Colorado lost its patent office; its inventors never stopped filing.

When the U.S. Patent and Trademark Office announced last October that it was permanently closing its Rocky Mountain Regional Outreach Office, the reaction around Denver was predictable. The office had opened in 2014, one of four created under the America Invents Act, and losing it felt like losing a seat at the table. The natural worry followed quickly: if the patent office leaves, do the filings leave too? Do the companies?

It is a fair question. It is also being answered incorrectly right now, and the way people are getting it wrong is worth understanding, because the same mistake surfaces any time someone reaches for recent patent data.

Start with what the record shows. Measured by published applications listing at least one Colorado inventor, Colorado filed 9,337 patent applications in 2013 and 10,586 in 2024, peaking at 11,018 in 2023. That is about 13% growth. Over the same period, national filings moved the other way, falling roughly 7%. Colorado’s share of all U.S. filings rose from 2.7% to 3.3%, and it has climbed in every period since bottoming out in 2016.

Even the most recent complete year holds up. Colorado filings did dip in 2024 by about 4%. But they fell slightly faster nationally, which means Colorado gained share in a down year. Whatever has been happening to this state’s inventors over the last decade, it is not decline.

Now the part that trips people up. Pull Colorado’s 2025 numbers today, and they appear to fall off a cliff, down more than 40%. Pull 2026, and they very nearly vanish. Stop there, and you have a dramatic story about an office closing and an ecosystem collapsing with it.

That story would be wrong, and the reason is a statute rather than an economy. Under federal law, a patent application stays confidential for 18 months after it is filed. Until it publishes, it does not appear in any public dataset. So a search run in August 2026 can see filings through roughly January 2025 and very little after that. The apparent collapse is not Colorado companies filing less. It is the public record catching up.

Two checks confirm as much. First, if the closure had driven filings down, you would expect a break in October 2025, the month it was announced. There is no break: Colorado recorded 484 applications that September and 483 that October. Second, and more telling, the publication delay suppresses every state by the same factor, so it cancels out if you look at Colorado’s share of the national total instead of its raw count. On that measure, Colorado sat at 3.27% across 2024, 3.15% in the months before the announcement, and 3.37% in the months after it. The number went up, and all three figures fall inside the ordinary month-to-month range.

So the honest answer to whether losing the patent office reduced Colorado patent filings is that nobody can know yet, and anyone who tells you otherwise is reading an artifact. Applications filed after the announcement begin publishing in the spring of 2027. That is when the question becomes answerable.

This is also hard to check independently. The patent office’s own published state-by-state tables stop at 2020. Anyone who wants Colorado’s recent numbers has to pull them directly from the agency’s data systems, which a general reader can’t casually verify. That gap is part of why a plausible claim about Colorado filings can circulate unchallenged.

None of this means the closure was harmless. It means filing volume is the wrong place to look for the harm. A regional office’s value was never that it processed the paperwork of nearby companies. Applications have been filed electronically from anywhere for years, and examiners have worked remotely for longer than that. The value was in the things that never show up in a filing count: walk-in help for someone with an idea and no lawyer, classes for first-time filers and a physical signal to a company deciding where to put an engineering team. If the closure cost Colorado something, it will show up in who is not filing, particularly independent inventors and small businesses without counsel. That is genuinely harder to measure than a total, and it is the thing worth watching.

The distinction matters for what Colorado does next. The patent office has said it is weighing where to place a future regional office. A state making that case is better served by an accurate account of its own strength, a decade of gaining share against a declining national trend, than by an argument built on a decline the data does not support. The first is verifiable and flattering. The second falls apart the moment somebody checks it, and somebody will.

For Colorado businesses, the practical takeaway is smaller and more useful. Nothing about the closure changed how or where you file, what your applications cost, or how they are examined. What changed is that the low-friction, in-person front door for people who did not know where to start is gone. If your company relied on that door, the substitute is not a building. It is knowing who to ask.

Charles W. Gray is a partner at Kilpatrick Townsend & Stockton in Denver, where he practices patent prosecution. He is the founder and leader of Kilpatrick Labs, the firm’s dedicated artificial intelligence research and development program. Charles has been recognized by IAM Patent 1000: The World’s Leading Patent Practitioners and by IAM Strategy 300: The World’s Leading IP Strategists, including as an IAM Strategy 300 Global Leader in 2025 and 2026. He has been named a Patent Leader by World IP Review and recognized by The Best Lawyers in America® for Patent Law. He earned his J.D. from the University of Denver College of Law.