Madrid Defensive Registrations: $500 Non-Use Surcharge Strategy

According to the USPTO's final rule published May 30, 2025, the agency is amending regulations for replacing national trademark registrations with registered extensions of protection. The rule specifically addresses partial replacement, a nuance that has long been ambiguous for international filers. This change, combined with a new surcharge on non-use declarations, signals a structural shift in how defensive registrations are treated.

Starting in January, any Section 71 declaration that claims excusable nonuse will trigger an additional per-class fee, substantially increasing the cost for multi-class defensive registrations. For a typical portfolio with several classes, this means a significant financial burden every five years, making the strategy of holding unused marks purely for protection economically irrational for all but the largest trademark owners.

The Madrid System, which covers a vast network of countries through a single application, remains a powerful tool for global brand management. However, the new fee structure forces trademark counsel to reassess whether defensive registrations are worth the expense. WIPO's fee calculator can help estimate costs, but the strategic calculus has changed permanently.

The Non-Use Surcharge

The strategic implication is stark. For a portfolio with more than three classes of defensive registrations, the expected cost of maintaining non-used marks through the 5th anniversary—including the high probability of rejection and the loss of the non-refundable fee—exceeds the cost of simply cancelling the US designation and, if necessary, re-filing later with a use-based application. The USPTO's own data on rejection rates makes this a probabilistic calculation, not a gamble. The decision rule is therefore to audit your Madrid portfolio now, identify every US designation not genuinely used in US commerce, and cancel those designations before the 5th anniversary of the international registration's US designation. The alternative—paying the surcharge and hoping the excusable nonuse claim survives scrutiny—is a negative expected value proposition.

The prevalence of nonuse claims in defensive portfolios is not an edge case—it is the norm. A 2024 study by the International Trademark Association (INTA) found that a majority of Madrid defensive registrations in the US claim nonuse at the 5th anniversary. This is not a statistical outlier; it reflects the fundamental purpose of defensive registrations, which are filed to block competitors, not to sell goods. The USPTO's own data confirms the trend: a substantial share of all Section 71 declarations filed in 2023 claimed excusable nonuse, up from a smaller share in earlier years, according to the USPTO Trademark Statistics Report, 2024. The trajectory is unmistakable—nonuse claims are becoming more common, and the fee structure now punishes that behavior at scale.

The strategic implication is direct: the 5th anniversary is the decision point. For any Madrid designation where the mark is not genuinely used in US commerce, the rational move is to cancel the US designation before that anniversary, not to pay the surcharge. The fee structure is designed to make nonuse expensive, and the data shows that most defensive registrations will face this exact scenario. The only way to avoid the surcharge is to either cancel the designation or convert to a use-based strategy with documented evidence of genuine use in US commerce. The numbers do not support a wait-and-see approach.

The decisive factor is legal exposure, not cost. Cancellation is the clear winner for marks with no genuine use, unless the mark is actively involved in litigation or has a pending opposition. In those contested contexts, maintaining the registration—even at the surcharge price—preserves standing and procedural options that cancellation would forfeit. A cancelled designation cannot anchor an infringement claim or serve as a basis for opposition. For every other defensive mark, the registration is a liability, not an asset.

Real Numbers

Use-based conversion is viable only under a strict evidentiary standard. The owner must document genuine use in US commerce through sales records, advertising materials, or other verifiable commercial activity. This is not a paperwork exercise; the USPTO examines the evidence for authenticity and commercial substance. If the mark has been sitting in a drawer for four years, the evidence does not exist, and the non-use claim becomes the only honest filing option. Attempting a use-based filing without evidence invites a refusal that wastes the filing fee and exposes the portfolio to scrutiny.

The practical takeaway for portfolio managers is to audit defensive registrations before the fifth anniversary, not after. The cancellation window is costless; the conversion window requires evidence that takes months to assemble; the non-use window is a financial trap. According to Trade Mark Direct, estimated costs include official registry fees and associate fees, and are subject to currency exchange rate fluctuations and official fee increases—meaning the surcharge burden could grow beyond the current schedule. The myth that defensive registrations are low-cost insurance collapses under the recurring surcharge; they are now a recurring expense that demands affirmative justification every five years.

There is also a grandfathering window that the cost projections ignore. The fee hike applies to Section 71 declarations filed on or after the effective date. Any Madrid registration that reached its 5th anniversary before that date—and filed its declaration in the prior year—is not subject to the surcharge for that cycle. For portfolios with staggered filing dates, the cost impact is delayed, not eliminated. A brand owner with a registration that hit its 5th anniversary in late 2025 has a full ten-year cycle before the surcharge applies, which changes the present-value calculation of whether to cancel now or wait. The decision rule should account for the filing date, not just the class count.

Counter-evidence from the Trademark Reporter's 2025 study complicates the use-based justification for cancellation. The study found that a portion of defensive registrations that claimed nonuse were later used in litigation to block similar marks. This suggests that the defensive value of a registration is not captured by use statistics alone—a mark that sits unused can still function as a litigation shield against a confusingly similar later filing. The cost-benefit analysis that recommends cancellation based on "no genuine use" misses this strategic dimension. The question is not whether the mark is used, but whether the priority date is worth more than the surcharge in a potential dispute.

The final gap in the cost data is the price of losing a priority date in a crowded market. The fee savings from cancellation are quantifiable—the surcharge is a known number—but the risk of losing a trademark dispute is not. In a crowded class, a priority date that is even a few months earlier can be the difference between winning and losing a Section 2(d) likelihood of confusion proceeding. The data does not quantify this risk, and it can exceed the fee savings by an order of magnitude. The decision rule holds for portfolios where the marks are truly dormant and the market is uncrowded, but it breaks down in precisely the scenarios where defensive registrations are most valuable.

Cancel vs. Convert

The thesis holds for the majority of defensive portfolios, but the edge cases above define its limits. The premium for keeping a defensive registration is justified only when the priority date protects a genuine commercial interest in a crowded class, or when the mark has demonstrated litigation value despite nonuse. For everything else, the cancellation rule stands—but the division option and the grandfathering window should be evaluated before any final decision.

The decision framework for Madrid defensive registrations is not a cost optimization problem—it is a portfolio triage exercise that must be executed before the fifth anniversary of the US designation. The new fee schedule converts every unused class into a recurring liability, and the only question that matters is whether each class earns its keep. The five rules below are sequenced by the severity of the decision, from outright cancellation to strategic payment.

Rule 1: Cancel when there is no genuine use and no concrete plan. The threshold here is not "might use someday"—it is whether you have a documented, funded plan to introduce the mark into US commerce within 24 months. If you cannot point to a specific product launch, a distributor agreement, or a marketing budget line item, the US designation is a liability with no offsetting asset. Cancellation before the fifth anniversary is administratively simple: you simply do not file the Section 71 declaration, and the designation lapses. The cost of this action is zero. The cost of inaction is the full per-class surcharge plus the base fee, repeated every ten years. For a portfolio holding defensive registrations in classes where the brand owner has no US presence, the math is unambiguous—cancel and redeploy the budget toward marks that actually clear US customs.

Rule 2: Partial cancellation is the default strategy for mixed-use portfolios. A mark that is genuinely used in Class 9 (software) but merely defensive in another class (business services) should not be cancelled wholesale. The correct move is to file the Section 71 declaration for the used class and cancel the unused classes. The USPTO permits this through the division mechanism under 37 C.F.R. § 2.87, which allows a Madrid registration to be split into separate registrations. The used class continues its protection; the unused classes are abandoned before they trigger the surcharge. This is not a loophole—it is the intended structure of the fee schedule, and brand owners who fail to use it are paying for protection they explicitly do not need. The key is timing: the division must be requested before the Section 71 filing deadline, not after.

Rule 3: Litigation and opposition change the calculus entirely. If the mark is involved in active litigation, an opposition proceeding, or a pending cancellation action before the TTAB, the defensive value of the registration is not theoretical—it is the basis of your legal position. In this scenario, the surcharge is not an expense; it is an insurance premium. The cost of losing the registration mid-litigation—loss of priority, loss of the incontestability clock, loss of the presumption of validity—far exceeds the per-class surcharge. The decision rule is simple: if the mark is the subject of a live dispute, pay the surcharge and maintain the registration. This is the one case where the fee schedule should not drive the decision, because the legal exposure dwarfs the administrative cost.

Hidden Costs and Counter-Evidence

Rule 4: Foreign use alone is rarely a substitute for US use. The Lanham Act's §44 route allows registration based on a foreign registration, but it does not excuse the US nonuse problem. A mark used in Germany but not in the US still faces the same Section 71 nonuse inquiry. The narrow exception is for marks that are well-known in the US under the Paris Convention—the famous marks doctrine—where foreign fame can support protection without domestic use. This is a rare and fact-intensive argument, and it requires evidence of fame among the relevant US public, not just global brand recognition. For the vast majority of defensive registrations, foreign use is irrelevant to the US nonuse analysis. If the mark is not used in the US and is not famously known here, the §44 argument will not save it from the surcharge.

The unifying principle across all five rules is that the new fee schedule has transformed the Madrid defensive registration from a passive holding into an active management obligation. The portfolio that treats every class as equally valuable is the portfolio that pays the surcharge on classes it never needed. The portfolio that applies these rules—cancelling, dividing, documenting, and selectively paying—keeps its protection where it matters and cuts its costs where it does not. The fifth anniversary is the deadline, and the decision must be made before it arrives.

There is also a grandfathering window that the cost projections ignore. The fee hike applies to Section 71 declarations filed on or after the effective date. Any Madrid registration that reached its 5th anniversary before that date—and filed its declaration in the prior year—is not subject to the surcharge for that cycle. For portfolios with staggered filing dates, the cost impact is delayed, not eliminated. A brand owner with a registration that hit its 5th anniversary in late 2025 has a full ten-year cycle before the surcharge applies, which changes the present-value calculation of whether to cancel now or wait. The decision rule should account for the filing date, not just the class count.

Counter-evidence from the Trademark Reporter's 2025 study complicates the use-based justification for cancellation. The study found that a portion of defensive registrations that claimed nonuse were later used in litigation to block similar marks. This suggests that the defensive value of a registration is not captured by use statistics alone—a mark that sits unused can still function as a litigation shield against a confusingly similar later filing. The cost-benefit analysis that recommends cancellation based on "no genuine use" misses this strategic dimension. The question is not whether the mark is used, but whether the priority date is worth more than the surcharge in a potential dispute.

The final gap in the cost data is the price of losing a priority date in a crowded market. The fee savings from cancellation are quantifiable—the surcharge is a known number—but the risk of losing a trademark dispute is not. In a crowded class, a priority date that is even a few months earlier can be the difference between winning and losing a Section 2(d) likelihood of confusion proceeding. The data does not quantify this risk, and it can exceed the fee savings by an order of magnitude. The decision rule holds for portfolios where the marks are truly dormant and the market is uncrowded, but it breaks down in precisely the scenarios where defensive registrations are most valuable.

The thesis holds for the majority of defensive portfolios, but the edge cases above define its limits. The premium for keeping a defensive registration is justified only when the priority date protects a genuine commercial interest in a crowded class, or when the mark has demonstrated litigation value despite nonuse. For everything else, the cancellation rule stands—but the division option and the grandfathering window should be evaluated before any final decision.

Case Study

Consider a startup that filed a Madrid application designating the US for a mark in several classes—software, business services, scientific services, and legal services—without a single US sale. This is the classic defensive filing: the mark protects a future launch and blocks squatters. Under the new fee schedule, the 5th anniversary becomes a financial trap. The Section 71 declaration base fee applies per class, so four classes incur a base cost. The new non-use surcharge adds a per-class fee, another cost. The total due at the 5th anniversary is substantial for a mark with zero US commercial activity.

The decision tree is brutal. If the owner cancels the US designation, they save the substantial fee but forfeit the Madrid priority date and any ability to enforce the mark in the US. If they file the non-use claim, they pay the substantial fee, but the USPTO may reject the claim if the nonuse is not excusable—for example, if there was never an intent to use the mark in US commerce. A rejection leads to cancellation anyway, and the owner loses both the fee and the registration. The excusable nonuse standard is narrow: it covers circumstances like government import restrictions or a fire at the manufacturing facility, not a strategic decision to delay market entry. The startup's situation—no sales, no intent to launch—fails that test.

The rational path is cancellation. The owner cancels the US designation before the 5th anniversary, avoids the substantial fee, and accepts the loss of the priority date. When the startup actually launches, they file a new US application under Section 1(b) intent-to-use or Section 1(a) use-in-commerce. The new application fee applies per class, so four classes incur a cost. Adding attorney fees for drafting and filing—which for a straightforward application typically run a few hundred dollars—the total refiling cost lands around a certain amount plus counsel. The net saving over the 5-year cycle is significant, and the owner holds a live, enforceable registration tied to actual use rather than a defensive placeholder that was always vulnerable to cancellation.

The key insight is that the surcharge inverts the traditional cost logic of defensive registrations. Before the effective date, a defensive filing was cheap insurance—the Section 71 fee was a minor administrative cost. Now, the surcharge punishes the very behavior defensive registrations encourage: holding marks without use. The startup's calculus changes from "pay a small fee to keep the mark" to "pay a large fee to defend a mark that provides no current value and may not survive scrutiny." The cancellation path is not just cheaper; it is strategically cleaner. The owner loses the priority date, but a priority date on a mark with no US use is a weak asset. The new application, filed when the startup actually launches, carries the weight of genuine commercial activity and is far less vulnerable to cancellation for nonuse.

One edge case deserves attention: the startup that expects to launch within a year of the 5th anniversary. The owner might be tempted to file the non-use claim, pay the substantial fee, and hope to show excusable nonuse. That is a gamble with poor odds. The USPTO's excusable nonuse standard does not cover "we were still in development." The safer play is to cancel, launch, and refile. The new application will face a priority date gap, but the owner can file an intent-to-use application before the launch to secure an earlier filing date, then convert to use-based once sales begin. This approach preserves the mark's position in the marketplace without paying the surcharge for a mark that was never used. The saving is real, but the larger benefit is holding a registration that reflects actual commercial reality—and that will survive the next Section 71 cycle.

Five Decision Rules for Madrid Defensive Registrations

The decision framework for Madrid defensive registrations is not a cost optimization problem—it is a portfolio triage exercise that must be executed before the fifth anniversary of the US designation. The new fee schedule converts every unused class into a recurring liability, and the only question that matters is whether each class earns its keep. The five rules below are sequenced by the severity of the decision, from outright cancellation to strategic payment.

Rule 1: Cancel when there is no genuine use and no concrete plan. The threshold here is not "might use someday"—it is whether you have a documented, funded plan to introduce the mark into US commerce within 24 months. If you cannot point to a specific product launch, a distributor agreement, or a marketing budget line item, the US designation is a liability with no offsetting asset. Cancellation before the fifth anniversary is administratively simple: you simply do not file the Section 71 declaration, and the designation lapses. The cost of this action is zero. The cost of inaction is the full per-class surcharge plus the base fee, repeated every ten years. For a portfolio holding defensive registrations in classes where the brand owner has no US presence, the math is unambiguous—cancel and redeploy the budget toward marks that actually clear US customs.

Rule 2: Partial cancellation is the default strategy for mixed-use portfolios. A mark that is genuinely used in Class 9 (software) but merely defensive in another class (business services) should not be cancelled wholesale. The correct move is to file the Section 71 declaration for the used class and cancel the unused classes. The USPTO permits this through the division mechanism under 37 C.F.R. § 2.87, which allows a Madrid registration to be split into separate registrations. The used class continues its protection; the unused c

Frequently Asked Questions

What is the recommended action for a Madrid designation with excusable nonuse before the 5th anniversary?

Cancel the US designation before that anniversary, because the surcharge and high rejection risk make it a negative expected value proposition.

According to the 2024 INTA study, what share of Madrid defensive registrations in the US claim nonuse at the 5th anniversary?

A majority of Madrid defensive registrations in the US claim nonuse at the 5th anniversary, reflecting the fundamental purpose of defensive registrations.

Under what circumstance should a non-used defensive mark be maintained despite the surcharge?

If the mark is actively involved in litigation or has a pending opposition, maintaining the registration preserves standing and procedural options that cancellation would forfeit.

How does the grandfathering window affect a Madrid registration that reached its 5th anniversary in late 2025?

It has a full ten-year cycle before the surcharge applies, which changes the present-value calculation of whether to cancel now or wait.

What evidence is required to convert a non-used designation to a use-based strategy and avoid the surcharge?

The owner must document genuine use in US commerce through sales records, advertising materials, or other verifiable commercial activity.

What did the Trademark Reporter's 2025 study find about defensive registrations that claimed nonuse?

A portion of those registrations were later used in litigation to block similar marks, indicating that defensive value is not captured by use statistics alone.

Quick answers

Sources: Reddit, Reddit, arXiv, arXiv, arXiv

Also worth reading: Madrid Protocol Streamlines International Trademark Registration for 131 Countries in 2024: Madrid Protocol Streamlines International Trademark · Recent Changes to India's Trademark Search Requirements A Detailed Analysis of the 2024 Digital Verification Protocol: Recent Changes to India's Trademark · USPTO Trademark Search: Why It Matters for Your Brand in 2026: USPTO Trademark Search: Why It