Category Archives: Statistics

What Issues Mattered in Trademark Prosecution in 2022?

Leave a reply

As a trademark practitioner, you know that not all Office Actions are created equal. If your Office Action just raises an issue with a specimen or a disclaimer request, there may be some back-and-forth with your examining attorney, but the issue will almost certainly get resolved sooner or later. You can get a feel for how likely an issue is to be resolved prior to publication with our Office Action analysis tool and the Examiner analytics it provides.

If, however, you run into a functionality refusal or a likelihood of confusion refusal, you’re likely looking at at least one round of substantive back-and-forth. Applications running into these sorts of refusals move forward to publication at a much lower rate, as we’ve detailed in some prior blog posts.

What sort of refusals, then, did the USPTO tend to issue in 2022? The big ones were:

Substantive Refusals
Likelihood of Confusion23%
Descriptiveness18%
Non-Substantive Refusals
Specimen30%
Classification25%
Disclaimer22%
Mark description14%
Domicile12%
All numbers are a percentage of times an issue was raised in any Office Action in 2022. Many Office Actions contain multiple issues, so these percentages will not add up to 100%.

Foreign registration-related suspensions and applicant entity clarification requests were present in around 4% of Office Actions each. No other issue exceeded 2% frequency.

A few interesting points arise from this data, particularly the large numbers of non-substantive refusals.

The Office Should Change the Process for Assigning Mark Descriptions. Mark description issues are far more common than I expected. It might be more efficient for the Office to simply assign mark descriptions and allow applicants to object, similar to the process for design codes. That process is highly effective, and applicants rarely push back on the Office’s preferred design coding (or, indeed, the description of the mark phrasing). The change of process would permit some otherwise perfectly acceptable applications from either going abandoned for lack of a response, and speed examination time for almost all of the rest.

Classification Issues Aren’t Going Away. The heavy incentives towards TEAS Plus filing are not enough to eliminate identification woes. While the Office’s ever-widening price gap between TEAS and TEAS Plus is an attempt to mitigate this issue, it can’t solve the whole problem — indeed, it will never solve the issue for any Madrid-based applications. Only further efforts to create a worldwide equivalent of the ID Manual would not only help speed up prosecution in the US, but could address the maddening inconsistencies between what constitutes an acceptable definition goods in various countries. If WIPO could incentivize (e.g. through filing fees) use of such “universally accepted” language, it would be a big improvement for the international trademark system.

Disclaimer Practice is a Drag. Disclaimer requests are really, really common, popping up in over a fifth of Office Actions. Disclaimer practice in the US is riven with inconsistencies; the same term showing up in a short mark will get a disclaimer request; if the mark has a pun in it or if the term is used in a tagline that’s barely any longer, it’ll avoid a disclaimer request. It’s all very silly.

There has to be a better way. A simple, hard-and-fast rule would remove much of this back-and-forth. For example, the Office could adopt a rule that a term that is used in active registrations more than X times (3? 5?) by different owners in a single class is presumptively not likely to cause confusion with another mark, unless there is other evidence of mark similarity like other terms in the mark, visual presentation, etc. That would have two big benefits. First, it would dramatically simplify the confusion analysis. (As a side effect, it would stop the Office from issuing the annoying refusals where 6 companies own registrations for TERMX + SECOND TERM, someone slips through a registration for TERMX alone for narrow goods, and then every subsequent applicant for TERMX + SECOND TERM style marks gets a 2(d) refusal that they have to fight about.) Second, it would obviate the need for disclaimer practice. Just count!

As with mark description issues, some otherwise perfectly fine applications are either going abandoned over disclaimer requests, or have to deal with extra attorney fees to reply (and increased total prosecution time). Neither are a good outcome, and the Office should consider a wholesale revamp of the disclaimer practice to fix it.

Conclusion. Hopefully, we’ve not only given you some insight on the types of issues applicants are running into most often, but also provided some interesting ideas about how the Office could revamp its existing processes to better avoid unnecessary prosecution problems.

Slowing Trademark Numbers and COVID-19: a Serious Slowdown Coming?

1 Reply

Thanks in large part to the COVID-19 virus, the US economy is looking a lot shakier than it was just months ago. Economic slowdowns can cut down on new trademark filings from both existing businesses (which may be afraid to launch products into a slow market) and from new businesses (who aren’t launching at the same rate they would in a brisker economy).

We took a look at the two countries that have been hardest hit by the virus and have substantial numbers of US applicants – China and Italy (sorry, Iran, you don’t qualify) – to see what might be in the future for the US. The numbers aren’t encouraging.

The first full 5-day week of March in 2020 saw new filings drop by 9% over either 2019 or 2018. Applications from the two main countries with large-scale COVID-19 outbreaks, China and India, dropped especially hard. China’s filing numbers in 2020 were only 41% of their 2019 numbers; Italian filings were only 26% of their 2019 levels.

March (first week) 2020: 8,376 (387 from China; 13 from Italy)
March (first week) 2019: 9,128 (924 from China, 50 from Italy)
March (first week) 2018: 9,172 (677 from China; 70 from Italy)

February 2020 was also off, with new filing dropping by 6% overall from 2019, and about the same rate from 2018. Chinese filings were about half their Feb. 2019 levels and just over 1/3 of their 2018 levels; Italian-based applicants were at only 25% of their 2019 levels.

Feb. 2020: 33,068 (1,215 from China; 54 from Italy)
Feb. 2019: 36,092 (2,194 from China; 202 from Italy)
Feb. 2018: 36,593 (3,066 from China; 284 from Italy)

The foreign-filing drops cannot entirely be blamed on the coronavirus outbreak. The changes for unrepresented foreign applicants were intended to, and, as our prior research suggests, has reduced application numbers in Jan. 2020 and overall.

However, thes precipitous drops in February and the first week of March do, I think, reflect a serious and substantial slowdown in economic activity. If the US sees a similar societal impact from the virus – and it impacts US filings the way it seems to have impacted Chinese and Italian applications in the US – the US trademark filing scene could be hurting, and soon.

Foreign Equivalents Refusals: Examination Outcomes

Leave a reply

This blog post takes a statistical look at the foreign equivalents doctrine, and how it influences examination outcomes in 2(d) refusals. In particular, I wanted to compare refusals where the Examining Attorney based the refusal on the foreign equivalents doctrine against Office Actions that did not raise the foreign equivalents doctrine but did emphasize that one or more terms in the mark are synonyms.

I looked at both 2019 and 2018 data, since so many of the 2019 applications have not made their way to a final decision yet.

Foreign equivalents refusals tend to result in a higher rate of refusals than “synonym” refusals. Roughly 10% more applications that had a foreign equivalents issue raised were refused in 2018 vs. “synonym” refusals. On a much smaller data set, the outcomes are much more closely equivalent in 2019, but outcomes are still worse for applications that receive a foreign equivalent refusal.

It’s also interesting that “synonym” refusals are just less common than “foreign equivalent” refusals – foreign equivalent refusals are about 5x more common than “synonym” refusals. Why? From dipping into specific examples, it appears that a decent percentage of the “synonym” refusals are not focused on the shared term in the mark – they’re the Examiner noting that an additional term, often weaker or descriptive, is essentially synonymous and so not sufficient to differentiate the two marks.

To see if I got different results, I restricted the search to only instances where there was a 1-word mark on each side of the refusal. (This is just space-delimited, so I’m sure it pulls in some compound marks, but it’s close enough.) This dropped the data set dramatically, especially for “synonym” refusals.

I expected the abandonment rates to be higher here, since these are one-word to one-word comparison. It was higher, a bit, for foreign equivalents, but perhaps not as much as I expected – maybe the inclusion of articles in some foreign languages tilted the multi-word comparison rates a bit higher. On the “synonym” refusals, these were if anything overcome at a higher rate than multi-word terms, supporting the hypothesis above that synonym-related refusals are most likely to matter when it is a tacked-on, secondary word (like MVD ANALYSIS vs MVD ASSAY) that is the synonym than a synonym-based comparison of the strongest term in a mark (like BOBCAT vs. LYNX).

Don’t forget to take advantage of TM TKO’s full set of indexed, text-searchable USPTO trademark prosecution histories to find comparable refusals – and successful responses! – for your Office Actions, whether you are facing issues related to foreign equivalents, synonyms, or any other prosecution hurdles.

Digitally Altered Specimen Refusals – By the Numbers

1 Reply

The US Patent & Trademark Office just came out with a new Examination Guide covering digitally altered specimens. Examination Guide 3-19 Examination of Specimens for Use in Commerce: Digitally Created or Altered and Mockup Specimens (USPTO, July 2019).

TM TKO has full, searchable prosecution histories as a part of the research platform, so I searched for any Office Actions related to this type of refusal. Although the Examination Guide is new, the refusal is not – it really started appearing in 2015, when the refusal was made about 1,000 times, and the refusal rates has continued to spike in each year since; 2019 is on pace for over 20,000 refusals. Since 2016, more than half of these refusals have been issued to Chinese applicants. However, a surprisingly small number of refusals have, to date, “held up” and resulted in the abandonment of the application – about 70% of the applications that have been finally disposed of have moved forward to registration, a far higher figure that I would have expected.

This raises interesting questions about what sort of conduct should result in an application being invalid. The “fraud” line of cases, like Medinol Ltd. v. Neuro Vasx Inc., 67 USPQ2d 1205 (TTAB 2003), was killed extremely dead via Bose Corp. v. Hexawave, Inc., 88 USPQ2d 1332 (TTAB 2007). The Board has, as a result, been very unwilling to expand revisit the idea of “fraud,” even where the applicant provided a demonstrably false specimen – deceptive intent is required, and so long as the applicant doesn’t say they had deceptive intent, they’re pretty much going to get off the hook. Tommie Copper IP, Inc. v. Gcool-Tech Usa LLC, Opposition No. 91223768 (May 9, 2018) [not precedential]. That a use-based application might be void ab initio due to non-use as of the use date claim might still be tenable, e.g., Parametric Technology Corporation v. PLMIC, LLC, Opposition No. 91174641 and PLMIC, LLC v. Parametric Technology Corporation, Opposition No. 91177168 (February 12, 2010) [not precedential], but the case law seems pretty clear that an applicant that knowingly submits a fake specimen and then cures with a real one prior to registration is basically going to be fine.

On the one hand, if the applicant ultimately can show that it is making real, in-commerce use of the mark, where’s the harm? The applicant got it right eventually and, for the purposes of having a registry that reflects reality, well, it does. On the other hand, there’s some real skepticism in the practice community that the specimens that are eventually accepted are actually real – that, instead, they’re just better fakes. Why would you fake something if you have a real product or service, after all?

Some practitioners take the position that the verified statement of the applicant should be taken at face value. I do not agree. If a simple Google Images or Tineye search can find the original photo that was doctored to match, or it’s otherwise equally blatant (a McDonald’s restaurant photo with “Mary’s Burgers” badly edited in place of “McDonald’s”), I see no reason why the USPTO should have to turn off its brain when examining specimens and ignore obvious fakes.

Represented vs. Unrepresented Applicants

How does being represented by an attorney impact the rate at which these refusals are issued, and the how does it impact the rates at which the refusals are overcome? The data for these queries treats pro se applicants who later add an attorney as “represented.”

Chinese Applicants

It is widely understood in the trademark practitioner community that a significant driver of these refusals are applications with Chinese applicants. That is definitely true – 45% or more of these refusals each year since 2016 are for Chinese applicants. Chinese applicants as a group overcome these refusals at a rate that is better than the norm. Chinese applicants have better success rates than unrepresented parties, although less than all represented parties do.

The issue of digitally altered specimens will undoubtedly continue to grow over time, as the technology becomes ever more omnipresent and the USPTO continues to work to implement its new Examination Guide. As always, TM TKO will be your best resource to research how the Office and other lawyers are addressing these kinds of refusals.

TM TKO launches new Office Action Analytics and Search Upgrades

1 Reply

TM TKO’s new Office Action analytics let you prepare smarter, better Office Action responses in less time. Our tools do the complex research for you, instantaneously. See a full example of the kind of Office Action analytics that TM TKO provides, or try it now from your Tools menu.

Issue-driven analysis: instantly provides you successful responses for similar marks for similar products or services to help you build on the successes of others. Pull any TSDR document in formatted PDF or plain-text format.

Examiner details: compare your Examining Attorney’s allowance rates on this issue with others in that Law Office and at the USPTO overall, and see recent successful responses for your issue and your Examiner.

Citation histories: do deep dives into citation histories at a click. It’s easy to understand complex webs of co-existence, assess your own chances of success, and see how the cited prior filings have interacted with other applications.

ThorCheck

Office Action analysis integrates with ThorCheck, TM TKO’s ground-breaking comparative research tool based on a precedential TTAB opinion, to find examples of co-existence of the same mark for two sets of goods with different owners. This evidence helps you push back against likelihood of confusion claims. You can also use ThorCheck to provide confusion evidence as the senior party in a Letter of Protest or opposition.

Search Upgrades

Office Action and prosecution data is now integrated across our platform.

Results in knockout search, manual search, and watch results all feature easily accessible citation data, full TSDR file histories with Office Actions and Responses tagged by issue, and quick links to prosecution analysis. Just click on the magnifying glass icon for a wealth of research options.

In-Depth Prosecution Stats – 2(d) Refusals in Class 1

Leave a reply

At TM TKO, we are really interested in prosecution trends. Today, we are going to zoom in on one type of refusal – 2(d) likelihood of confusion refusals – in one class, International Class 1. International Class 1 primarily consists of industrial chemicals and similar products.

In particular, we wanted to test out class relationships – how strong was the likelihood of getting a refusal if the prior mark was in-class, was in a coordinated class (but not in-class), or was in an “unrelated” class (not in-class and not in a coordinated class), and how likely were the refusals to “stick” and cause the application to go abandoned?

While fewer than 12% of applications filed at any time received 2(d) citations, the 2(d) citation rate jumps to about 30% from 2014 to the present. It’s perhaps unsurprising that an increasing 2(d) rate results from an increasingly crowded registry.

As for the content of the cited marks, 2(d) refusals were not quite twice as common in the same class (i.e. the prior registration had International Class 1) than in all the classes noted by the Office as “coordinated” classes combined (any of International Classes 5, 17, 35, 42, 44, A, or B). “Out-of-class” citations – all other classes except Class 1 or any coordinated classes – were surprisingly common, with slightly more “out-of-class” citations than in-class citations, and about twice the number of out-of-class citations as coordinated class citations. There are many more marks not in one of Class 1’s coordinated classes than not, of course, which may account for the difference – currently, there are some 67,000 active filings in Class 1, just short of 950,000 in Class 1 or a coordinated class of Class 1, and more than three million neither in Class 1 nor a coordinated class of Class 1.

Cl1_2d_core

The importance of class and coordinated class are more apparent in looking at how frequency applicants overcome the initial 2(d) citation or citations. Applicants that receive a 2(d) citation in Class 1 or a coordinated class overcome it roughly 60% of the time; applicants that receive a 2(d) citation in an “unrelated” class are about 10% more likely to be successful. The same trends hold since 2014 as over all citations, suggesting that – at least for this class – citations in unrelated classes were noticeably less likely to hold up after further argument by the applicant.

Cl1_2d_coor

Within the coordinated classes, refusals in Class 1 tended to be the most difficult to overcome; citations to Class 42 (traditionally a grab-bag of services, though now more narrowly defined) and Class 200 were the most likely to be overcome by applicants. Refusals in Classes A and B were less common, but comparatively difficult to overcome, perhaps because they can include especially broad descriptions of goods.

We will be looking at other classes and other prosecution in more depth over the coming weeks and months. If you have specific topics of interest, please let us know at inquiries@tmtko.com.