Ask why trademark applications fail and most people picture a refusal: an examiner who finds a conflicting mark, or a bigger company that files an opposition. The USPTO's own records tell a quieter story. We followed every U.S. application filed between January 1 and June 30, 2024 — 277,556 of them — to September 30, 2026, and looked at how each one ended.
The short version: about a third were abandoned, and most of those died because a deadline passed. Examiners' refusals and opponents matter, but far less often than silence.
Where 277,556 Applications Ended Up
| Outcome by September 30, 2026 | Applications | Share |
|---|---|---|
| Registered | 155,483 | 56.0% |
| Abandoned | 86,382 | 31.1% |
| Still pending | 34,304 | 12.4% |
| Other (registered then cancelled, unclear status) | 1,387 | 0.5% |
Some of the pending applications will still be abandoned, so the final rate for this cohort will end up higher. Part of the reason it isn't finished: an approved intent-to-use application can stay open for up to three years after its notice of allowance while the owner proves use.
Why They Were Abandoned
Every abandoned application carries a status code that records how it ended. Here is how the 86,382 abandonments break down:
| How the application ended | USPTO status | Applications | Share of abandonments |
|---|---|---|---|
| No response, or a late response, to an Office Action | 602 | 52,778 | 61.1% |
| No Statement of Use filed (intent-to-use) | 606 | 27,553 | 31.9% |
| Withdrawn by the applicant (express abandonment) | 601 | 2,259 | 2.6% |
| Lost an opposition or other inter partes case | 604 | 1,873 | 2.2% |
| Withdrawn by the applicant after publication | 605 | 800 | 0.9% |
| Incomplete response | 600 | 566 | 0.7% |
| Refusal upheld on appeal | 603 | 235 | 0.3% |
| Petition to revive denied | 614 | 227 | 0.3% |
| Petition to revive filed, not yet decided | 612 | 61 | 0.1% |
| Defective Statement of Use | 607 | 30 | 0.0% |
Two lines account for 93.0% of everything. The first, "failure to respond or late response", is what happens when an Office Action's deadline passes without an answer — three months, or six with a paid extension. The second applies only to intent-to-use filings: the USPTO approved the mark, but the owner never filed proof that it was in use. 20.6% of all intent-to-use applications in the cohort ended this way.
The data records how each application ended, not why the owner let it go. Some applicants abandon on purpose — a product is dropped, a rebrand happens, an Office Action shows the mark can't be saved and answering it isn't worth the cost. Others may simply have missed the deadline. The status code can't tell those apart, so we describe these as deadlines that passed, not as mistakes.
An Office Action Changes the Odds
| Applications filed Jan–Jun 2024 | Applications | Abandoned by Sept. 2026 |
|---|---|---|
| Got at least one Office Action | 145,755 | 44.0% |
| Got no Office Action | 131,801 | 16.8% |
| Got a final refusal | 18,187 | 45.8% |
| Got no final refusal | 259,369 | 30.1% |
An Office Action roughly multiplies the chance of abandonment by 2.6. That is partly the substance of the objection — some marks can't be registered as filed — and partly the deadline it creates. Either way, the practical lesson is the same: the day an Office Action arrives, put its response date somewhere you can't miss it. Our deadline calculator works it out from the issue date and gives you a calendar file.
Filing Basis Matters More Than Anything Else
| Filing basis (as filed) | Applications | Registered | Abandoned |
|---|---|---|---|
| Already in use — Section 1(a) | 139,559 | 75.2% | 20.7% |
| Intent to use — Section 1(b) | 131,777 | 35.2% | 42.7% |
| Foreign application or registration — Section 44 | 5,233 | 76.2% | 14.1% |
Filing on intent to use protects a name before launch, and it's often the right call. But it adds a second hurdle after approval, and many products never launch under the name that was filed. Of the intent-to-use filings, 42.7% were already abandoned by September 2026 — about twice the rate for marks that were in use when filed — and a large share of the rest were still in their Statement of Use period.
What Doesn't Matter as Much
Logos and plain word marks were abandoned at almost the same rate: 31.5% for standard-character marks, 31.4% for designs with words, 23.0% for designs with no words. Applications covering three or more classes were abandoned slightly less often (26.2%) than single-class ones (31.5%), even though, as our Office Action report shows, they draw more objections. The records don't say why.
Abandonment by Class
Among single-class applications, these classes had the highest and lowest abandonment rates (classes with at least 300 applications):
| Class | What it covers | Applications | Abandoned |
|---|---|---|---|
| 38 | Telecommunications | 793 | 43.5% |
| 34 | Tobacco & vaping | 1,535 | 42.3% |
| 32 | Beer & non-alcoholic drinks | 2,299 | 40.5% |
| 25 | Clothing, footwear & headwear | 21,188 | 39.1% |
| 30 | Coffee, bakery & staples | 5,591 | 38.5% |
| 33 | Wine & spirits | 3,587 | 37.6% |
| Class | What it covers | Applications | Abandoned |
|---|---|---|---|
| 20 | Furniture | 8,687 | 22.9% |
| 12 | Vehicles | 4,722 | 22.4% |
| 27 | Carpets & floor coverings | 1,018 | 21.7% |
| 11 | Lighting, heating & appliances | 6,773 | 21.6% |
| 6 | Metal goods | 2,371 | 20.7% |
| 26 | Lace, ribbons & hair accessories | 2,309 | 19.5% |
Drinks, food, tobacco and clothing sit near the top; furniture, vehicles, lighting and metal goods near the bottom. The data shows the pattern, not its cause.
Abandoned Isn't Always Final
An application abandoned because a response or Statement of Use was missed can be brought back with a petition to revive, if the delay was unintentional and the petition is filed within two months of the notice of abandonment. In this cohort, 9,838 applications (3.5%) had a petition to revive granted at some point, and 6,982 of them were registered or still pending at the end of September 2026. The median abandoned application in the cohort died 12.4 months after filing.
How We Counted
- Who's counted: every U.S. application with a filing date between January 1 and June 30, 2024. Madrid Protocol filings (Section 66(a), serial numbers starting with 79) are excluded because they follow a different process.
- Outcome: each application's USPTO status code on September 30, 2026, grouped with the USPTO's own live/dead indicator from its status code table: registered, abandoned, pending, or other.
- Reason: the USPTO status description for abandoned applications (codes 600–609, 612 and 614).
- Office Action and final refusal: at least one non-final action (event codes GNRT/CNRT) or final refusal (GNFR/CNFR) in the prosecution history, as in our Office Action report.
- Filing basis: the basis claimed when the application was filed. An application that claimed intent to use for any goods or services counts as 1(b).
- Revival: a "petition to revive — granted" event (PETG) in the prosecution history.
- What the data can't show: why an owner stopped. A missed deadline and a deliberate decision to drop a mark produce the same record.
- USPTO Open Data Portal: Trademark Full Text XML, daily applications (TRTDXFAP)
- USPTO Open Data Portal: Trademark Full Text XML, annual backfile (TRTYRAP)
- Public research copy of both files used for the counts (illeatmyhat/uspto-trademarks)
- USPTO Office of the Chief Economist: Trademark Case Files Dataset (event code descriptions)
- USPTO: Trademark Applications Daily XML documentation v2.0 (legal entity type codes)
- 37 C.F.R. § 2.62: period for response to an Office Action
- 37 C.F.R. § 2.89: extensions of time to file a Statement of Use
- 37 C.F.R. § 2.66: reviving an abandoned application — unintentional delay, two-month deadline
Bulk data is updated daily by the USPTO; figures here are fixed at September 30, 2026. They are our calculations, not statistics published by the USPTO. Links checked 19 October 2026.