: You lose some enforcement power, but you get a registration in ~3-6 months instead of fighting for 12+ months.

Commercial Impression (0.78 effectiveness)

Even if the marks are visually or phonetically similar, they may create entirely different commercial impressions.

When it works:

• Marks suggest different concepts (e.g., "Delta" for airlines vs. "Delta" for faucets)
• Design elements or stylization create distinct impressions
• The overall "feel" of the marks is different despite surface similarity

How to argue it:

1. Describe the connotation of each mark
2. Explain how consumers perceive each mark in context
3. Cite case law on commercial impression (e.g., In re Chatam Int'l Inc., 380 F.3d 1340 (Fed. Cir. 2004))

Example: "BLUEBIRD" for financial services vs. "BLUEBIRD" for children's toys. Both use the same word, but the commercial impression—serious vs. playful—differs dramatically.

Goods/Services Coexistence (0.77 effectiveness)

This is your most common argument: the goods/services don't actually compete, so confusion is unlikely.

When it works:

• Different end users (e.g., B2B software vs. consumer software)
• Different distribution channels (e.g., high-end boutiques vs. mass-market retail)
• Different use cases (e.g., "software for architects" vs. "software for accountants")

How to argue it:

1. Compare the IDs side by side
2. Highlight limitations and distinctions
3. Explain how the products/services serve different markets
4. Cite DuPont factor: relatedness of goods/services

Example: Examiner cites a mark for "clothing" against your mark for "athletic footwear." Argue that clothing and footwear are distinct product categories, sold in different departments, with different customer bases.

Pro tip: If the examiner relied on broad language like "clothing and footwear are both fashion items," push back. Cite the ID language verbatim and show non-overlap.

Third-Party Registrations (0.77 effectiveness)

If the register is crowded with similar marks, the cited mark is entitled to a narrower scope of protection.

When it works:

• You find 5+ third-party registrations with similar elements
• The marks coexist in the same or related classes
• The cited mark is not particularly distinctive

How to do it:

1. Search the USPTO Trademark Search for marks with the same or similar terms
2. Print registrations that show coexistence
3. Argue that the cited mark is weak and consumers are accustomed to distinguishing among similar marks

Example: If you're fighting a refusal based on "SUMMIT" for financial services, and you find 10 other "SUMMIT"-formative marks in Class 36, that's strong evidence of mark weakness.

Cite: In re E.I. DuPont, 476 F.2d at 1361 (third-party use is relevant to mark strength and consumer recognition).

Weakness of Cited Mark (0.77 effectiveness)

If the cited mark is descriptive, generic, or highly suggestive, it's entitled to less protection.

When it works:

• The cited mark describes a feature of the goods/services
• The mark is a common term or phrase
• There's evidence of third-party use of the same term

How to argue it:

1. Show that the mark is descriptive or suggestive (cite dictionary definitions, industry usage)
2. Provide evidence of third-party use (websites, advertisements, other registrations)
3. Argue that the mark's scope of protection is narrow

Example: "CLOUD" for cloud computing services is highly suggestive. If the examiner cites a "CLOUD"-formative mark, argue that the term is widely used and consumers can distinguish among variants.

Cite: In re Dixie Restaurants, Inc., 105 F.3d 1405 (Fed. Cir. 1997) (weak marks entitled to narrow scope of protection).

Step 3: Build Your Argument—Practical Tips

Once you've chosen your primary strategy, strengthen it with these tactics:

Use precise comparisons

Don't just assert differences—show them. Create a side-by-side comparison:

Factor	Applicant Mark	Cited Mark
Visual appearance	[Describe]	[Describe]
Sound/pronunciation	[Describe]	[Describe]
Meaning/connotation	[Describe]	[Describe]
Commercial impression	[Describe]	[Describe]

This forces the examiner to see the differences on paper.

Cite DuPont factors explicitly

Don't make the examiner guess which factors you're addressing. Say it outright:

"With respect to DuPont factor 2 (similarity of goods/services), Applicant's goods are distinct from the cited mark's goods in the following ways..."

Use evidence, not argument

The strongest responses are evidence-heavy:

• Screenshots of the cited mark in use (to show commercial impression)
• Third-party registrations (to show crowded register)
• Dictionary definitions (to show meaning differences)
• Industry publications (to show distinct markets or channels)

Address the examiner's specific points

The Office Action will explain why the examiner found confusion likely. Don't write a generic response—address each point.

If the examiner said, "Both marks include the term 'PEAK' and are used in connection with outdoor recreation," your response should tackle both elements:

1. Explain why "PEAK" is weak in the outdoor industry (third-party use)
2. Show that your goods (e.g., hiking boots) are distinct from the cited mark's goods (e.g., camping tents)

Don't overargue

More is not better. Pick 2-3 strong arguments and develop them fully. A scattershot response ("the marks are different AND the goods are different AND consumers are sophisticated AND...") signals desperation.

Focus on your best argument and support it with evidence.

Step 4: Structure Your Response

A well-organized response makes the examiner's job easier—and increases your chances of success.

Suggested structure:

I. Introduction

• Thank the examiner for their review
• State that you're respectfully traversing the Section 2(d) refusal
• Briefly preview your argument (1-2 sentences)

II. Background

• Describe your mark and goods/services
• Describe the cited mark and goods/services
• Acknowledge the examiner's concerns

III. Legal Standard

• Briefly summarize the DuPont test
• Note that no single factor is determinative
• State which factors are most relevant to this case

IV. Argument

• Primary argument (e.g., goods/services coexistence)
  • Evidence
  • Analysis
  • Citation to DuPont factor(s)
• Secondary argument (if applicable)
  • Evidence
  • Analysis

V. Conclusion

• Summarize why confusion is unlikely
• Request withdrawal of the refusal and approval for publication

Attachments:

• Third-party registrations
• Consent agreement (if applicable)
• Evidence of cited mark abandonment
• Dictionary definitions, screenshots, etc.

Length: Quality over quantity

Some responses succeed in 3 pages. Others require 15 pages. Don't pad—but don't skimp on evidence.

Aim for 8-12 pages with exhibits. Shorter if your argument is straightforward (e.g., cited mark abandoned). Longer if you're making multiple arguments or the case is complex.

Common Mistakes to Avoid

Mistake 1: Ignoring the examiner's rationale

The examiner explained their reasoning in the Office Action. If you don't address it, you're talking past each other.

Fix: Quote the examiner's language and respond directly.

Mistake 2: Arguing facts not in evidence

Don't claim "consumers are sophisticated" without proof. Don't say "our mark is famous" without evidence.

Fix: Use market surveys, sales data, customer demographics, or industry publications to support factual claims.

Mistake 3: Attacking the cited mark owner

It's not the owner's fault the examiner cited their mark. Avoid language like "the cited mark is weak and shouldn't have been registered."

Fix: Focus on coexistence, not blame. Frame it as "both marks can coexist without confusion."

Mistake 4: Missing the deadline

You have 6 months from the issue date to respond. Miss it, and the application is abandoned.

Fix: Calendar the deadline immediately. If you need more time, file an extension (you get one automatic 6-month extension under TMEP § 1714).

Mistake 5: Amending the mark to avoid confusion

You cannot amend a mark to make it "less similar" to the cited mark. The USPTO treats that as a new mark, requiring a new application.

Fix: Work with the mark as filed. If the mark truly can't proceed, advise the client to file a new application with a different mark.

Mistake 6: Over-relying on intent-to-use status

Some attorneys argue, "We're not using the mark yet, so there's no actual confusion." This doesn't matter. The USPTO examines based on potential confusion, not actual use.

Fix: Focus on DuPont factors, not your basis for filing.

When to Consider Alternatives

Not every refusal is worth fighting. Here's when to consider other paths:

Switch to the Supplemental Register

When: Your mark is descriptive and you're not claiming acquired distinctiveness.

Benefit: Quick approval (no likelihood of confusion analysis).

Trade-off: Weaker enforcement rights, no incontestability.

Our data: Supplemental Register arguments succeed 80% of the time when the mark is genuinely descriptive.

Amend your goods/services

When: You can narrow your ID to eliminate overlap with the cited mark.

Benefit: Preserves your mark on the Principal Register.

Trade-off: You lose coverage for certain goods/services.

Our data: Goods/services amendment arguments succeed 76% of the time when the amendment meaningfully distinguishes the IDs.

Request a consent agreement

When: You have a relationship with the cited mark owner or can demonstrate non-competitive use.

Benefit: Highest success rate (90%) of any strategy.

Trade-off: Requires cooperation from a third party.

Abandon and refile with a different mark

When: The mark is inherently confusing and no argument will work.

Benefit: Avoids wasting time on an unwinnable case.

Trade-off: You lose your filing date and application fees.

Red flags for abandonment:

• Near-identical marks in the same class
• Highly related goods/services
• Strong cited mark with extensive use
• No credible distinction argument

If your analysis shows a less than 30% chance of success, counsel the client to pivot.

The Data Advantage: How GleanMark Can Save You 4-6 Hours Per Response

Drafting a Section 2(d) response typically takes 4-6 hours of attorney time:

• 1 hour analyzing the cited mark
• 1-2 hours researching case law and third-party registrations
• 2-3 hours writing and formatting the response

GleanMark's AI-powered Office Action drafting tool automates the research phase:

• Instant analysis of the cited mark (status, goods/services, prosecution history)
• Automated third-party registration search (finds similar marks in seconds)
• Strategy recommendations based on the data in this article
• Draft argument sections pre-populated with case citations and DuPont factor analysis

You still review and refine—but the heavy lifting is done. That 6-hour response becomes a 90-minute review.

Want to see it in action? Search our trademark database or run a clearance search to see how our tools surface insights other platforms miss.

Final Thoughts: Strategy Over Volume

Our analysis of 1,687 Section 2(d) cases revealed a clear pattern: attorneys who win focus on 2-3 strong arguments backed by evidence. The responses that fail tend to throw everything at the wall—10 weak arguments instead of 2 strong ones.

Choose your strategy based on the facts. Use the data in this guide to pick arguments with proven effectiveness. Build your case with evidence, not assertion.

And remember: the examiner is not your adversary. They're applying a legal standard to protect the public from confusion. Your job is to show them—clearly, respectfully, and convincingly—why confusion is unlikely in this specific case.

Average time to registration after a successful Section 2(d) response: 128 days. You're not starting over—you're navigating a detour. With the right strategy, you'll get there.

---

For a deeper dive into the data behind these strategies, see our analysis of 101 winning arguments. To understand how the examiner evaluates likelihood of confusion, read our DuPont factors guide.