“IP renewals” is not one deadline — it is two separate, differently-shaped obligations that a university technology transfer office (TTO) has to track across every patent and trademark it owns or co-owns: patent maintenance fees, which keep a granted patent in force under U.S. law, and trademark renewal filings, which keep a registered mark on the federal register. Both run on fixed statutory clocks, both escalate in cost the longer an asset stays alive, and both are easy to miss precisely because the deadlines are years apart and rarely land on anyone’s daily calendar. For a TTO managing a portfolio that can run from a few dozen to several hundred active assets, renewal management is less a legal question than an operations problem: knowing what is coming due, deciding whether it is still worth paying for, and having a process that does not depend on one person remembering.
This guide covers the deadline structure and current fee mechanics for both asset types, then focuses on the part that is specific to institutional portfolio management: docketing, the maintain-or-abandon decision, and how renewal obligations interact with active license agreements.
Patent maintenance fees: the recurring cost of keeping a patent alive
A granted U.S. utility patent does not stay in force automatically for its full statutory term. Under 35 U.S.C. § 41(b), the USPTO requires three maintenance fee payments, due at fixed points measured from the grant date (not the filing date):
| Due window | Grace period | What happens if missed |
|---|---|---|
| 3.0–3.5 years after grant | Additional 6 months, with surcharge | Patent expires at the end of the grace period if unpaid |
| 7.0–7.5 years after grant | Additional 6 months, with surcharge | Patent expires at the end of the grace period if unpaid |
| 11.0–11.5 years after grant | Additional 6 months, with surcharge | Patent expires at the end of the grace period if unpaid |
The fees themselves are not flat — they escalate sharply at each tier and are set by USPTO’s own fee-setting authority rather than fixed statutory amounts, and they differ by entity size (large, small, or micro entity, a separate USPTO determination based on the applicant’s size and prior patenting/licensing activity). A late payment made during the six-month grace period carries a flat surcharge on top of the base fee regardless of entity size; a patent that lapses even after the grace period can only be revived on a showing that the delay was unintentional, with an additional petition fee on top of the overdue amount. Design and plant patents are the exception: 35 U.S.C. § 41(b) expressly states that no maintenance fee may be charged to keep either in force, so once granted they require no further USPTO payment for their full term.
CASRAI’s guide to patent term, maintenance fees, and extensions covers the full current fee schedule, the mechanics of patent term adjustment (PTA) and patent term extension (PTE), and how the 20-year statutory term interacts with all three — this page does not re-derive that detail. The point that matters for portfolio management specifically: because the fee escalates roughly four-fold between the first and third tier, the 11.5-year maintenance payment is the single most consequential renewal decision point in a patent’s life. A patent with no licensee, no active field-of-use interest, and no near-term commercial prospect is a legitimate, deliberate candidate for abandonment at that point rather than an automatic multi-thousand-dollar renewal.
Trademark renewals: a different clock, and one that never actually ends
A trademark registration works on an entirely different logic from a patent. A patent has a fixed expiration date no renewal can extend past its statutory term; a federal trademark registration, by contrast, has no fixed expiration date at all — it can be renewed indefinitely, in ten-year cycles, for as long as the mark stays in genuine use in commerce under the Lanham Act (15 U.S.C. § 1051 et seq.). The obligation is not really “renewal” in the patent sense — it is periodic proof that the registration still reflects a mark actually being used, and a registration that lapses because nobody files that proof is cancelled regardless of how long it had already been on the register.
Per USPTO’s own post-registration maintenance guidance, the filing structure for a standard (non-Madrid) U.S. registration is:
- Between the 5th and 6th year after registration: file a Section 8 Declaration of Use (or Excusable Nonuse), with a specimen showing the mark in current use, or the registration is cancelled.
- Between the 9th and 10th year after registration: file a combined Section 8 Declaration of Use and Section 9 Application for Renewal — this is the first true “renewal” filing, and it repeats.
- Every 10 years after that (19th–20th year, 29th–30th year, and so on): file the same combined Section 8/9 declaration and renewal.
Each deadline carries a six-month grace period after it, available for an additional per-class fee. As of USPTO’s 2025 fee schedule, the Section 8 declaration alone runs $325 per class, and the combined Section 8/9 renewal filed at the 10-year mark runs roughly $650 per class (the sum of both components) — filing during the grace period adds a further per-class surcharge on top. Fee amounts are set administratively and do change; confirm the current schedule against USPTO’s own fee page before budgeting a specific renewal cycle rather than relying on a figure that may be more than a year old by the time you read it.
Two consequences matter operationally. First, because there is no final expiration, a trademark renewal deadline missed once does not fix itself the way a patent’s fixed term does — it simply lapses the registration outright if the grace period also passes, and re-establishing rights means starting a new application from scratch, potentially after a competing party has already registered something confusingly similar. Second, because the cycle repeats indefinitely, a mark with no ongoing renewal decision point built into institutional process will just keep renewing by default every ten years unless someone actively decides otherwise — the opposite failure mode from a patent, where the default is abandonment absent a deliberate payment.
Where trademark renewal sits inside a university’s IP structure
It is worth being explicit that university trademark portfolios and TTO patent portfolios are usually managed by different institutional offices, even though both fall under “intellectual property.” Institutional marks — the university’s name, seal, wordmark, mascot, and athletic logos — are typically administered by a dedicated trademark licensing or brand office, often working with an outside collegiate licensing agent to handle enforcement and royalty collection across consumer-goods licensees. The technology transfer office’s own patent portfolio is a separate function, licensing faculty inventions rather than institutional identity marks. The two intersect mainly when a university-affiliated startup or licensee needs separate authorization to reference the university’s name or marks in its own branding — a distinct grant, negotiated alongside but separately from the underlying patent license. See CASRAI’s trademark dictionary entry for the fuller distinction between trademark, patent, copyright, and trade secret protection.
Where a TTO does directly hold trademark obligations is narrower: product or technology marks tied to a specific licensed invention (a branded assay name, a licensed software product name) can sit inside the same disclosure-to-license file a patent does, and in that case the Section 8/9 renewal deadlines become exactly the kind of thing a portfolio docketing process needs to track alongside patent maintenance dates, not a separate system nobody owns.
Portfolio-level renewal management: docketing, decisions, and budget
A TTO with even a modest active portfolio is tracking maintenance and renewal deadlines that are years apart, staggered across dozens or hundreds of individually-numbered assets, each with its own grant date, entity-size classification, and grace-period cutoff. The practical failure mode is not usually ignorance of the rule — it is a deadline landing during a personnel transition, a spreadsheet cell that didn’t get updated, or a decision that quietly defaults to “renew” because nobody was asked. A few structural practices reduce that risk:
- Centralized docketing, not memory or a shared calendar alone. Most institutional TTOs and the outside patent counsel they work with use dedicated IP docketing software, or delegate renewal tracking and payment to a commercial patent/trademark annuity (renewal) payment service, precisely because missing a maintenance deadline is an unrecoverable, high-consequence error that a generic calendar reminder is not a reliable enough control for. Whichever system is used, it needs to surface upcoming deadlines with enough lead time — typically 6-12 months — for a real maintain-or-abandon decision to happen, not just a payment reminder the week of the grace-period cutoff.
- A documented decision point at each maintenance tier, not an automatic renewal. Build an explicit review step into the docketing workflow at each patent maintenance fee window and each trademark renewal window, asking the same short set of questions every time: is there an active license or a live negotiation, is the technology still within a licensee’s field of use, has commercialization interest materialized since the last review, and does the remaining term or renewal cost still justify the fee at the entity’s current tier. This is the same discipline AUTM (the Association of University Technology Managers) frames as active portfolio management rather than passive maintenance — a portfolio’s value comes from deliberately pruning assets with no commercial path, not from carrying every filed patent to its natural expiration.
- Budget for the escalation curve, not a flat per-asset cost. Because both patent maintenance fees and, to a lesser extent, trademark renewal costs increase at later tiers, and because a portfolio’s maintenance obligations compound as more inventions are filed each year, annual IP-maintenance budgeting should model the escalating curve across the full live portfolio rather than a single average per-asset figure — a portfolio that looks affordable in year one can carry a materially larger maintenance bill five years later purely from the fee schedule’s own structure, independent of portfolio growth.
- Tie the renewal decision to the license agreement, not just the patent office deadline. Where a patent or product trademark is subject to an active license, most agreements place diligence, minimum royalty, or reporting obligations on the licensee that are a more reliable signal of continued commercial value than the TTO’s own periodic review — a licensee that is current on diligence milestones and royalty payments is strong evidence a maintenance fee is worth paying; a licensee that has gone quiet is a prompt to have that conversation before, not after, the next fee comes due.
For the underlying licensing mechanics that make license status a meaningful input to a renewal decision, see CASRAI’s guide to patent licensing. For the cost side of building a portfolio in the first place, see cost of filing a patent, and for the obligations that specifically attach to federally funded inventions — which can affect whether abandoning a patent is even a decision the institution is free to make unilaterally — see the Bayh-Dole Act’s disclosure and commercialization-diligence requirements.
Frequently asked questions
Do trademark renewals ever end, the way a patent’s term does?
No. A patent has a fixed statutory term (20 years from filing for a utility patent) that no renewal can extend past its natural expiration. A trademark registration has no such ceiling — it can be renewed every ten years indefinitely, for as long as the mark stays in genuine use in commerce and the owner keeps filing the required declarations.
What happens if a TTO misses a patent maintenance fee deadline entirely, including the grace period?
The patent expires. It can potentially be revived only on a showing that the delay was unintentional (or, for older lapses, unavoidable), which requires an additional petition and fee on top of the overdue maintenance payment and its surcharge — revival is not guaranteed, and is a materially more expensive and uncertain path than simply paying on time.
Is trademark renewal handled by the same office that manages the patent portfolio?
Often not, for institutional marks — university name, seal, and athletic branding are typically handled by a separate trademark licensing or brand office, sometimes with an outside collegiate licensing agent, rather than the TTO. A TTO’s own renewal responsibility is usually narrower: product or technology marks tied to a specific licensed invention, which do belong in the same docketing process as that invention’s patent.
Does an issued patent’s maintenance fee amount depend on the institution’s entity size?
Yes. USPTO sets separate large-entity, small-entity, and micro-entity maintenance fee rates, with small and micro-entity rates substantially lower than the large-entity rate. Entity size is a distinct USPTO determination with its own qualifying criteria — verify current status before relying on a reduced rate for a specific patent rather than assuming eligibility.
