Case, Board, and file wrapper citations in this article link to the documents in ProSearch. A free Design Patent Pro account is all that is needed to open them.
Introduction
Every reference in a design patent obviousness rejection must now be analogous art to the claimed design. LKQ v. GM, 102 F.4th 1280, 1296-99 (Fed. Cir. 2024) (en banc). In LKQ, the court applied the first prong of the utility test, the same field of endeavor, and said the second prong, a reference reasonably pertinent to the problem, does not seem to apply to designs in the same way. It left the rest for later cases. This article examines how examiners and applicants have handled the requirement in the two years since, using 84 file wrappers in which the question was argued rather than only recited. The short answer is that applicants raise the argument often, examiners answer it by defining the field more broadly than the article, and the argument decides the case in about one file in eight. The article closes with what the record suggests about when the argument is worth making and how to make it.
1. What the Court Decided and What It Left Open
The en banc court reaffirmed that “an analogous art requirement applies for obviousness of design patents.” 102 F.4th at 1296. It then addressed the two-part utility test. The first part asks whether the reference is from the same field of endeavor as the claimed invention. The second asks whether a reference outside that field is reasonably pertinent to the particular problem the inventor faced. The court held that the first part applies, and that for designs the field is defined by the article of manufacture. The primary reference will typically be in the same field of endeavor as the claimed design’s article of manufacture, but it need not be, so long as it is analogous art. Id. at 1297-98. On the second part the court wrote that it “does not seem to apply to design patents in the same way, and how to translate this part of the test into the design context is less apparent.” Id. at 1297. It declined to “delineate the full and precise contours of the analogous art test for design patents” and said the question is one of fact to be addressed case by case. Id. The court quoted the older formulation from Hupp v. Siroflex of America, 122 F.3d 1456, 1462 (Fed. Cir. 1997), that the scope of the prior art is “designs of the same article of manufacture or of articles sufficiently similar that a person of ordinary skill would look to such articles for their designs.” Id. at 1297.
Three things follow. The requirement applies to secondary references as well as primary references. The field is tied to the article, so the title and claim language that name the article matter. And nobody has said what “sufficiently similar” means, whether the second prong exists for designs, or what evidence establishes the field. The Office’s memorandum of May 22, 2024 told examiners that both primary and secondary references must be analogous art and repeated the court’s language. It did not add a rule.
2. The Record
The corpus is 84 file wrappers of design applications that issued between August 1, 2024 and September 20, 2026, retrieved with the ProSearch query “obvious to a designer of ordinary skill” AND (“not analogous” OR “non-analogous” OR “nonanalogous” OR “reasonably pertinent”). The first phrase is examiner form language for a design obviousness rejection. The others are the words an applicant or an examiner uses when the requirement is contested rather than recited. A broader query for the phrase “analogous art” alone returns 174 files, and a query for “field of endeavor” returns 168, most of them examiner recitals that the applied references are in the same field. The 84 are the files in which someone argued the point.
In 58 of the 84 the exchange took place entirely after May 21, 2024. In 15 it began before and continued after. Ten are pre-LKQ exchanges that used the utility test on their own. The corpus is drawn from issued patents, so applications that were abandoned after the argument failed are not in it. That is a real limit for this topic, and the success figures below should be read as figures for applications that eventually issued. Thirty-two of the 84 are also in the 171-file corpus used in the companion article on the reason to combine.
Figure 1. Five counts from the 84 files.
3. What Examiners Are Finding
The Office’s guidance calls for an analogous art finding for each reference. In practice the finding is often missing or is only a recital. In 35 of the 84 files the examiner never made an express finding on analogous art. In 8 the only statement was the form sentence that “an analogous art requirement applies to each reference.” In 41 the examiner named a field of endeavor. In 4 of those the examiner also addressed the second prong.
What the examiner named as the field is the most consistent pattern in the corpus. Where a field was named, it was the claimed article itself in only 2 files. In 30 it was a category broader than the article. In 8 it was a function or a problem. Examples of the broader category: a mailing tube and a wine bottle wrapping are both “packaging items” (29/866,534). A printer control panel and a point-of-sale printer are both “types of printers,” and “[o]peration panels are elements on printers and are in the same field of endeavor” (29/744,291). A wine case is in the field of “partitioned container systems” (29/895,014). A knee support and a sock are “form fitting soft goods with stitching that are worn on the lower extremities” and “in foot/legwear art areas” (29/866,742). A toy robot is the same article of manufacture as a delivery robot, with “delivery” treated as intended use (29/671,792).
Examples of the functional definition: a dental capsule and an auto-injector are both “devices which inject material into the body” (29/781,414, 29/781,417, 29/781,418). An oral irrigator tip and a saliva suction tip are “articles for introduction, removal or collection of fluids” (29/826,538). An extractor insert and a nut share the field of “applying torque, even if the claimed design is for the specific purpose of applying torque to a damaged bolt,” so “any torque applying device would be within the general field of endeavor” (29/854,126). A wall mount for an audio device and an arm rest bracket are analogous because “both the prior art and the claimed design are mounts/brackets that hold or support objects” (29/789,995). Two filter pouches are in the field of “filtering liquid,” with the examiner quoting MPEP 904.01(c) that analogousness “depends upon the necessary essential function or utility of the subject matter covered by the claims, and not upon what it is called by the applicant” (29/880,660). One examiner stated the principle directly: “even ornamental designs for articles of manufacture ultimately serve a function of some kind, though considering this function is chiefly a means to determine analogousness, not patentability” (29/875,881).
Figure 2. How the examiner defined the field of endeavor. Where a field was named, the examiner used the claimed article itself in two files.
Graphical user interfaces are a category of their own. Examiners in Art Unit 2923 treat all GUIs and icons as one field because the article of manufacture is the display screen. An interview summary in 29/810,013 records the position: “all graphical user interfaces/icons are considered analogous art; designing GUIs is a single field of endeavor, and 2D planar display screens are the article of manufacture on which the design is embodied, regardless if an amended title describes additional environment or articles.” The same examiner said in 29/949,803 that “the designer skilled in the art is considered to be a designer of graphical user interfaces, who designs many different GUIs for different clients,” and that analogousness is “based upon an ornament for the surface of a display screen, not on a functional computer program or device.” The Board adopted this view in the one decision in the corpus that reached the question, discussed in Part 6.
Two older formulations survive. Seven files recite the In re Glavas passage that “[t]he principle of nonanalogous arts, therefore, cannot be applied to design cases in exactly the same manner as to mechanical cases,” followed by the “so related” test that LKQ overruled. One of them is an Examiner’s Answer (29/663,384). Another examiner cited In re Antle and In re Nalbandian for the proposition that “one skilled in the art is charged with knowledge of all related art” as the reason an arm rest bracket was analogous to an audio wall mount (29/789,995). Those citations answer the level of skill, not the field.
Four examiners used the second prong against an applicant. In 29/771,650 the examiner defined “pad” from the Oxford English Dictionary as an object “used to absorb or spread a liquid” and reasoned that a designer of vapor pads “would be motivated to look at catamenial pads for their ability to absorb a large volume of liquid,” so the references were both in the same field and “reasonably pertinent to the particular problem faced by the inventor.” In 29/854,126 a fastening nut was “reasonably pertinent to the problem faced” of loosening a damaged bolt. In 29/693,559 and 29/838,770 the examiner quoted the utility formulation that a reference is pertinent if it “logically would have commanded itself to an inventor’s attention in considering his problem.” The court said this prong does not seem to apply to designs in the same way. These examiners applied it in the utility form without comment.
Evidence for the field, where any was offered, came from dictionaries (29/771,650 and 29/866,534 by the examiner, 29/787,268 by the applicant), the Office’s design classification (29/866,534, 29/838,770), the classification of the cited utility patents (29/790,572, where the examiner said the first step in deciding analogousness “is to look at where the utility applications are classified”), and, on the applicant’s side, product listings, a video, and a parent utility application (29/963,012, 29/858,525, 29/921,272). Most files contain no evidence on the field at all. The examiner names a category and the applicant names the article.
4. What Applicants Are Arguing
Applicants raised the argument in 70 of the 84 files. Most argued the first prong only: the reference shows a different article, so it is in a different field. Nineteen argued both prongs. Thirty-nine cited LKQ by name, and several quoted the May 22, 2024 memorandum. Thirty-three aimed the argument at every applied reference, 19 at the secondary reference only, and 14 at the primary reference only.
The article-level argument takes a simple form. The title says what the article is. The reference is a different article. Therefore the reference is in a different field. Applicants argued that a shoe is not a plant pot (29/909,266), a rail stiffener is not packaging material (29/803,617), a seed pod is not aquarium filtration media (29/839,768), a wireless tracking device is not a blister envelope (29/940,303 and 29/940,304), a vapor pad is not a menstrual pad (29/771,650), and a medical control console is not a patient monitor (29/882,266). For GUIs, applicants argued that the software context defines the field, so a shopping cart screen is not a home screen (29/841,307), a slide-to-close control is not an arrow icon (29/841,323), and a media player interface is not a glucose monitoring icon (29/949,803).
A smaller group of applicants proposed rules for the second prong. Two argued that a reference outside the field is analogous only with evidence that “the particular problem being solved in the cited art is of particular and significant relevance to the specific article of manufacture in the claimed design,” and that a design specification cannot supply the “particular problem” because the specification does not describe one (29/669,161 and 29/937,714). One quoted the court’s statement that the second prong does not seem to apply to designs in the same way and argued that the article of manufacture is the only proper focus (29/882,266). One argued that the examiner must make a second-prong showing once the same-field showing fails, and that the test is otherwise the same as for utility patents (29/853,212). One applied the full utility test to a yoga mat and used the parent utility application to prove the field (29/921,272). No examiner in the corpus adopted any of these proposals.
A few applicants used the title as the tool. In 29/771,650 the applicant amended “Vapor Pad” to “Vapor Pad for Vaporizer, Diffuser, Humidifier, or the Like” and argued under Curver Luxembourg v. Home Expressions, 938 F.3d 1334 (Fed. Cir. 2019), that the title limits the claim to that article. In 29/992,406 the applicant narrowed the title of a toy ball for the same purpose. In 29/810,013 the applicant proposed adding “gaming machine” to a GUI title, and the examiner said the amendment would not narrow the field because the article remains the display screen.
5. How the Arguments Are Faring
The argument decided the case in 10 of the 84 files. In 30 the examiner rejected the argument and then allowed the application on another ground, almost always a visual difference between the claimed design and the primary reference. In 14 the examiner rejected the argument and maintained the rejection, and the file ended by amendment, continued prosecution, or appeal. In 11 the examiner allowed the application as “persuasive” without saying which argument persuaded. In 3 an amendment mooted the rejection before the examiner ruled. In 13 the applicant never contested the point. Three went to the Board and were reversed on the reason to combine, not on analogous art.
Figure 3. How the analogous art argument ended. The examiner rejected the argument in 44 files and accepted it in 10.
The ten wins have a common feature. In each the reference was an article that no one would confuse with the claimed one, or the applicant made it so by amending the title. The examiner in 29/616,972 withdrew a rejection of a paper product over “a tape edge for a diaper enclosure and not a paper product,” citing In re SurgiSil for the point that “the claim identifies a paper product and the claim is limited to paper products and does not cover other articles of manufacture.” The examiner in 29/787,268 withdrew because a website user avatar was not an icon, after the applicant supplied a dictionary definition of “icon.” The examiner in 29/803,617 accepted that packaging material is not analogous to a rail stiffener for a photovoltaic device. The examiner in 29/806,782 accepted that a hammer tip is not analogous to an erasing wheel. The examiner in 29/839,768 accepted, before LKQ, that aquarium filtration media is not analogous to a seed pod. The examiner in 29/851,947 wrote that the primary reference “is not analogous art to the claimed design” without saying more. The examiner in 29/778,252 conceded that an origami paper was not analogous to a GUI and substituted new references.
Two of the ten deserve separate mention. In 29/771,650 the examiner had applied both prongs against the applicant, using the dictionary definition of “pad.” After the title amendment and an appeal brief the examiner withdrew: “The title change in the response disclosed in the applicant arguments of 8/13/24 has sufficiently narrowed the scope of the claim so that the references are considered nonanalogous, that the references are not pertinent to the problem faced by the inventor.” That is the only file in the corpus in which an examiner applied the second prong and then conceded it. In 35/519,721 and 35/519,733, companion applications for a mobility station structure, the examiner raised the point. The interview summary records that the examiner advised the applicant that after LKQ all cited references must be analogous art and suggested arguing that the secondary reference was not. The withdrawal reads: “While the primary reference is considered analogous, the secondary reference would not be considered entirely analogous to the claim for a structure. The rejection of record under 35 USC 103 has been withdrawn.”
The largest category, 30 files, is the argument rejected and the application allowed anyway. The pattern is visible in 29/866,534. The applicant argued that a wine bottle wrapping and a packaging insert are not analogous to a mailing tube. The examiner allowed on visual differences and then added an Examiner’s Comment: “Examiner respectfully disagrees in that the claimed design and the prior art all are considered packaging items. Packaging is defined as ‘materials used to wrap or protect goods.’ Wrapping for a bottle and a packaging insert are items that are understood to be used for those purposes and are, therefore, analogous to the claimed design and its packaging classification.” The examiner in 29/744,291 did the same for a printer control panel: “Although the rejection is withdrawn, the argument that the primary and secondary references are not analogous is incorrect.” In 29/882,266 the examiner allowed a medical control console on proportion and wrote that the applicant’s own description of operating rooms and recovery rooms was “proof of analogousness” because the articles are used in similar hospital spaces. Examiners are answering the argument even when nothing turns on it. The answers are the closest thing to a rule the Office has produced.
In 30 files the examiner’s response to the argument was to define the field more broadly than the article named in the title. That number is the same as the number of files in which a field was defined as a broader category, and it is the same set. The examiner did not take the applicant’s article and ask whether the reference shows the same article. The examiner took the reference and the article and named a category that contains both.
6. What the Board Has Said
One Board decision in the corpus addresses analogous art under LKQ. In Appeal No. 2023-001595 (Application No. 29/678,877, decided September 20, 2024), the claim was for a display screen with a machine learning confidence score GUI, showing a lightbulb and a line. The applied references were a chemical test dial, a battery settings screen, and a washing machine cycle display. The appellant argued that none was in the “machine learning confidence score graphical user interface” field and none was pertinent to machine learning confidence scores. The Board, citing LKQ at 1296, rejected the argument: “We agree with the Examiner that ‘the “field of invention” of the present claim is not understood to be machine learning, but the ornamentation of the surface of a display screen’ through a graphical user interface. Indeed, the Appellant does not dispute that EM ‘0002, Shuttleworth, and Coffman are display screens having a graphical user interface, each of which depict a design having a line and a lightbulb. Hence, the applied prior art are within the same field of endeavor.” The Board added that even if the confidence score context were considered, “a designer skilled in the art designing a lightbulb icon to illustrate a ‘confidence score’ would be expected to draw from any similar-appearing lightbulb symbols found in the prior art, regardless of what they might be intended to signify in their original contexts,” and quoted the Hupp formulation. The Board then reversed the rejection because the examiner had supplied no record-supported reason to combine. 102 F.4th at 1299.
That decision is the only appellate word in the corpus on how the field is defined for a design, and it defines the field for a GUI by the display screen, not by what the screen shows. The other two Board reversals in the corpus, Ex parte Thornburg, Appeal No. 2023-002969 (29/570,477), and Ex parte Jafarzadeh, Appeal No. 2025-002153 (29/671,792), did not reach analogous art. In Jafarzadeh the applicant had argued that the court gave no guidance on the second prong and argued the point under the Office guidance. The Board reversed on the reason to combine.
7. Why the Argument Rarely Decides the Case
The record suggests four reasons.
First, the test has no content beyond the article, and the article has no fixed level of generality. The court tied the field to the article of manufacture and quoted Hupp for “articles sufficiently similar that a person of ordinary skill would look to such articles for their designs.” Nothing in the opinion says whether a mailing tube is the article or packaging is the article. An examiner who says packaging is not wrong under any stated rule. An applicant who says mailing tube is not wrong either. With no rule for the level of generality, the examiner’s choice controls in prosecution, and examiners choose the category. The 30 files in which the examiner widened the field are the result.
Second, the function-based definition has a home in the MPEP. MPEP 904.01(c) says that what arts are analogous “depends upon the necessary essential function or utility of the subject matter covered by the claims, and not upon what it is called by the applicant.” That sentence was written for utility claims and for search. Examiners are quoting it to define the field for design claims (29/880,660, 29/854,126). A design claim has no function in the utility sense, and the court did not adopt this test. But no authority says examiners cannot use it, and it produces fields like “applying torque” and “devices which inject material into the body” that swallow the article-level argument.
Third, for GUIs the article is the screen, and the Board has said so. The applicant’s software context is treated as intended use, in the same way that the examiner in 29/671,792 treated “delivery” as intended use for a robot. A title amendment cannot change what the screen is. Every software-context argument in the corpus lost on that point. The one GUI win turned on the reference not being an icon at all.
Fourth, the argument is usually made without evidence and answered without evidence. The applicant asserts that a designer of one article would not look to another. The examiner asserts that both are packaging, or containers, or fluid-handling devices. The court said the question is one of fact. Almost no file treats it as one. The few files with evidence, a dictionary definition, a classification, a parent application, a product listing, are the files in which the examiner engaged with the field rather than renaming it.
The reason to combine, by contrast, is a requirement the examiner must satisfy affirmatively with a stated reason, and the Board has reversed seven times in the companion corpus where the reason was missing. The analogous art requirement, as applied, is a requirement the examiner satisfies by naming a category. The two requirements look parallel in the opinion. In practice they are not.
8. What Practitioners Might Consider
Draft the title to name the article at the level of generality the applicant wants the field to have, and make sure the specification and drawings support that title. The field follows the article, and the article follows the title. A title of “Pad” invites a field of absorbent objects. A title of “Vapor Pad for Vaporizer, Diffuser, Humidifier, or the Like” produced a concession. The cautions are the ones that apply to any title amendment: the claim is narrower for enforcement, SurgiSil and Curver cut both ways, and the amendment needs support in the original disclosure. The examiner in 29/771,650 said as much in the interview.
Make the argument where the reference is a different article and the uses are unrelated. The ten wins are packaging against a rail stiffener, filtration media against a seed pod, a hammer tip against an erasing wheel, origami against a GUI, a menstrual pad against a vapor pad after the title change. The argument did not win where the articles share a setting or a user, such as two devices in a hospital, two garments on a leg, two containers, two printers. In those files the examiner named the shared category and moved on.
Do not make the argument alone. In 30 files the applicant lost on analogous art and won on the visual comparison in the same response. The visual argument is the one examiners accept. The analogous art argument is the one they answer. Put both in every response, and put the visual argument first.
Put evidence of the field in the record. The court called this a fact question. An examiner who has to answer a dictionary definition, a classification listing, a declaration from a designer in the field, or a parent application describing the field has to engage with the field. An examiner who has to answer an assertion can answer with an assertion. The files with evidence are the files in which the field was discussed on its terms.
For GUIs, do not expect the software context to define the field. The Board and Art Unit 2923 have both said the article is the screen. If a reference is not a GUI or an icon at all, say so, as in 29/787,268. If it is, argue the visual differences.
Where the examiner has made no analogous art finding, ask for one. In 35 of the 84 files the examiner made none, and in 8 the examiner recited only the form sentence. The Office’s guidance calls for a finding on each reference. A response that identifies the missing finding, quotes the guidance, and states the field the applicant contends is correct puts the examiner to the choice of naming a field or withdrawing. Naming a field creates a record the applicant can answer with evidence.
Where the examiner uses the second prong, note that the court said it does not seem to apply to designs in the same way. No examiner or Board panel in the corpus has explained how a design specification supplies the “particular problem” the utility test requires. The applicants in 29/669,161 and 29/937,714 argued that it cannot. The point is preserved for appeal even though no examiner has accepted it.
Where the examiner recites Glavas, or relies on Antle and Nalbandian for the field, say so. The “so related” test was overruled. Knowledge of all related art is a statement about the level of skill, not about the field. Neither supplies the finding the guidance calls for.
Conclusion
The analogous art requirement is being argued in design prosecution far more often than it is being decided. Applicants raise it in most files where the issue appears. Examiners answer it by defining the field at a level above the article, by function, or, for GUIs, by the screen. The argument decided the case in 10 of 84 files, each involving articles with unrelated uses or a title amendment that made the uses unrelated. The one Board decision on the question defined the field for a GUI by the display screen. No court or Board panel has said what level of generality the field takes for a physical article, whether the second prong exists for designs, or what evidence establishes the field. Until one does, the examiner’s category will usually stand, and the practitioner’s best tools are the title, the visual comparison, and evidence of the field placed in the record.
Limits of the Record
The corpus is 84 issued design patents. Applications abandoned after the analogous art argument failed are not in it, so the figures for the argument’s success are figures for applications that eventually issued. Classification of examiner findings and outcomes was done from the text of the wrappers and reviewed against the wrappers where the first reading was unclear. Quotations are from the wrappers. Page citations to LKQ are to 102 F.4th 1280. Application numbers are given so that any file can be checked.
